PDA

View Full Version : Why not add time after 40 moves?



Jesper Norgaard
05-02-2010, 02:02 PM
Since I haven't got an answer to why (or in which official documents!) it is a compelling rule that electronic clocks may not add time for the next period after for instance 40 clock presses but may only do it when one of the players has used up all his time for the 1. period, I will instead turn the arguments upside-down and argument why I think it is better that the clock adds extra time after 40 moves, to really test if there are any valid arguments for the opposite.

In the good old days (which we still are part of in the sense that still some games are handled by analog clocks), a solution had to be found for 2 hours for 40 moves and ˝ hour for the rest of the game. Since there is only flag fall on whole ours, it is necessary that the arbiter jumps in and add the ˝ hour manually so that the next flag fall will happen on the whole hour instead of the half hour (where there is no flag). It is pretty clear that this has to happen manually and can only happen when the players actually have 40 moves on the scoresheet. Probably it has been determined over time that the best solution is to only allow that when both 40 moves are present on the scoresheet and a flag fall of one of the players has happened, to avoid this scenario: Both players agree before any flag fall that 40 moves have been made, and ask the arbiter to add ˝ hour to both clocks, and play goes on. Suddenly one player figures out that 40 moves have not been made, and claims that the opponent lost on time. In this case it would be necessary to check flag fall at the half hour (an impossible task) in addition to reconstructing the moves - conclusion, in the analog clock it is best to wait until one of the players time has completely expired, before the times get adjusted. So far so well.

Perhaps this argument has just been "transferred" directly to the electronic clocks? But if that is the case, no attention has been paid to that flag falls on an electronic clock always can be checked by simple calculation, you don't even need the shown flagfall (or red button) to determine that. Every second on the clock has been clearly defined and can be used to calculate down to the second how long time has been used by either player.

Let us look on the 2 scenarios:

(A) Electronic clock adds time to the 2. time period after 40 moves (i.e. clock presses):
When overstepping time at move 37 (if both players have fumbled with the buttons and pressed extra times when no move was made on the board) the players will not discover the flag fall because time has already been added, unless they manually check the scoresheet. But checking the scoresheet, the players must check that time for the 2. period is not added until 40 moves have been written on the scoresheet.

(B) Electronic clock adds time to the 2. time period when one of the players uses all his time for the 1.st. time period:
When overstepping time at move 37 the clock will automatically add extra time at flag fall. Again it is possible to check that the clock has added extra time before the 40 moves have been written on the scoresheet. I really don't see much difference between the two examples. In this example, although the time has been added to both clocks, the player that has less time than ˝ hour is the one who lost on time, even though his time is 29:58 against 35:11 for the other player.

So what is the reason that I feel that (A) is a better definition than (B)? Apparently they are equivalent in that what must be checked is that time is only added after 40 moves have been completed on the scoresheet?

There are several examples that the electronic clocks work better under (A) than under (B). I do not doubt that some of them works fine under both definitions (for instance the DGT XL). But we have several examples that electronic clocks don't work that well under (B). The DGT 2000 if 1. period is set to 0 moves instead of 40 moves, will simply backlock after one of the players time run out, showing "-0.00" and not adding any more time to the clocks. In my clock the Excalibur Game Time II (B) works in the way that time after the flag fall is correctly counted, but the flag fall (and red button) will never be acquitted no matter how many more moves are made. In fact to remove this intermediate flag fall, the arbiter (or worse, one of the players!) needs to stop the clock, "adjust" the time by just clicking OK on all settings, and after this operation the flag fall is acquitted (no red button anymore).

I have no doubt that there are more trouble with electronic clocks managing (B) while most of them work correctly under (A).

This means there are 2 arguments (there might be more depending on other clock models) that (A) is better than (B). Are they at all any arguments from arbiters for situations where (B) is better than (A)?

CameronD
05-02-2010, 02:17 PM
Some electronic clocks dont show seconds until under 20 minutes, is their a way to retrieve this data from the machine.

Jesper Norgaard
05-02-2010, 03:05 PM
Some electronic clocks dont show seconds until under 20 minutes, is their a way to retrieve this data from the machine.
I think most clocks can be stopped, and then asked to adjust the time, and then it will be possible to see the seconds for each player. Some clocks can be set up to always show the seconds too in normal mode (like Excalibur Game Time II).

Bill Gletsos
05-02-2010, 03:55 PM
I believe the following scenario highlights the problem with adding extra time automatically at move 40.

Players A & B have pressed the clock more times than the moves actually made.

After the clock thinks 40 moves have been made it adds 30 minutes additional time but neither player notices the time added.
Player A's clock now shows 33 minutes and players B's 31 minutes.

A & B both make a move (or 2).

7 minutes after the clock added the 30 minutes Player A notices that it is only move 39 on the score sheet.

If A's clock now reads 29 minutes and B's 28 minutes then if the 30 minutes had not been added the clocks would show -1 & -2 minutes respectively. However as the clocks have not hit zero (due to the additional time being added based on the move counter) the clock will not show which clock hit zero first.

This scenario cannot happen if the second time control is not added until one clock hits zero.

Ian Rout
05-02-2010, 04:01 PM
The reason that time is added after the first expiry and not 40 moves (in the analogue era and then carrying over to digital) is that it is the player's responsibility to know that they have played 40 moves. They are not entitled to get advice from a clock or arbiter.

CameronD
05-02-2010, 04:05 PM
The reason that time is added after the first expiry and not 40 moves (in the analogue era and then carrying over to digital) is that it is the player's responsibility to know that they have played 40 moves. They are not entitled to get advice from a clock or arbiter.

Just out of interest, I know a lot of players who draw a line at the 40 move mark on the scoresheet, is this allowable??

Kevin Bonham
05-02-2010, 04:18 PM
Just out of interest, I know a lot of players who draw a line at the 40 move mark on the scoresheet, is this allowable??

Technically it might be considered to be making notes but I have never known of any cases of anyone actually caring.


The scoresheet shall be used only for recording the moves, the times of the clocks, the offers of a draw, and matters relating to a claim and other relevant data.

Garrett
05-02-2010, 04:24 PM
I wonder if anyone objected, and the player was required to get a new scoresheet, if it would be okay to fold the scoresheet at the 40 move mark (therby leaving a crease) ?

But yes, I cannot imagine anyone caring.

Jesper Norgaard
05-02-2010, 06:00 PM
I believe the following scenario highlights the problem with adding extra time automatically at move 40.

Players A & B have pressed the clock more times than the moves actually made.

After the clock thinks 40 moves have been made it adds 30 minutes additional time but neither player notices the time added.
Player A's clock now shows 33 minutes and players B's 31 minutes.

A & B both make a move (or 2).

7 minutes after the clock added the 30 minutes Player A notices that it is only move 39 on the score sheet.

If A's clock now reads 29 minutes and B's 28 minutes then if the 30 minutes had not been added the clocks would show -1 & -2 minutes respectively. However as the clocks have not hit zero (due to the additional time being added based on the move counter) the clock will not show which clock hit zero first.

This scenario cannot happen if the second time control is not added until one clock hits zero.

I must admit that I had not foreseen this scenario. Quite ingenious! However, is this not biting it's own tail in a way? What you obtain as an extra is the benefit of being able to tell who first lost on time when it is determined that in fact both players exceeded the time control. If in fact just one player has exceeded the time control (the far more likely scenario) there is no benefit.

But going along with this idea, then the first flag fall in the first period must be carried on indefinitely because we can't really tell how many moves have been made until there is an actual reconstruction of moves. This means that all along 2. period one flag will be continuously blinking. Suppose we talk about 3 periods like in the Gibraltar tournament. Now the scenario could be that the flag was blinking for Player 1 all along 2. period, but Player 2 hit zero first in 2.period, so all along 3. period Player 2's flag was blinking, right until both players perhaps exceed the final time limit, and then it depends on the last zero-hit whose flag is blinking. Quite confusing in my opinion! A big price to pay just to be able to tell if both players exceeded a particular period, who hit zero first. You would have a (possibly immaterial) blinking flag all through 3-4 hours of play, and shifting between players, in fact in every single game, still without meaning anything until it was in fact the last flag fall, or until it was determined that both of the players in some period had exceeded the time.

With time added at 40 moves, and then 20 more moves, the flag would in fact not show until the very end, when one of the players exceeded the last time. Unless of course they both forgot to press the clock in one or several moves, when the flag would be shown e.g. when time was exceeded at move 39 according to the clock, when in fact there were 40 moves on the score sheet.

There is an additional problem. If there is always added an increment to each move, you can't really tell from a blinking flag if the flag means that in this period the time was exceeded, and positive time was restored with the increment, or if the time was not exceeded, but the flag is displaying from the previous period, where it might have been immaterial because in that period sufficient moves were made.

Quite mind-boggling stuff, and I might have overlooked something again, but it still seems to me that the (A) time control is more, shall we say, appealing to common sense. It will show in the last period who lost on time first, but will usually not show any false positives before that. With the (B) time control we are obliged at false positives all along the last 2 time periods.

Jesper Norgaard
05-02-2010, 06:08 PM
The reason that time is added after the first expiry and not 40 moves (in the analogue era and then carrying over to digital) is that it is the player's responsibility to know that they have played 40 moves. They are not entitled to get advice from a clock or arbiter.
You are right about the arbiter may not advice about how many moves have been made, but not the clock:

6.14 Screens, monitors, or demonstration boards showing the current position on the chessboard, the moves and the number of moves made, and clocks which also show the number of moves, are allowed in the playing hall. However, the player may not make a claim relying solely on information shown in this manner.

So the clock may show how many moves have been punched (not made) :P

Bill Gletsos
05-02-2010, 09:50 PM
The other issue is that by adding 30 minutes after 40 moves have been made (according to the clocks move counter) the the requirement of Article 6.3 is not actually met when the first time control period actually elapses.


Immediately after a flag falls, the requirements of article 6.2 a. must be checked.

When using a chess clock, each player must make a minimum number of moves or all moves in an allotted period of time and/or may be allocated an additional amount of time with each move. All these must be specified in advance.

Jesper Norgaard
06-02-2010, 01:32 AM
The other issue is that by adding 30 minutes after 40 moves have been made (according to the clocks move counter) the requirement of Article 6.3 is not actually met when the first time control period actually elapses.
So that after move 40, if he just moved in the same second, and the clock added 30 minutes (for new period) and 30 seconds (for the move) and his clock shows 30:30 (as opposed to 30:31 where he still had at least a fraction of a second left) then he should have lost on time? Well, I tend to think that this should be checked by the hardware, so it should actually raise the flag in this situation. Hopefully most clock designs will have that functioning correctly.

Or is your concern more than fraction-of-a-second? Well if clock makers have made their homework right, I think that since you can't move at exactly at 0 thousands of a millisecond, either you would have surpassed by a few milliseconds (in which case the clock should flag you) or would have a few milliseconds in reserve) in which case the clock should show 30:31 with no flag fall.

Santa
01-03-2010, 11:41 AM
Just out of interest, I know a lot of players who draw a line at the 40 move mark on the scoresheet, is this allowable??

On scoresheets I used in Victoria, one had to turn the page at move 40.

I think I might invite a player complaining about his opponent's making such a mark to do the same.

But then, I'm no IA.

Kevin Bonham
19-07-2011, 06:28 PM
Just curious, don't know the tourney or the clocks, have never tried the new FIDE time, but ... didn't FIDE make a statement that the clocks should not be set to automatically add 30 minutes at move 40 because that might lead to confusion. Instead 30 minutes are only added to both clocks when one player oversteps 90 minutes and 0 seconds. If this happens before 40 moves have been made by that player, he loses on time (at least if discovered before both flags fall)?

I'm not sure about this. Haven't heard of it being added manually in Aus tournaments using this or similar time controls. A concern with stopping the clock only at flagfall for the stated time control is that since 30 secs/move are added from move 1, players might take a long time after move 40 to reach zero, if they even reached it at all. Sounds a bit confusing to administer.

Keong Ang
19-07-2011, 07:43 PM
Just curious, don't know the tourney or the clocks, have never tried the new FIDE time, but ... didn't FIDE make a statement that the clocks should not be set to automatically add 30 minutes at move 40 because that might lead to confusion. Instead 30 minutes are only added to both clocks when one player oversteps 90 minutes and 0 seconds. If this happens before 40 moves have been made by that player, he loses on time (at least if discovered before both flags fall)?

If both flags have fallen in the first period, if the clock indicates which flag fell first, the corresponding player loses. Otherwise, the game continues (in this case to finish).

"90 minutes for 40 moves + 30 minutes with 30 seconds cumulative increment for each move starting from the first move" is a standard FIDE rate of play for a tournament to be eligible for title norms.
I believe it is the same time control used at the Olympiad.

...it is also the same time control used at the George Trundle NZ Masters that is currently entering round 5 of 9.

With DGT XL clocks, we set them up to automatically add extra 30 minutes on move 40.
As far as I know, all international tournaments set up the clocks to automatically add the 30 minutes with 40 moves as the first time control. FIDE specifies 40 moves as the uniform first time control in games with multi time control rate of play. Players benefit from uniformity.

One reason for manually adding 30 minutes instead of doing it automatically is because the clocks that are used are incapable of handling multiple time controls automatically. This only happens when the organiser forgot to check what rates of play the clocks are capable of beforehand.

Errors with "add 30 minutes at move 40" occur when players click and counter-click the clock without making a move. The arbiter needs to regularly check that all clocks move counters accurately reflect that on the scoresheets. Usually the error that occurs is 30 minutes is added before move 40 is made due to the clock being clicked and counter-clicked repeatedly by the players.

I have also experienced cases where a player complains that the clock did not add 30 minutes even though more than 40 moves have been made. Checking the scoresheets would show eg. 42 moves, therefore a valid complaint. Checking the move counter on the clock would show eg. 36 moves. Conclusion, the players have been making moves without clicking the clock. Simple solution is to click until correct number of moves registered. The extra 30 minutes would be added automatically and all 30 second increments would also be automatically added.

Garvinator
19-07-2011, 08:59 PM
With DGT XL clocks, we set them up to automatically add extra 30 minutes on move 40.
As far as I know, all international tournaments set up the clocks to automatically add the 30 minutes with 40 moves as the first time control. FIDE specifies 40 moves as the uniform first time control in games with multi time control rate of play. Players benefit from uniformity.We have been through this before Keong. The clocks are meant to be set to not use the move counter at all. Once one players clock reaches zero, the dgt xl adds the 30 minutes to both players clocks, if the move counter has not been set.

From move 41, both players have received their 30 minutes, it is just not shown until one players clock reaches zero.


One reason for manually adding 30 minutes instead of doing it automatically is because the clocks that are used are incapable of handling multiple time controls automatically. This only happens when the organiser forgot to check what rates of play the clocks are capable of beforehand.All three dgt versions 2000, xl and 2010 can handle up to four time periods, set from the beginning. Have a look at the back of the clock and you will even see some pre-set options that specify up to four different time periods.


Errors with "add 30 minutes at move 40" occur when players click and counter-click the clock without making a move. The arbiter needs to regularly check that all clocks move counters accurately reflect that on the scoresheets. Usually the error that occurs is 30 minutes is added before move 40 is made due to the clock being clicked and counter-clicked repeatedly by the players.This is an incorrect procedure and Keong, we have been through this before. Clearly the last conversation on this topic has fallen from your memory.

With the circumstance that can happen with the scenario you mention above is that a player receives the extra 30 minutes from the clock, starts thinking for five minutes, then the opponents checks the scoresheets and realises that only 38 moves have been made.

Player A has then used more than the 40 moves in 90 mins plus 30 secs (so 40 moves in 110 minutes total) and so would lose on time for overstepping the time period.

I would imagine any player who had this happened to them would be very pissed off and probably rightly so.

If the move counter is not set, then this can never occur.


I have also experienced cases where a player complains that the clock did not add 30 minutes even though more than 40 moves have been made. Checking the scoresheets would show eg. 42 moves, therefore a valid complaint. Checking the move counter on the clock would show eg. 36 moves. Conclusion, the players have been making moves without clicking the clock. Simple solution is to click until correct number of moves registered. The extra 30 minutes would be added automatically and all 30 second increments would also be automatically added.This is another false procedure and is against the laws of chess. It is not the arbiters responsiblity to add more time to the player because they forgot to press their clock, which is what you are doing by the part I have bolded.

The player is responsible for pressing their clock, not the arbiter. If they forgot to press their clock, then they accept whatever happens to them because of it.

From the fide laws of chess:


6.2 a. When using a chess clock, each player must make a minimum number of moves or all moves in an allotted period of time and/or may be allocated an additional amount of time with each move. All these must be specified in advance.

b. The time saved by a player during one period is added to his time available for the next period, except in the ‘time delay’ mode.
In the time delay mode both players receive an allotted ‘main thinking time’.

Each player also receives a ‘fixed extra time’ with every move. The countdown of the main time only commences after the fixed time has expired. Provided the player stops his clock before the expiration of the fixed time, the main thinking time does not change, irrespective of the proportion of the fixed time used.

6.3 Immediately after a flag falls, the requirements of article 6.2 a. must be checked.

It is 6.3 that matters most here. An arbiter can only add the extra time after one players clock has reached zero. That was the procedure when analogs were being used and it is still the same procedure now.

If you do not use a move counter, then there is no need for the arbiter to do anything, as long as the second time period has been set correctly. All the arbiter has to do is to have a quick look at both scoresheets once both players have made move 40.

And that is it.


Also, from the fide laws of chess-

Article 13: The role of the Arbiter (See Preface)
13.1 The arbiter shall see that the Laws of Chess are strictly observed.

13.2 The arbiter shall act in the best interest of the competition. He should ensure that a good playing environment is maintained and that the players are not disturbed. He shall supervise the progress of the competition.

13.3 The arbiter shall observe the games, especially when the players are short of time, enforce decisions he has made and impose penalties on players where appropriate.

13.4 The arbiter can apply one or more of the following penalties:

warning
increasing the remaining time of the opponent
reducing the remaining time of the offending player
declaring the game to be lost
reducing the points scored in the game by the offending party
increasing the points scored in the game by the opponent to the maximum available for that game
expulsion from the event.

13.5 The arbiter may award either or both players additional time in the event of external disturbance of the game.

13.6 The arbiter must not intervene in a game except in cases described by the Laws of Chess. He shall not indicate the number of moves made, except in applying Article 8.5, when at least one flag has fallen. The arbiter shall refrain from informing a player that his opponent has completed a move or that the player has not pressed his clock.

13.7 Spectators and players in other games are not to speak about or otherwise interfere in a game. If necessary, the arbiter may expel offenders from the playing venue. If someone observes an irregularity, he may inform only the arbiter.
Unless authorised by the arbiter, it is forbidden for anybody to use a mobile phone or any kind of communication device in the playing venue and any contiguous area designated by the arbiter.

In 13.6, it clearly states that the arbiter can not indicate the number of moves made, unless checking 8.5

antichrist
19-07-2011, 11:29 PM
why not add time after 40 moves?

because the arbiters would have to get off their backside and actually do something as they had to with the mechanical clocks.

A bit of exercise for chessplayers is a plus

Keong Ang
20-07-2011, 01:27 PM
We have been through this before Keong. The clocks are meant to be set to not use the move counter at all. Once one players clock reaches zero, the dgt xl adds the 30 minutes to both players clocks, if the move counter has not been set.

From move 41, both players have received their 30 minutes, it is just not shown until one players clock reaches zero.

By not setting the move counter at all, you're just recreating all the problems that analogue clocks have. Plus creating issues with monitoring the 30 second increments.

What exactly is the problem with using the clock's move counter for it's design purpose? After all isn't the move counter meant to correctly handle moving into the next time control for a 90' + 30' on move 40 + 30" increment from move 1 rate of play?

I've just done some experiments with a DGT XL clock. What error is supposed to happen that makes it essential to not use the move counter?


All three dgt versions 2000, xl and 2010 can handle up to four time periods, set from the beginning. Have a look at the back of the clock and you will even see some pre-set options that specify up to four different time periods.

Yes I know. All these FIDE approved clocks are alright. I was referring to other clocks that cannot handle multi time controls that are significantly cheaper to buy.


This is an incorrect procedure and Keong, we have been through this before. Clearly the last conversation on this topic has fallen from your memory.

With the circumstance that can happen with the scenario you mention above is that a player receives the extra 30 minutes from the clock, starts thinking for five minutes, then the opponents checks the scoresheets and realises that only 38 moves have been made.

Player A has then used more than the 40 moves in 90 mins plus 30 secs (so 40 moves in 110 minutes total) and so would lose on time for overstepping the time period.

I would imagine any player who had this happened to them would be very pissed off and probably rightly so.

If the move counter is not set, then this can never occur.

After being flamed by Bill and you, it's not easy to forget!! :lol:

The scenario I'm referring to tends to happen when illegal moves are made, and would actually be handled under article7.4.a-b.
Also can happen with players fumbling of clock.

Regular checking of move counter against scoresheets would pick up these infractions. Usually arbiter is alerted by hearing repeated clicking in the tournament hall.

Even in rates of play without multiple time controls, arbiters need to pick up on this as too many 30 second increments may have been added.


This is another false procedure and is against the laws of chess. It is not the arbiters responsiblity to add more time to the player because they forgot to press their clock, which is what you are doing by the part I have bolded.

The player is responsible for pressing their clock, not the arbiter. If they forgot to press their clock, then they accept whatever happens to them because of it.

Exactly what articles in the Laws of Chess is this against?

When a player complains to the arbiter that his clock did not add 30 minutes after 40 moves have been made, the arbiter needs to apply article6.10.a. Checking players scoresheets and move counter would show eg. 42 moves on scoresheet and 36 moves on clock.
As the arbiter, I'd suspect that the players haven't been pressing the clock, but how to prove it?
Every indication on the clock conclusively shows that only 36 moves have been registered despite 42 moves having been made. Accusing the player of forgetfulness is not going to work because the player can validly claim that the clock is faulty by not registering clock presses. Keep in mind that the arbiter is saying BOTH players are being forgetful here!
The correct resolution is to adjust the clock time and move counter to accurately show what's on the scoresheets. Simplest way to do it is to click until correct number of moves is registered.

Note that the arbiter is not adding time here. The players are simply getting the correct increments and time controls according to the scoresheets. No additional thinking time is added, all time that has elapsed before they discovered the "error" is lost to the players.

This is exactly the procedure thought to me during my FA seminar. It was also a FA exam question. If it is a false procedure that breaks the rules, I need to know!! :eek:


It is 6.3 that matters most here. An arbiter can only add the extra time after one players clock has reached zero. That was the procedure when analogs were being used and it is still the same procedure now.

If you do not use a move counter, then there is no need for the arbiter to do anything, as long as the second time period has been set correctly. All the arbiter has to do is to have a quick look at both scoresheets once both players have made move 40.

And that is it.

If you do not use the move counter, the arbiter has to ensure that flag falls are caught before move 40 is played.

Picture this situation. Flag fall's at move 39. Since move counter is not used, clock automatically adds 30 minutes to both players. Neither player notices a flag has fallen before move 40 and the game continued at a faster pace. At move 41 the arbiter comes by to check scoresheets and flag fall indicator means nothing now as it is past move 40.
Wouldn't a player have escaped losing the game here?

If move counter was used, the clock would have registered a flag fall before the 40 move time control and the clock would have locked with zeros and flag on one player's display.


In 13.6, it clearly states that the arbiter can not indicate the number of moves made, unless checking 8.5

That's why it is better to set the clock up with move counter. It is much better to have the clock showing the actual time remaining. Much less confusing for the players. At move 40 white player would have 30 minutes added while black player has not since still at move 39. If black's flag falls now, it's correct and because the clock locks up, no big dispute.

Jesper Norgaard
24-07-2011, 01:00 PM
why not add time after 40 moves?
because the arbiters would have to get off their backside and actually do something as they had to with the mechanical clocks.
antichrist, why are you so clueless? The question here is whether the clock should add time after 40 clock presses or not. It would not make any arbiter get off their backside. In any event the adding of the time for a digital clock is done automatically either at 40 clock presses or when a flag fall is reached, and that is an advantage over mechanical clocks.


By not setting the move counter at all, you're just recreating all the problems that analogue clocks have. Plus creating issues with monitoring the 30 second increments.

What exactly is the problem with using the clock's move counter for it's design purpose? After all isn't the move counter meant to correctly handle moving into the next time control for a 90' + 30' on move 40 + 30" increment from move 1 rate of play?

I've just done some experiments with a DGT XL clock. What error is supposed to happen that makes it essential to not use the move counter?

The question is not whether the clock counts the moves. The question is whether time for a new period is added based on it or not. The error most obviously encountered would be that the score sheets are wrong, and if they are, the arbiter would be applying a wrong correction. Therefore correcting time based on moves should only be done in a possible reconstruction of moves after a period (e.g. after a flag fall), or perhaps if one of the players makes a specific claim to the arbiter that the move counter and time should be corrected. It is not a good idea that the arbiter runs around checking and correcting move counters and adjust clocks all the time. It is uncalled for and it is error-prone.

Note that most digital clocks can be set to count the moves and show that count on demand, even if it is not used to add time for a new period.



This is exactly the procedure thought to me during my FA seminar. It was also a FA exam question. If it is a false procedure that breaks the rules, I need to know!! :eek:

It is not incorrect to adjust time based on exact moves made, however, players are not faultless with score sheets, so only in a reconstruction of moves, this would be free from errors, and therefore should only take place on an actual reconstruction or on a player claim. I am not aware how the FA exam question was worded. Can you elaborate?



If you do not use the move counter, the arbiter has to ensure that flag falls are caught before move 40 is played.

That is the normal situation, the arbiter must always check any flag fall, and if not the last period, if enough moves have been played. That is equivalent to mechanical clock management. This task will be more troublesome for the arbiter if he will never see the flag fall because the clock already added the time for the new period, erroneously or not.


Picture this situation. Flag fall's at move 39. Since move counter is not used, clock automatically adds 30 minutes to both players. Neither player notices a flag has fallen before move 40 and the game continued at a faster pace.

Perhaps the players won't notice, but the arbiter should. If time is added for a new period, it means to the arbiter that the flag fall was accounted for, so the number of moves must have been completed - otherwise someone lost on time. It is easy to see that a flag fall occurred if you think about it.


At move 41 the arbiter comes by to check score sheets and flag fall indicator means nothing now as it is past move 40. Wouldn't a player have escaped losing the game here?

Yes, but firstly it is also the players responsibility to check flag fall, but secondly the arbiter was late. He should be around at "fixed time + 40 x increment" or perhaps rather "fixed time + 38 x increment". If any flag fall has occurred is easy to check on the remaining time shown, which would have added time if a flag had fallen. Any flag fall before that time would probably be erroneous based on too few clock presses, and is therefore a bogus flag fall, that should be checked, but would probably always after reconstruction reach a conclusion that the flag fall was erroneous. If the flag fall came early because one player used a lot more time than the opponent, the reconstruction and adjustment according to clock presses will reveal if the player lost on time or not.


If move counter was used, the clock would have registered a flag fall before the 40 move time control and the clock would have locked with zeros and flag on one player's display.

This has the additional problem that the clock freezes and does not count time even though the flag fall was not correct because the players actually pressed the clock too few times. This means that from the flag fall to the flag fall is noticed, the used time for both players are not accounted for. A very unfortunate situation. The arbiter can't correct himself out of this situation since there is no way to guess how much time each player used of that fraction of time.

This is perhaps the biggest problem of all when letting the clock decide based on clock presses when the flag has fallen.

antichrist
24-07-2011, 06:07 PM
Originally Posted by antichrist

why not add time after 40 moves?
because the arbiters would have to get off their backside and actually do something as they had to with the mechanical clocks.

antichrist, why are you so clueless? The question here is whether the clock should add time after 40 clock presses or not. It would not make any arbiter get off their backside. In any event the adding of the time for a digital clock is done automatically either at 40 clock presses or when a flag fall is reached, and that is an advantage over mechanical clocks.

AC
probably so clueless coz I only played maybe two comps under digital clocks and was totally turned off by the experience, so has not been back since in all those years to remember their ins and outs. sorry boss (that is main reason not playing comp any more) Plus I dont like the Lcd or whatever image they use. A certain angles they are not clear. You are only needed a split second glance at mechanical clock.

Keong Ang
28-07-2011, 12:15 AM
The question is not whether the clock counts the moves. The question is whether time for a new period is added based on it or not. The error most obviously encountered would be that the score sheets are wrong, and if they are, the arbiter would be applying a wrong correction. Therefore correcting time based on moves should only be done in a possible reconstruction of moves after a period (e.g. after a flag fall), or perhaps if one of the players makes a specific claim to the arbiter that the move counter and time should be corrected. It is not a good idea that the arbiter runs around checking and correcting move counters and adjust clocks all the time. It is uncalled for and it is error-prone.

The procedure is for the arbiter to check that the scoresheets show sufficient moves when the clock adds the 2nd control's time. Insufficient moves recorded in the scoresheet means the players have been clicking the clock repeatedly without making moves. Definately incorrect procedure by the players.

The arbiter only needs to check the move counter to prove infractions on the players part.

Obviously players who make moves without pressing their clocks are going to be left alone by the arbiter until one of them claims the clock is faulty!!


Note that most digital clocks can be set to count the moves and show that count on demand, even if it is not used to add time for a new period.

Yes, the clock should be set to count moves. If an increment (eg. 30seconds) is used, an arbiter needs to check that the same "repeat clicking" error is not occuring. Many minutes can be added by such incorrect procedure on the players part.

FIDE is moving towards endorsing clocks that display the move count at all times. I think it's recorded in the technical committee's minutes that there is no reason to not allow it to be displayed.


It is not incorrect to adjust time based on exact moves made, however, players are not faultless with score sheets, so only in a reconstruction of moves, this would be free from errors, and therefore should only take place on an actual reconstruction or on a player claim. I am not aware how the FA exam question was worded. Can you elaborate?

The FA seminar had rather lengthy instruction on how the following articles in the Laws of Chess interplay in situations during a tournament.


6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

6.9
Except where one of the Articles: 5.1.a, 5.1.b, 5.2.a, 5.2.b, 5.2.c applies, if a player does not complete the prescribed number of moves in the allotted time, the game is lost by the player. However, the game is drawn, if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves.

6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

b.
If during a game it is found that the setting of either or both clocks was incorrect, either player or the arbiter shall stop the clocks immediately. The arbiter shall install the correct setting and adjust the times and move counter. He shall use his best judgement when determining the correct settings.

8.2
The scoresheet shall be visible to the arbiter throughout the game.

8.7
At the conclusion of the game both players shall sign both scoresheets, indicating the result of the game. Even if incorrect, this result shall stand, unless the arbiter decides otherwise.

The exam questions based on the Laws of Chess would be describing a situation that occurred. With the usual "What is the arbiter's decision?" question before we need to write down the answers quoting the relevant articles of the Laws of Chess. There's really only time to write down what laws were broken, what laws to apply and what actions the arbiter must take.

In most situations, the correct answer would conclude with, penalties to apply and "adjust the clock's move counter".

FIDE puts a lot of effort instructing prospective arbiters on the correct use of clock move counters so that the correct number of clock presses by the players together with correct recording of moves during the course of a game through multiple time controls happens.

Not setting the move counter to automatically enter the next time control after a prescribed number of moves is an infraction of article6.10.b. and "the offending arbiter who set the clock should hang his head in shame" :eek:


Perhaps the players won't notice, but the arbiter should. If time is added for a new period, it means to the arbiter that the flag fall was accounted for, so the number of moves must have been completed - otherwise someone lost on time. It is easy to see that a flag fall occurred if you think about it.

The idea is that the clock should lock up when a flag falls. Then it does not matter when the players or arbiter finally notices it, the result is conclusive due to the clock. For multiple time control rates of play, the clock should only enter the next time control when the required number of moves have been played.


Yes, but firstly it is also the players responsibility to check flag fall, but secondly the arbiter was late. He should be around at "fixed time + 40 x increment" or perhaps rather "fixed time + 38 x increment". If any flag fall has occurred is easy to check on the remaining time shown, which would have added time if a flag had fallen. Any flag fall before that time would probably be erroneous based on too few clock presses, and is therefore a bogus flag fall, that should be checked, but would probably always after reconstruction reach a conclusion that the flag fall was erroneous. If the flag fall came early because one player used a lot more time than the opponent, the reconstruction and adjustment according to clock presses will reveal if the player lost on time or not.

The trouble is, the games all start at the same time and therefore multiple boards would near flag fall around the same time.

Anyway, if a flag falls, it is a totally different situation. In fact it is a simpler situation, the player whose flag falls loses. No reconstruction is required.

The arbiter only needs to reconstruct and adjust if a player makes a claim before a flag fall.


This has the additional problem that the clock freezes and does not count time even though the flag fall was not correct because the players actually pressed the clock too few times. This means that from the flag fall to the flag fall is noticed, the used time for both players are not accounted for. A very unfortunate situation. The arbiter can't correct himself out of this situation since there is no way to guess how much time each player used of that fraction of time.

This is perhaps the biggest problem of all when letting the clock decide based on clock presses when the flag has fallen.


This is where article6.10.a. applies.
That is why FIDE approved clocks must lock up when a flag falls. The game is over. The player whose flag has fallen has lost and the clock's move counter conclusively says at which move the game ended. All subsequent "moves" are invalid as they were made after flag fall.

The arbiter should not be trying to resurrect a game that is over in this case.

A FA exam question covered this situation. Two players continued playing without noticing a flag fall for 15 moves before the arbiter comes and notices a flag has fallen. Answer: Clock provides conclusive proof to the arbiter on the exact move the game ended (through move counter) and also which player lost art.6.1.a. Decision, player whose flag fell lost the game at move that clock's move counter registers. Action, tell players game is over and collect signed scoresheets.

Instead of being the biggest problem, it is actually the best solution to the problem. Arbiters can manage larger tournaments with this technology.

Players are encouraged to record clock times on their scoresheets, very useful if the clock malfunctions. Also useful as evidence of a clock's evident defect.

antichrist
28-07-2011, 09:30 AM
I can recall a few moves being made after the flagfall has fallen with mechanical clocks but 15 moves, that is more moves than GM's draw with. Coz with digital clocks there is no mechanical to quickly peep at between mvoes that is how can occur. Which agrees with how I stated ages ago that I can't read the digital clocks easily at angles so makes me lose concentration etc. - the mechanical I can guess if at angle coz can still see clearly.



So with digital clocks if everything runs perfectly (like nuke re-actors) then arbiters can conduct tournaments, but with mechanical clocks if another "apprentice" arbiter is also utilised it is all better for the game. This arbiter could concentrate on the lower end of the draw to begin with.

As KB says the rules are not more but the implementation seems a lot more involved when considering the posts up above

Jesper Norgaard
28-07-2011, 05:35 PM
In a perfect world, each player clicks the clock after each move, and writes each move on his score sheet. The clock has been set correctly from the beginning, and has no malfunction during the game. This is the optimum situation for which I agree that whether a new period begins after 40 clock presses by each player or after one players clock reaches zero, is of equal value.

In the real world the following errors are common and should be considered:
(1) Arbiter or players set the clock wrongly from the start
(2) Players forget to click after a move
(3) Players forget to write down moves or write moves double
(4) The clock malfunctions during the game or is paused for a long time by mistake

Thefore the clock can't be the arbiter. It can only passively record time passing and count clock presses. Relying on the clock alone is a serious arbiter mistake. It's a can of worms, and not recognizing it will just mislead you to wrong decisions automatically.

Let me show you an example where errors are slight compared to the above threats, but still we have a problem, Houston.

The clock is set to add new time after 40 clock presses, 90 minutes for 40 moves, and 30 minutes for the rest of the game. 30 seconds added after each move.

During the game, player A two times make a move and player B rapidly responds with a prepared move, so both clicks are lost. Player A doesn't know he is still entitled to press the clock in this situation, and player B is not doing anything to recover the lost clock presses. Player A forgets to write 2 half-moves around move 12, and player B forgets to write 2 half-moves around move 26. After only 2˝ hours of play player A's flag falls, since player B played quickly. His scoresheet shows 39 moves and player B's score sheet shows 39 moves at this moment. Unfortunately the arbiter is not around and the flag fall is not discovered until 10 minutes later. Since the clock has only received 38 clock presses from each, it freezes the clock and flags player A. At this point they play 3 more moves each, but naturally, nobody knows exactly how much time each player used for these moves, because the clock froze and did not record the time or clock presses after the "flag fall".

Now the flag fall is discovered by the arbiter and he rushes to make a decision. The clock shows the flag fall happened at move 38, but as we already know, this happened actually at move 40, because we know the players had forgotten to press the clock twice. However, both players are now at move 42 at the score sheet, while the correct game would actually be at move 43 (since they both missed to write 2 half-moves).

I suppose arbiter Keong Ang might forfeit player A, because he lost on time at move 38. However at that point player A was entitled to 1 minute extra time because he had made 2 moves where 30 seconds were not added at each, so in fact he had not lost on time when the clock froze. Now a reconstruction takes place, and it is determined that the game is 43 moves long, even though both score sheets had only 42 moves. The clock is still on 38 moves. At this point we can't say anything about how many moves had been played in the game when the clock had recorded 38 clock presses and a flag fall. We only know that in fact at this point 43 moves have been made on the board.

The situation would have been different if the flag fall had been discovered immediately, and after the flag fall, the reconstructed number of moves were equal to the clock count because then we know that the last thing that happened in the game, was in fact the flag falling, and none of the players would have been entitled to extra time, so the flag fall should stand.

In the current case arbiter Keong, after careful deliberation, must conclude that the flag fall is not certain (in fact we know it is false, he could not know for certain). He must order the game to continue. However, now he can't adjust the time on both players clocks for the 3 moves and 10 minutes that happened after "flag fall". He would have to guess who is discounted the 10 minutes. If he does not discount the 10 minutes some way, he might prolong how long the game could take by 10 minutes.

The same scenario but with no time added at move 40, and no flag fall registered whether or not 40 clock presses has taken place when one clock exceeds the time:
Player A clock reaches zero exactly at the same time as the previous scenario, but the clock does not freeze, and each player is added 30 minutes, and the clocks continue. They play 3 more moves, and for each move the corresponding clock is running, so the 10 minutes are recorded accurately for each player. Now the arbiter comes around at the same time, and sees that a flag fall has happened (since they have 30 minutes extra each), but on the score sheet the number of moves shown is 42 so he does nothing (not even reconstruction, which would have revealed 43 moves played). In effect at this point, no time forfeit is registered. Maybe the arbiter will notice that only 41 clock presses on the clock, while 42 moves on the score sheet, so he adds a clock press to each players clock. If he had reconstructed, he would have known that he should actually have added 2 clock presses to each player.

In the first scenario, inexperienced arbiters would surely have forfeited player A even though he in fact still had time left (around 1 minute) when the "flag fall" happened. And even an experienced arbiter would have problems, because the 10 minutes elapsed after "flag fall" is not accounted for, so who should they be taken from, player A or B, and with which proportion?

In the second scenario all of these problems would have been avoided.


The procedure is for the arbiter to check that the scoresheets show sufficient moves when the clock adds the 2nd control's time. Insufficient moves recorded in the scoresheet means the players have been clicking the clock repeatedly without making moves. Definately incorrect procedure by the players.

Definitely. What we don't know is where they erred, on the score sheet or on the clock presses. Perhaps the players did not write moves and the clock presses are correct. Correcting the clock presses back if that was the case would definitely be a mistake.



The arbiter only needs to check the move counter to prove infractions on the players part.
Correct, but which correction to apply is more difficult. And you will have false positives and true negatives - for instance if they recorded 2 moves extra and pressed 2 times extra, it would seem that everything is correct, while it is in fact not.



FIDE is moving towards endorsing clocks that display the move count at all times. I think it's recorded in the technical committee's minutes that there is no reason to not allow it to be displayed.
Yes and that matches the laws of chess:
"6.14
Screens, monitors, or demonstration boards showing the current position on the chessboard, the moves and the number of moves made, and clocks which also show the number of moves, are allowed in the playing hall. However, the player may not make a claim relying solely on information shown in this manner."
FIDE in fact is not "moving towards", they already moved!



FIDE puts a lot of effort instructing prospective arbiters on the correct use of clock move counters so that the correct number of clock presses by the players together with correct recording of moves during the course of a game through multiple time controls happens.

Perhaps they are making too much effort in mechanical reactions and too little on common sense. Some of your comments make me suspect that. But I am not in the position to judge whether FIDE is doing that.



Not setting the move counter to automatically enter the next time control after a prescribed number of moves is an infraction of article6.10.b. and "the offending arbiter who set the clock should hang his head in shame" :eek:

This is quite simply rubbish. The article in no way indicates *how* it should be set, only that if it was set wrong it should be corrected. Using 40 moves or 0 moves for the first period can by no means cause an infraction of 6.10.b. In fact there is no part of the laws of chess that indicates which is correct (unfortunately). I have asked for this clarification, but no arbiter has until now been able to point to an official document of any kind that tells which one is recommended, let alone to be "the only correct way" as you suggest. However all my comments here should indicate to you why I am finally leaning to 0 moves for the first period is correct, because it avoids that the clock freezes when in fact the game should have continued, and therefore the clock loses control of both clock presses and time elapsed.



The idea is that the clock should lock up when a flag falls. Then it does not matter when the players or arbiter finally notices it, the result is conclusive due to the clock. For multiple time control rates of play, the clock should only enter the next time control when the required number of moves have been played.

This seems to be your own idea, and an unfortunate one. Where do you see that in the laws of chess that the clock should lock up? I disagree on all 4 points. The only thing correct on all of this is that after a flag fall in the last period, the clock should lock up, but I still don't think it says so in the laws. A mechanical clock does not lock up, but it is irrelevant.



Anyway, if a flag falls, it is a totally different situation. In fact it is a simpler situation, the player whose flag falls loses. No reconstruction is required.

Hilariously wrong.



The arbiter only needs to reconstruct and adjust if a player makes a claim before a flag fall.

Wrong again.



This is where article6.10.a. applies.
That is why FIDE approved clocks must lock up when a flag falls. The game is over. The player whose flag has fallen has lost and the clock's move counter conclusively says at which move the game ended. All subsequent "moves" are invalid as they were made after flag fall.

Again wrong. Note that the clock can *never* tell at which move the game ended, only at which clock press the flag fall was recorded. There is a world of difference.


The arbiter should not be trying to resurrect a game that is over in this case.

Sigh.



A FA exam question covered this situation. Two players continued playing without noticing a flag fall for 15 moves before the arbiter comes and notices a flag has fallen. Answer: Clock provides conclusive proof to the arbiter on the exact move the game ended (through move counter) and also which player lost art.6.1.a. Decision, player whose flag fell lost the game at move that clock's move counter registers. Action, tell players game is over and collect signed scoresheets.

Instead of being the biggest problem, it is actually the best solution to the problem. Arbiters can manage larger tournaments with this technology.

Letting the clock be the arbiter, will perhaps permit the arbiter to make quicker decisions, but also more wrong decisions based on pseudo-evidence.



Players are encouraged to record clock times on their scoresheets, very useful if the clock malfunctions. Also useful as evidence of a clock's evident defect.
I must applaud this, although of course, fraud could happen here but is not likely. As a last remark, thank you for the input on the contents of the FA exams, and your feedback, although we seem to disagree on a number of points.

antichrist
28-07-2011, 10:32 PM
All this is all too sickly for me to even read - and the benefits of digital clocks are....

Clicks not registering blah blah - play it again Sam

Keong Ang
29-07-2011, 11:53 PM
In a perfect world, each player clicks the clock after each move, and writes each move on his score sheet. The clock has been set correctly from the beginning, and has no malfunction during the game. This is the optimum situation for which I agree that whether a new period begins after 40 clock presses by each player or after one players clock reaches zero, is of equal value.

In the real world the following errors are common and should be considered:
(1) Arbiter or players set the clock wrongly from the start
(2) Players forget to click after a move
(3) Players forget to write down moves or write moves double
(4) The clock malfunctions during the game or is paused for a long time by mistake

Thefore the clock can't be the arbiter. It can only passively record time passing and count clock presses. Relying on the clock alone is a serious arbiter mistake. It's a can of worms, and not recognizing it will just mislead you to wrong decisions automatically.

Let me show you an example where errors are slight compared to the above threats, but still we have a problem, Houston.

The clock is set to add new time after 40 clock presses, 90 minutes for 40 moves, and 30 minutes for the rest of the game. 30 seconds added after each move.

During the game, player A two times make a move and player B rapidly responds with a prepared move, so both clicks are lost. Player A doesn't know he is still entitled to press the clock in this situation, and player B is not doing anything to recover the lost clock presses. Player A forgets to write 2 half-moves around move 12, and player B forgets to write 2 half-moves around move 26. After only 2˝ hours of play player A's flag falls, since player B played quickly. His scoresheet shows 39 moves and player B's score sheet shows 39 moves at this moment. Unfortunately the arbiter is not around and the flag fall is not discovered until 10 minutes later. Since the clock has only received 38 clock presses from each, it freezes the clock and flags player A. At this point they play 3 more moves each, but naturally, nobody knows exactly how much time each player used for these moves, because the clock froze and did not record the time or clock presses after the "flag fall".

Now the flag fall is discovered by the arbiter and he rushes to make a decision. The clock shows the flag fall happened at move 38, but as we already know, this happened actually at move 40, because we know the players had forgotten to press the clock twice. However, both players are now at move 42 at the score sheet, while the correct game would actually be at move 43 (since they both missed to write 2 half-moves).

I suppose arbiter Keong Ang might forfeit player A, because he lost on time at move 38. However at that point player A was entitled to 1 minute extra time because he had made 2 moves where 30 seconds were not added at each, so in fact he had not lost on time when the clock froze. Now a reconstruction takes place, and it is determined that the game is 43 moves long, even though both score sheets had only 42 moves. The clock is still on 38 moves. At this point we can't say anything about how many moves had been played in the game when the clock had recorded 38 clock presses and a flag fall. We only know that in fact at this point 43 moves have been made on the board.

The situation would have been different if the flag fall had been discovered immediately, and after the flag fall, the reconstructed number of moves were equal to the clock count because then we know that the last thing that happened in the game, was in fact the flag falling, and none of the players would have been entitled to extra time, so the flag fall should stand.

In the current case arbiter Keong, after careful deliberation, must conclude that the flag fall is not certain (in fact we know it is false, he could not know for certain). He must order the game to continue. However, now he can't adjust the time on both players clocks for the 3 moves and 10 minutes that happened after "flag fall". He would have to guess who is discounted the 10 minutes. If he does not discount the 10 minutes some way, he might prolong how long the game could take by 10 minutes.

The same scenario but with no time added at move 40, and no flag fall registered whether or not 40 clock presses has taken place when one clock exceeds the time:
Player A clock reaches zero exactly at the same time as the previous scenario, but the clock does not freeze, and each player is added 30 minutes, and the clocks continue. They play 3 more moves, and for each move the corresponding clock is running, so the 10 minutes are recorded accurately for each player. Now the arbiter comes around at the same time, and sees that a flag fall has happened (since they have 30 minutes extra each), but on the score sheet the number of moves shown is 42 so he does nothing (not even reconstruction, which would have revealed 43 moves played). In effect at this point, no time forfeit is registered. Maybe the arbiter will notice that only 41 clock presses on the clock, while 42 moves on the score sheet, so he adds a clock press to each players clock. If he had reconstructed, he would have known that he should actually have added 2 clock presses to each player.

In the first scenario, inexperienced arbiters would surely have forfeited player A even though he in fact still had time left (around 1 minute) when the "flag fall" happened. And even an experienced arbiter would have problems, because the 10 minutes elapsed after "flag fall" is not accounted for, so who should they be taken from, player A or B, and with which proportion?

In the second scenario all of these problems would have been avoided.

In the first scenario, the arbiter observed that the clock's flag has already fallen. What knowledge does the arbiter have about the situation?
From your scenario description, the arbiter has no knowledge of players missing clock presses and missed move recording because he has not observed it. Obviously, if the arbiter had observed such errors earlier, he would have already penalised the players for infractions of article6.7.a. and article8.1.

Practically what would have happened is that the arbiter would observe that the playerA has had a flag fall and neither player has noticed it. The arbiter would check the move counter and that is conclusive (article6.10.a) that the game ended at move38. All other moves in the scoresheet are discounted as occuring after flag fall.

You must remember that in scenario one, there is nothing for the arbiter to suspect that something has gone wrong. What is observed are two players who are still playing on despite a flag fall. The only relevant points to consider are that the flag has fallen, the move counter says flag fell at move38, players scoresheets show 42moves. Extra human factors that an experienced arbiter should disregard is the opinion that these two players are tardy with the clock and therefore most probably tardy with the scoresheet.

An inexperienced arbiter would waste time and effort to reconstruct the game in such a scenario. Even if the game was reconstructed, there is no record of when moves were made. All a reconstruction would prove is that the players are skipping recording of moves (infraction of article8.1) and that the flag fell earlier than what was recorded as move38 on the scoresheets. But this is an irrelevant issue.

In scenario two, you have just described how the players have got away with it. You are really trying to cater for tardy players by setting the clock wrong. The clock should show the exact amount of time remaining for each player. Otherwise, you would have situations where the player can claim that the clock erroneously states 10minutes remaining when there is actually 40minutes. That is a valid article6.10.b claim, but that is a different matter to what we're currently discussing.

You are clearly mistaken to say that the arbiter "must conclude that the flag fall is not certain", as article6.10.a forces the arbiter to decide that the flag conclusively fell at move38. The fact that there is no evident defect in the clock forces such a conclusion. The arbiter needs to ignore suspicions in his mind that the players have been forgetting to press the clock.

You are also wrong to state that I'd be forfeiting playerA if I was the arbiter. The correct answer is playerA lost the game on time. It is not a forfeit because both players have made at least one move. Score is zero for playerA and one for playerB.

Scenario one is a typical FA exam question. The correct answer requires you to separate irrelevant information that is given. An actual arbiter in scenario one would have no way of knowing how the situation came about. All that he would know is that flag has fallen on move38 and the players have continued playing several moves without noticing.


Definitely. What we don't know is where they erred, on the score sheet or on the clock presses. Perhaps the players did not write moves and the clock presses are correct. Correcting the clock presses back if that was the case would definitely be a mistake.


Correct, but which correction to apply is more difficult. And you will have false positives and true negatives - for instance if they recorded 2 moves extra and pressed 2 times extra, it would seem that everything is correct, while it is in fact not.

These situations are very rarely picked up during tournaments. How many arbiters actually check that a position on the board is could be reached by x number of moves on the scoresheet?


Yes and that matches the laws of chess:
"6.14
Screens, monitors, or demonstration boards showing the current position on the chessboard, the moves and the number of moves made, and clocks which also show the number of moves, are allowed in the playing hall. However, the player may not make a claim relying solely on information shown in this manner."
FIDE in fact is not "moving towards", they already moved!

I'm referring to FIDE moving towards endorsing clocks that display number of moves made constantly. Currently FIDE approved clocks eg. DGT XL does not show the number of moves unless a button is pressed.

Article6.14 is something the arbiter has to be mindful of when a player makes a claim because the player must back it up with scoresheets and the actual position on the board. Basically it is an instruction for arbiters to disregard Screens, monitors or demonstration boards as the sole source of evidence. They can only be used to support a claim based on eg. scoresheets and clock.


Perhaps they are making too much effort in mechanical reactions and too little on common sense. Some of your comments make me suspect that. But I am not in the position to judge whether FIDE is doing that.

Did you know that there is nothing in FIDE regulations that requires arbiters to have common sense? :uhoh: :eh:


3. Requirements for the title of FIDE Arbiter (FA).
All of the following:
3.1 Thorough knowledge of the Laws of Chess, the FIDE Regulations for chess competitions, the Swiss Pairing Systems.
3.2 Absolute objectivity, demonstrated at all times during his activity as an arbiter.
3.3 Sufficient knowledge of at least one official FIDE language.
3.4 Skills to operate electronic chess clocks of different types and for different systems.
.....
4. Requirements for the title of International Arbiter (IA).
All of the following:
4.1 Thorough knowledge of the Laws of Chess, the FIDE Regulations for chess competitions, the Swiss Pairing Systems,
the FIDE Regulations regarding achievement of title norms and the FIDE Rating System.
4.2 Absolute objectivity, demonstrated at all times during his activity as an arbiter.
4.3 Obligatory knowledge of English language, minimum at conversation level; and of chess terms in other official FIDE languages.
4.4 Minimum skills at user level to work on a personal computer. Knowledge of pairing programs endorsed by the FIDE,
Word, Excel and E-mail.
4.5 Skills to operate electronic chess clocks of different types and for different systems.
.....

Also from the preface to the Laws of Chess

The Laws assume that arbiters have the necessary competence, sound
judgement and absolute objectivity. Too detailed a rule might deprive the arbiter of his freedom of
judgement and thus prevent him from finding the solution to a problem dictated by fairness, logic
and special factors.

As you can see, "common sense" is not a listed requirement. Instead the emphasis tends to be on mechanical reactions to situations so that the result would be the same globally. ;)


This is quite simply rubbish. The article in no way indicates *how* it should be set, only that if it was set wrong it should be corrected. Using 40 moves or 0 moves for the first period can by no means cause an infraction of 6.10.b. In fact there is no part of the laws of chess that indicates which is correct (unfortunately). I have asked for this clarification, but no arbiter has until now been able to point to an official document of any kind that tells which one is recommended, let alone to be "the only correct way" as you suggest. However all my comments here should indicate to you why I am finally leaning to 0 moves for the first period is correct, because it avoids that the clock freezes when in fact the game should have continued, and therefore the clock loses control of both clock presses and time elapsed.

What do you mean "no part of the laws of chess that indicates which is correct"? Everywhere you read "adjust move counter" in the Laws of Chess, doesn't it say to you that the move counter is performing an important job? If you set the time control to 0 moves, you are making the clock's move counter irrelevant.

I had to go through a thorough session in the FA seminar on just how and why we must set the clock's move counter correctly. There was one fellow candidate who was always insisting that the move counter should be set to 0 moves, he did not pass the exam.

Also, it is part of FIDE's testing procedure for clocks that they can lock up (freeze) when flag falls before sufficient clock presses are made. For clocks to get FIDE approval, they need to be able to regulate the passage of multiple time controls. Clocks that allow players to "get away" are not approved.


This seems to be your own idea, and an unfortunate one. Where do you see that in the laws of chess that the clock should lock up? I disagree on all 4 points. The only thing correct on all of this is that after a flag fall in the last period, the clock should lock up, but I still don't think it says so in the laws. A mechanical clock does not lock up, but it is irrelevant.

The Laws of Chess do not specify that a clock should lock up because of the need to allow for all sorts of clocks to be used. Especially in federations with limited resources.

A clock needs to lock up to negate the problems and complications of situations where both flags fall. There are rates of play that you require the clock to still count down after flag fall so that both flags can fall, but that's a different topic.

For our example "90 minutes for 40 moves + 30 minutes with 30 seconds cumulative increment for each move starting from the first move" rate of play, the clock should be able to lock up so that article6.2.a and article6.9 can be applied without question in light of article6.10.a.

If clocks that can be set to do that are available, the setting should be used.

Anyway, quoting from FIDE Tournament Rules
10. Chess equipment
.....
(c) If electronic chess clocks are used, they must function in full accordance with the FIDE Laws.
(1) The display at all times should show the time available to complete a player’s next move.
(2) The displays must be legible from a distance of at least 3 metres.
(3) From a distance of at least 10 metres a player must have a clearly visible indication which clock is running.
(4) In case of passing a time control, a sign on the display must give clear signal which player passed the time limit first.
(5) For battery powered clocks, a low-battery indication is required.
(6) In case of a low-battery indication the clock must continue to function flawlessly for at least 10 hours.
(7) Special attention should be given to the correct announcement of passing time controls.
(8) In case of accumulative or delay timing systems, the clock should not add any additional time if a player passed the last time control.
(9) In case of time penalties it must be possible that time and move counter corrections are executed by an arbiter within 60 seconds.
(10) It must be impossible to erase or change the data in display with a simple manipulation.
(11) Clocks must contain a short user manual on the clock. Electronic chess clocks used for FIDE events must be endorsed by FIDE Technical Commission.

10.c.1 makes setting the move counter mandatory when interpreting Laws of Chess articles. If the clock says there's 30 minutes left, it must mean there's 30 minutes left. Not, that there's 30 minutes left if your scoresheet says it has 40 or more moves otherwise your flag has fallen and you've actually lost your game on time!!

Similarly, if the clock displays 12 seconds remaining, it means when that the flag will fall once 12 seconds has passed. Not that another 30 minutes will be added.

Let me know what you think when you've read it. The Laws of Chess do not contain all the regulations an arbiter has to be mindful of.


Hilariously wrong.


Wrong again.


Again wrong. Note that the clock can *never* tell at which move the game ended, only at which clock press the flag fall was recorded. There is a world of difference.


Sigh.

Why are you always trying to find reasons for the arbiter to distrust the clock and resurrect games that have ended?
It is not in the best interest of the competition and hinders the progress of the competition.
We need game results quickly.

You seem to not understand article6.10.a, especially the first words "Every indication given by the clocks is considered to be conclusive
..."

This means that the arbiter must start off with the opinion that everything the clock says is correct. If the clock cannot be proven wrong, by default the clock is right.


Letting the clock be the arbiter, will perhaps permit the arbiter to make quicker decisions, but also more wrong decisions based on pseudo-evidence.

The logic is that the evidence the clock provides (moves, flag fall, etc) is correct. The clock does not replace the arbiter. Instead, the arbiter uses the information provided by the clock to make the decision. The information provided by the clock must be assumed to be correct unless it is obviously wrong.

Wasn't I admonished by Garvinator several posts ago for adjusting the clock to reflect correct number of moves before flag fall. Now you seem to be saying that an arbiter needs to intervene and resurrect a game after flag fall.
What happened to the earlier position about the players having to bear the consequences of forgetting to press their clocks?


I must applaud this, although of course, fraud could happen here but is not likely. As a last remark, thank you for the input on the contents of the FA exams, and your feedback, although we seem to disagree on a number of points.

Thank you for the opportunity to let me think about some of the things the FA seminar drilled into me.
One thing that I heard was that "Arbiters are not gods, they can and do make wrong decisions, but they must make a decision in every case."

Jesper Norgaard
31-07-2011, 07:23 AM
First a couple of introductory notes.

(1) You seem to think that if the addition of time for a new period is not based on the move counter, the importance of the move counter is lost. My argument is that the move counter is a very important parameter for all time controls with increments, and it is reliable in the sense that the count it shows directly tells how much extra time both players have received in multiplum of increments (here 30s). Any adjustment of the clock by the arbiter must take it into account.
(I would call the move counter a click counter, because the clock cannot count the moves, it can only count clicks).

(2) The reason that I object to your argument that a flag fall always stand and does not need to be checked at all, lies in article 6.14 where it says that a claim from a player based solely on for instance a clock's move counter, so neither an arbiter can make a claim solely on the clock's move counter. He must take into account the move number based on the score sheets, and if necessary (especially if there is a claim or there are discrepancies between the two score sheets) he must reconstruct to get the right number of moves. If the moves demonstrated on the score sheets does not match the move counter on the clock, it is quite clear that the time added based on increments is wrong, exactly by the difference given multiplied here with 30 seconds. My example was trying to show that both players had gotten too few increments since the clock move counter would have been two less than the actual moves made. However, at the point the arbiter intervened, this discrepancy could not be shown because the move counter was stuck on 38. We return again to the problem that we don't know at which *move* the move counter froze at 38. If it was at move 38, then the flag fall should stand. If it was at move 41, it should not stand. By freezing up, the clock has ruined the evidence. We now have 42 moves on the score sheets, so at the time of intervention, the players have made enough moves to pass the period.

(3) If the flag fall had been discovered directly, the example would perhaps be simpler. The clock would show 38 clicks. The score sheets would show 39 moves. Each clock therefore lacks 30 seconds, so the flag fall does not stand. If the score sheets had shown 38 moves or less, the flag fall would stand. The clocks should be adjusted, and player A still has 30 seconds to complete his 40.th. move. If they had actually done a reconstruction based on a claim of one of the players (seeing that the score sheets don't match after move 12), then they would have got that the players were actually at move 40, and 1 minute should be added to each clock. However, only one single difference e.g. more moves than clicks means that the flag fall doesn't stand. You might say player A must be penalised for not pressing the clock, but I would disagree with that.

(4) Some interesting cases in Geurt Gijssen's column is geurt144.pdf (April 2010) and geurt145.pdf (May 2010)
Pierre Becker made the same claim as you, the clock is always right. Geurt rejects this and states "If both clocks showed thirty-nine moves, then the last move was played and also completed by the player of the black pieces. In that case only the flag of the player of the white pieces could fall when he was thinking about his forty first move. I assume that they made forty moves as both score sheets indicated. The game should continue and the arbiter has to adjust the clocks. I am afraid
that I have to inform you that your decision was wrong."

Harald Eder claims that his opponent's flag fall should stand. In April 2010 Geurt agrees with this, but then IA Günter Mitterheimer (in May 2010) writes to him with his view on the case, and Geurt changes his mind and confirm that it was not sure that the flag fall should stand, so the game should continue. What is especially interesting for our debate is the following comment from Geurt: "By the way, in 2006 there was a FIDE decision that the move counter should not be activated, but, as far as I can see, in many events, including the match Anand - Topalov, the move counter was activated." To this day I have still not seen more documentation on this "decision" by FIDE. It is clear that there is great confusion among arbiters whether one or the other or both ways of setting the clock are correct. I think it would be fair to come up with a decision which one is valid (as they did in 2006) and then make the other way invalid. But that has not happened officially by FIDE in writing to all public.

(5) I can only pity the poor chap in your exam that insisted on 0 moves for period 1, and did not pass the exam. He was only using the FIDE decision of 2006! I hope he was not flunked just because of this, e.g. unless he made more "mistakes".

(6) Resuming my opinion about adding 30 minutes when 40 clock presses have been made, it is possible to do that, but then no flag fall determined by the clock is valid in other periods than the last if there are less move counts on the clock than moves completed on the score sheets. You can construct cases with this characteristic where there was in fact still time left for the player that was flagged, so the claim must be null and void. I dislike the fact that with this time control, the clock freezes before move 40 and therefore does not account for subsequent clock presses and time used, if they players or the arbiter did not discover right away the flag fall. Also it is too easy to get to the situation that the players make too many clock presses, and the clock automatically gives time for the new period even if not enough moves have been made in the game. The players will see the added time and not discover they are still in time-trouble. If a player is flagged in this situation, we could talk about a violation of the first rule for certification of clocks
(1) The display at all times should show the time available to complete a player's next move.
A player might have 4 minutes left for the last move in reality, but because the clock already added time, it shows that he has 34 minutes left. If he thinks for 5 minutes about the move, he might to his surprise be claimed to have exceeded the time. With the 0 moves for first period he will see 4 minutes is left, and is not that likely to exceed this without checking on the score sheet if he has passed the number of moves required for the first period.


In the first scenario, the arbiter observed that the clock's flag has already fallen. What knowledge does the arbiter have about the situation?
From your scenario description, the arbiter has no knowledge of players missing clock presses and missed move recording because he has not observed it.

He knows that the score sheets show 42 moves and that the move counter is 38. You are right about that he does not know about the missing move recording by both players, but comparing score sheets even casually, should reveal the discrepancy.



Obviously, if the arbiter had observed such errors earlier, he would have already penalised the players for infractions of article6.7.a. and article8.1.

I don't think it would be correct to penalise honest mistakes of the players in either article 6.7.a and article 8.1. I have never heard of a decision like that. Please enlighten me. On the other hand, if the player deliberately ignores the two articles, and even refuses to follow them, a penalty even up to losing the game for not following the laws of chess, might be the arbiter's only option. Of course it is true that the arbite can implement small time penalties for any specific kind of situation ocurring, but I doubt an appeal to an appeal committee would hold.



Practically what would have happened is that the arbiter would observe that the playerA has had a flag fall and neither player has noticed it. The arbiter would check the move counter and that is conclusive (article6.10.a) that the game ended at move 38.
This article states "Every indication given by the clocks is considered to be conclusive in the absence of any evident defect". The clock shows time left for the players, and a move counter which in turn shows the number of clicks made. The actual moves made must be taken from the game via the game position and the score sheets, as per 6.14. I am not questioning the indications of the clock at all as you might have noticed. I am questioning the interpretation. A click counter is not a move counter even though you call it that.



You must remember that in scenario one, there is nothing for the arbiter to suspect that something has gone wrong.
Yes the move counter vs. the score sheets show a discrepancy.



An inexperienced arbiter would waste time and effort to reconstruct the game in such a scenario.
But he would then save himself the potential embarrassment of being overturned on an appeal.


The clock should show the exact amount of time remaining for each player.
Yes of course it can never be exact as per the faulty clock presses of the players (this is documented of even world-class players making these kind of mistakes). But it should not show 34 minutes when the player in fact has 4 minutes left. It is true that the clock could not know that. And that is exactly my point.


You are also wrong to state that I'd be forfeiting playerA if I was the arbiter. The correct answer is playerA lost the game on time. It is not a forfeit because both players have made at least one move. Score is zero for playerA and one for playerB.
I meant to distinguish "the arbiter declares that the player has lost on time" in contrast to "the player has lost on time". There is a clear difference in terms of legality. I should not have used the f-word I admit. If the arbiter is right, the score is zero-one as you explain.


Scenario one is a typical FA exam question. The correct answer requires you to separate irrelevant information that is given.
The score sheets are not irrelevant information. They are in fact the prime piece of evidence, and has always been. Check 6.14 and ask yourself why it is formulated like that.



These situations are very rarely picked up during tournaments. How many arbiters actually check that a position on the board is could be reached by x number of moves on the scoresheet?
I think it is prudent that the arbiter checks the score sheets, but he should only be blamed for lack of objectivity if he ignores in his decision that there are a different number of moves on the score sheets and the clock, not if he doesn't notice that the two score sheets are different.



I'm referring to FIDE moving towards endorsing clocks that display number of moves made constantly. Currently FIDE approved clocks eg. DGT XL does not show the number of moves unless a button is pressed.
I am not aware that FIDE has not endorsed a clock solely because the move counter was clearly visible all the time. Do you?
What is happening is that DGT and other clock producers are moving towards putting the move counter displayed directly. But the endorsement of this is already part of the laws as per 6.14. In other words, FIDE is not moving towards anything new here.


Article 6.14 is something the arbiter has to be mindful of when a player makes a claim because the player must back it up with scoresheets and the actual position on the board. Basically it is an instruction for arbiters to disregard Screens, monitors or demonstration boards as the sole source of evidence. They can only be used to support a claim based on eg. scoresheets and clock.
Unfortunately the move counter on the clock is directly mentioned in 6.14. If a flag fall of the clock is based on the move counter, then that fact is no more a fact than the rest of 6.14.


Did you know that there is nothing in FIDE regulations that requires arbiters to have common sense? :uhoh: :eh:
I would hope FIDE would want arbiters to have common sense. In the exams, hopefully they are tested whether they have chess arbitrarial common sense. As a matter of fact, I do think that this is tested.


Also from the preface to the Laws of Chess

The Laws assume that arbiters have the necessary competence, sound
judgement and absolute objectivity. Too detailed a rule might deprive the arbiter of his freedom of
judgement and thus prevent him from finding the solution to a problem dictated by fairness, logic
and special factors.

As you can see, "common sense" is not a listed requirement. Instead the emphasis tends to be on mechanical reactions to situations so that the result would be the same globally. ;)
You are being very mechanical here. You seem to think that the preface does not use the word "common sense" then it doesn't care about it. To me the whole paragraph beams of "common sense". How could you "find the solutions to a problem dictated by fairness, logic and special factors" without using common sense? To me they are saying arbiters should look beyond there own little nosetip.



Also, it is part of FIDE's testing procedure for clocks that they can lock up (freeze) when flag falls before sufficient clock presses are made. For clocks to get FIDE approval, they need to be able to regulate the passage of multiple time controls. Clocks that allow players to "get away" are not approved.
The clocks must not allow the player to use all his time in the last period, and then recover from that by making more moves - the flag fall should stand. You think that players can get away with what should have been a flag fall in a period other than the last, when the clock was assigned 0 moves for this period. I invite you to show an example.



A clock needs to lock up to negate the problems and complications of situations where both flags fall. There are rates of play that you require the clock to still count down after flag fall so that both flags can fall, but that's a different topic.
If you care to go back in the thread, Bill Gletsoes shows an example where it is not possible to determine with multiple flag falls in the first period which flag actually fell first, using 40 moves for the first period. This does not happen with 0 moves for first period.



Wasn't I admonished by Garvinator several posts ago for adjusting the clock to reflect correct number of moves before flag fall. Now you seem to be saying that an arbiter needs to intervene and resurrect a game after flag fall.
What happened to the earlier position about the players having to bear the consequences of forgetting to press their clocks?

Yes Garvinator stated that the arbiter should never correct for the players faulty clock presses. I don't agree with that, I have only said the precaution that it should be on a claim or on a reconstruction or on evaluating a flag fall.



Thank you for the opportunity to let me think about some of the things the FA seminar drilled into me.
One thing that I heard was that "Arbiters are not gods, they can and do make wrong decisions, but they must make a decision in every case."
Hopefully a decision with common sense, even though they are not gods.

Actually you don't have to reconstruct the whole game to determine there are discrepancies in the current game where White forgot two half-moves at move 12 and Black forgot two half-moves at move 26

(Yue-Carlsen, Medias 2010, but they didnt make the scoring mistakes, I invented them!)
White's Score Black's Score
1. d4 Nf6 --- 1. d4 Nf6
2. c4 g6 --- 2. c4 g6
... ...
11. Bxc4 Nd7 --- 11. Bxc4 Bxc4
12. Qd2 Qa5 --- 12. 0-0 Nd7
13. Rfd1 Rad8 --- 13. Qd2 Qa5
... ...
25. Nxc3 Ne5 --- 25. e6 Bxc3
26. Red1 Kf8 --- 26. Red1 Kf8
27. Rac1 Ke7 --- 27. Rac1 Ke7

It should be clear just from quickly skimming and comparing the two scores that they are not identical. A reconstruction would now be relevant, and would reveal that white missed 11...Bxc4 12.0-0 and black missed 26.Nxc3 Ne5, so the whole game is one move longer than both score sheets. The score sheets will only be identical at each point even though two half-moves had been omitted, if the players in fact made the same mistake at the same point. To verify all other situations simply compare the two score sheets, and if they are identical, we are 99% sure that no moves are lacking.

In my opinion flag falls are happening infrequently in a period that is not the last, and so the arbiter should be very sure to get it right. There must be no doubts, e.g. no scenario could reach the facts presented without the flag fall being in effect. Imagine you play a game and the arbiter decides you lose the game because you had only completed 39 moves, then you go home and put the game into the chess database and you discover that 40 moves had been completed. I would be very pissed in this situation, even if I technically could be said to not have fulfilled all obligations (of 8.1).

Keong Ang
03-08-2011, 08:46 PM
First a couple of introductory notes.

(1) You seem to think that if the addition of time for a new period is not based on the move counter, the importance of the move counter is lost. My argument is that the move counter is a very important parameter for all time controls with increments, and it is reliable in the sense that the count it shows directly tells how much extra time both players have received in multiplum of increments (here 30s). Any adjustment of the clock by the arbiter must take it into account.
(I would call the move counter a click counter, because the clock cannot count the moves, it can only count clicks).

Obviously the clock only counts clicks. The players are responsible and obligated to click the clock correctly. Hence all moves registered on the clock are "legally correct". I'm not only saying that the move counter is not important, but is so important that it must be activated to regulate the transition between time controls based on moves.


(2) The reason that I object to your argument that a flag fall always stand and does not need to be checked at all, lies in article 6.14 where it says that a claim from a player based solely on for instance a clock's move counter, so neither an arbiter can make a claim solely on the clock's move counter. He must take into account the move number based on the score sheets, and if necessary (especially if there is a claim or there are discrepancies between the two score sheets) he must reconstruct to get the right number of moves. If the moves demonstrated on the score sheets does not match the move counter on the clock, it is quite clear that the time added based on increments is wrong, exactly by the difference given multiplied here with 30 seconds. My example was trying to show that both players had gotten too few increments since the clock move counter would have been two less than the actual moves made. However, at the point the arbiter intervened, this discrepancy could not be shown because the move counter was stuck on 38. We return again to the problem that we don't know at which *move* the move counter froze at 38. If it was at move 38, then the flag fall should stand. If it was at move 41, it should not stand. By freezing up, the clock has ruined the evidence. We now have 42 moves on the score sheets, so at the time of intervention, the players have made enough moves to pass the period.

To clarify article6.14 as thought to me during my FA seminar.
Article6.14 is to allow displays that let spectators follow the games better. When visible in the playing hall, the players obviously can also see them. Article6.14 is really an escape clause for players to avoid infractions of article12.3.a.

The "and clocks which also show the number of moves" refers to the large screen showing each player's time remaining and moves made. It is not referring to the chess clock beside the board that the players are clicking.

"However, the player may not make a claim relying solely on information shown in this manner." actually allows the arbiter to disregard displayed and broadcast information in favour of the actual chess clock and scoresheets.

You've also misinterpreted article6.14 by stating that it also binds the arbiter. In chess law, there are 3 entities; arbiters, players and spectators. A clause that specifically binds players does not necessarily bind the arbiter. Take for example, the analogous situation regarding mobile phone rules for players and spectators. Notice that arbiters are not bound by the mobile phone rule?


(3) If the flag fall had been discovered directly, the example would perhaps be simpler. The clock would show 38 clicks. The score sheets would show 39 moves. Each clock therefore lacks 30 seconds, so the flag fall does not stand. If the score sheets had shown 38 moves or less, the flag fall would stand. The clocks should be adjusted, and player A still has 30 seconds to complete his 40.th. move. If they had actually done a reconstruction based on a claim of one of the players (seeing that the score sheets don't match after move 12), then they would have got that the players were actually at move 40, and 1 minute should be added to each clock. However, only one single difference e.g. more moves than clicks means that the flag fall doesn't stand. You might say player A must be penalised for not pressing the clock, but I would disagree with that.

We're in agreement here.
If flag fall had been observed as it happened, corrections and adjustments can be made because of the information available to the arbiter.

The arbiter must apply penalties in accordance to article6.7.a, article8.1 and article13.1 the word "shall" means it is compulsory.
The penalty I apply is article13.4.a ;)
I'd just mark a "W" for warning on the offending player's scoresheet. If this keeps happening repeatedly, obviously more serious penalties need to be applied.


(4) Some interesting cases in Geurt Gijssen's column is geurt144.pdf (April 2010) and geurt145.pdf (May 2010)
Pierre Becker made the same claim as you, the clock is always right. Geurt rejects this and states "If both clocks showed thirty-nine moves, then the last move was played and also completed by the player of the black pieces. In that case only the flag of the player of the white pieces could fall when he was thinking about his forty first move. I assume that they made forty moves as both score sheets indicated. The game should continue and the arbiter has to adjust the clocks. I am afraid
that I have to inform you that your decision was wrong."

This is a case where it is "almost clear" that the players missed a clock press. So the scoresheets trump the clock's move counter. Similar to the situation that I've mentioned in an earlier post. Also, Pierre Becker's situation was based on accepting the clocks move counter despite knowing that the players had missed a clock press.


Harald Eder claims that his opponent's flag fall should stand. In April 2010 Geurt agrees with this, but then IA Günter Mitterheimer (in May 2010) writes to him with his view on the case, and Geurt changes his mind and confirm that it was not sure that the flag fall should stand, so the game should continue. What is especially interesting for our debate is the following comment from Geurt: "By the way, in 2006 there was a FIDE decision that the move counter should not be activated, but, as far as I can see, in many events, including the match Anand - Topalov, the move counter was activated." To this day I have still not seen more documentation on this "decision" by FIDE. It is clear that there is great confusion among arbiters whether one or the other or both ways of setting the clock are correct. I think it would be fair to come up with a decision which one is valid (as they did in 2006) and then make the other way invalid. But that has not happened officially by FIDE in writing to all public.

Geurt also mentioned that if the clocks move counter was checked, there would be no doubts if it agreed with the scoresheets.

Geurt mentions a 2006 decision. Where are the minutes?
If it was a 2006 decision, it would have been binding from June in the year following the congress (GA?) that approved it.

I suspect the decision was not approved.

Arbiters are trained to activate the move counter to regulate transitions between time controls in multiple time control rates of play that uses number of moves.

Anyway, I've noticed that Geurt is not held up as an example of arbiter accuracy in this forum, and in the limited FIDE circles I have contact with, the opinion is similar.


(5) I can only pity the poor chap in your exam that insisted on 0 moves for period 1, and did not pass the exam. He was only using the FIDE decision of 2006! I hope he was not flunked just because of this, e.g. unless he made more "mistakes".

I don't know what other mistakes he made. Candidates only know who did not pass because a fellow candidate's name did not appear on the list of those who passed.

Activating the clock's move counter was the only thing that I recall him being in disagreement on. There were sufficient questions (to fail!) where the only correct answer is based on the move counter being activated in the first place. If he had tried to be smart and wrote something like "this would never have happened if the move counter was set to 0", he's asking for failure.


(6) Resuming my opinion about adding 30 minutes when 40 clock presses have been made, it is possible to do that, but then no flag fall determined by the clock is valid in other periods than the last if there are less move counts on the clock than moves completed on the score sheets. You can construct cases with this characteristic where there was in fact still time left for the player that was flagged, so the claim must be null and void. I dislike the fact that with this time control, the clock freezes before move 40 and therefore does not account for subsequent clock presses and time used, if they players or the arbiter did not discover right away the flag fall.

I did debate this during my FA seminar. If the clock freezes, it destroys all subsequent time evidence. If the arbiter decides that the flag fall was wrong, the remedy would be to restart play from the last reliably identified position before flag fall. It's like taking back moves to me.

All candidates were instructed to follow standard operating procedure. The clock is right until it is proven that it was not registering clicks (evident defect clause). A player "forgetting" to click the clock is not a considered. It is the players responsibility to click and write moves correctly, not the arbiter's. If neither player observed flag fall, it can be safely assumed (no matter how wrong!) that they had played all subsequent moves written on scoresheets after the flag fell.

That's why arbiters prefers the clock to freeze, it's easy to rule here especially with the full weight of article6.10.a. If a player claims that the clock was not registering clicks (before flag fell), it's just a simple matter to reset the clock and click a few times to see if the clicks are registered. If the clicks are, claim is dismissed.

The player stands a better chance where clock times are recorded on the scoresheet. That's when a scoresheet trumps the clock. But that was not the scenario described.


Also it is too easy to get to the situation that the players make too many clock presses, and the clock automatically gives time for the new period even if not enough moves have been made in the game. The players will see the added time and not discover they are still in time-trouble. If a player is flagged in this situation, we could talk about a violation of the first rule for certification of clocks
(1) The display at all times should show the time available to complete a player's next move.
A player might have 4 minutes left for the last move in reality, but because the clock already added time, it shows that he has 34 minutes left. If he thinks for 5 minutes about the move, he might to his surprise be claimed to have exceeded the time. With the 0 moves for first period he will see 4 minutes is left, and is not that likely to exceed this without checking on the score sheet if he has passed the number of moves required for the first period.

That's why all arbiters are instructed to go around, preferably with a clipboard, recording clock times and moves as recorded on both players scoresheet.
You'd pick up those "frantic clock clickers" this way. Then, penalties are applied.

Keong Ang
03-08-2011, 09:03 PM
He knows that the score sheets show 42 moves and that the move counter is 38. You are right about that he does not know about the missing move recording by both players, but comparing score sheets even casually, should reveal the discrepancy.

A frozen clock tells the arbiter that the game ended at 38moves as on the move counter. That's evidence that article6.10.a makes irrefutable. The only way out is a defective clock. Testing the clock would show no defect. Players forgetting to press their clock is not considered a "clock defect".

Decision is simple, game over at move38. If we reconstruct on scoresheet, we can say that due to player forgetting to record move, what was recorded as say, move12, was actually move13. That means the position as recorded at move37 was actually when the flag fell and moves 38 to 42 were played after flag fell without being noticed. There is nothing on the scoresheets that would prove that a player forgot to press the clock.

Now I know this is harsh. PlayerA has paid the ultimate penalty for an infraction of article6.7.a (forgetting to press clock). But this happened because nobody noticed flag fall in this scenario until 10minutes later, in fact nobody would know with certainty how much time actually elapsed since flag fall. So nobody knows when the flag actually fell. The only evidence is the move counter on the clock.


I don't think it would be correct to penalise honest mistakes of the players in either article 6.7.a and article 8.1. I have never heard of a decision like that. Please enlighten me. On the other hand, if the player deliberately ignores the two articles, and even refuses to follow them, a penalty even up to losing the game for not following the laws of chess, might be the arbiter's only option. Of course it is true that the arbite can implement small time penalties for any specific kind of situation ocurring, but I doubt an appeal to an appeal committee would hold.

This is an interplay between article6.7.a, article8.1, article13.1 and article13.6
It is the arbiter's compulsory duty.

Human nature of wanting to be nice, plus a "common sense" attitude causes such a decision to be rare.


This article states "Every indication given by the clocks is considered to be conclusive in the absence of any evident defect". The clock shows time left for the players, and a move counter which in turn shows the number of clicks made. The actual moves made must be taken from the game via the game position and the score sheets, as per 6.14. I am not questioning the indications of the clock at all as you might have noticed. I am questioning the interpretation. A click counter is not a move counter even though you call it that.

As I've explained earlier, article6.14 is about large displays that also have clocks. A player can say that the huge screen of the DGT broadcast of his eboard says 42moves have been made and the chess clock's move counter is wrong, but the issue is that he needs to prove it from the scoresheet. Put crudely it's a "technology be damned" clause that works against the player.

In the scenario that we're using, the game position and score sheets observed by the arbiter who arrived some time after flag fall are immediately suspect. The 1st assumption is that the players must have made moves after flag fall. This means the current position and number of moves recorded on the scoresheet is erroneous. Logically, the procedure is to disregard all moves after move38.


Yes the move counter vs. the score sheets show a discrepancy.

And the move counter trumps the scoresheets in this case.
All moves after move38 are discounted as "occuring after flag fall".

Now the situation and decision would be different if flag fall was observed as it happened. Or the move counter discrepancy was detected before flag fall. Corrections can then be made.


But he would then save himself the potential embarrassment of being overturned on an appeal.

Are arbiters even the least bit concerned about this?
I've been instructed that there are numerous players who attempt to pressure the arbiter by invoking the "appeal" threat, but we are supposed to ignore it.

Let's not forget that playerB would find the arbiter biased with all that effort he's providing trying to resurrect the game for playerA. Whatever the arbiter's decision, an appeal could come from either front.


Yes of course it can never be exact as per the faulty clock presses of the players (this is documented of even world-class players making these kind of mistakes). But it should not show 34 minutes when the player in fact has 4 minutes left. It is true that the clock could not know that. And that is exactly my point.

That's why the players have the responsibility and obligation to ensure the clock is pressed correctly and the moves are recorded. Errors on the players part if detected should be corrected and adjusted for with the appropriate penalty applied.


I meant to distinguish "the arbiter declares that the player has lost on time" in contrast to "the player has lost on time". There is a clear difference in terms of legality. I should not have used the f-word I admit. If the arbiter is right, the score is zero-one as you explain.

The score sheets are not irrelevant information. They are in fact the prime piece of evidence, and has always been. Check 6.14 and ask yourself why it is formulated like that.

Article6.14 is formulated to allow players to get around article12.3.a when they cannot help noticing information displayed for spectators in the "playing hall". A very badly worded article.

Exactly what is a "playing hall"? Is it the "playing area" or "playing venue"?
Such inaccurate use of words in legislation throws into doubt what is meant by other words. As an arbiter, the use of the word "clocks" with the "s" at the end of it definitely means that it is not referring to the chess clock that players are meant to click. This is in keeping with the meaning of the whole of article6.14. Anyway, it's a clause that applies to the player and not the arbiter.


I think it is prudent that the arbiter checks the score sheets, but he should only be blamed for lack of objectivity if he ignores in his decision that there are a different number of moves on the score sheets and the clock, not if he doesn't notice that the two score sheets are different.

The arbiter knows that there are a different number of moves on the score sheets and the clock's move counter. However, in this scenario, the conclusion is that the players had played on despite playerA's flag having fallen at move38. A difference in the two score sheets further reinforces the arbiter's view that the scoresheets are not reliable. Further weakening playerA's chances of getting the clock's move counter evidence rejected.


I am not aware that FIDE has not endorsed a clock solely because the move counter was clearly visible all the time. Do you?
What is happening is that DGT and other clock producers are moving towards putting the move counter displayed directly. But the endorsement of this is already part of the laws as per 6.14. In other words, FIDE is not moving towards anything new here.

I'm in favour of the move counter being displayed constantly. Makes errors detectable faster.

It seems to be quite a new thing for the chess clock to have a move counter visible constantly. Refer to http://www.fide.com/fide/minutes/5050-81st-fide-general-assembly-minutes-and-annexes.html Annex38, Technical Commission report.

1. Information about TC activities since 2009
The main task of the Commission was testing of chess clocks. There are requests to include the check of
chess equipment to the duties of TC as well.
Mr. Mikko Markkula informed about his way to test chess clocks. He wants to have the time for the
second period added on a clock immediately when the preset number of 40 moves was completed for the
first period by a player playing white and then after of 40 moves completed by a player playing black.
Chairman Andrzej Filipowicz gave his view, that after move 40 the clock should add the time for the
following period. Furthermore he cannot see any reason why the clock should not show the number of
moves on its display.
The testing report for electronic clocks has to be updated. The new guidelines will be forwarded to the
FIDE General Assembly.

It was an issue for the move counter to be constantly displayed until amended at 81st FIDE General Assembly Khanty-Mansiysk.
Corrected a ludicrous situation where all clocks constantly showed the number of moves except THE chess clock.

Also, it is quite obvious that the move counter is meant to be activated to add time after move40. Why else would so much effort be spent on testing this particular ability before a clock can be FIDE endorsed?


Unfortunately the move counter on the clock is directly mentioned in 6.14. If a flag fall of the clock is based on the move counter, then that fact is no more a fact than the rest of 6.14.

You need to understand that article6.14 does NOT refer to the chess clock that players are using. It refers to the display clocks that the spectators look at.

Hence the last sentence, to prevent a player using such information to trump the official equipment.


I would hope FIDE would want arbiters to have common sense. In the exams, hopefully they are tested whether they have chess arbitrarial common sense. As a matter of fact, I do think that this is tested.

If there are regulations for a situation, an arbiter is expected to apply the regulations strictly.

In the exams, the only way to test the candidates fairly is to set the questions where there can be only one correct answer to a given situation as it is regulated for. There are no "common sense" answer type questions as they are unfair. What passes as common sense for candidates from one part of the world may well be the exact opposite to the examiner.

No, you thought wrong, arbitrarial common sense is not tested. I've been through the exam, and passed, they only want mechanical answers. I noticed that the unsuccessful candidates were the types that would think too much common sense into decision making. I'd want to hire them to run my business, but unfortunately they don't make the cut to be arbiters. Come to think of it, there's a chance that I may have passed because I was too jet-lagged and sleep deprived to have much common sense then!!


You are being very mechanical here. You seem to think that the preface does not use the word "common sense" then it doesn't care about it. To me the whole paragraph beams of "common sense". How could you "find the solutions to a problem dictated by fairness, logic and special factors" without using common sense? To me they are saying arbiters should look beyond there own little nosetip.

Notice the "Where cases are not precisely regulated..." clause?
Essentially kills all the feel good guff. If an article regulates a situation, common sense decisions are not permitted. The arbiter is to strictly follow the regulations.

The wording is clear. First, determine if a situation is regulated. If it is, apply the regulation strictly. If not, then "common sense" decision making is allowed ("fairness, logic and special factors" clause).

This results in arbiters having a very mechanical algorithm to follow. The way to make them snap out of it is to have a situation that is not regulated. Then the human being with common sense emerges from the robotic arbiter!


The clocks must not allow the player to use all his time in the last period, and then recover from that by making more moves - the flag fall should stand. You think that players can get away with what should have been a flag fall in a period other than the last, when the clock was assigned 0 moves for this period. I invite you to show an example.

Rate of play, 40moves/90minutes, +30minutes, with 30seconds increment from move one.

Both playerA and playerB are clicking the clock and recording their moves accurately.
On move38 playerA's flag fell, but because move counter is set to 0, extra 30 minutes automatically added.
Neither player observes flag fall (transition into 2nd time period).
Both players continue playing at a faster pace until move42.
Arbiter notices playerA is in next time control, checks scoresheets and since it is move42, concludes no flag fall.

Hasn't playerA just got away with what should have been a flag fall?


If you care to go back in the thread, Bill Gletsoes shows an example where it is not possible to determine with multiple flag falls in the first period which flag actually fell first, using 40 moves for the first period. This does not happen with 0 moves for first period.

Yes, this is the situation where clicking too many times by the player causes problems. That is the scenario that setting the move counter causes problems. During the FA seminar, the training is for arbiters to regularly patrol the playing area with clock and move check sheets. We write down the clock times and move number on the scoresheets. The idea is to pick up discrepencies between clock times and scoresheets move numbers. This is how we detect the frantic clock clickers. Also a regular procedure to keep arbiters in compliance with article13.3

Your scenario is where clicks are missing. Something where activating the move counter actually works.


Yes Garvinator stated that the arbiter should never correct for the players faulty clock presses. I don't agree with that, I have only said the precaution that it should be on a claim or on a reconstruction or on evaluating a flag fall.

We're in agreement here.
Faulty (or forgotten) clock presses need to be corrected and adjusted for.

One thing to note is that the patrolling with clipboard can also detect missed clock presses. However, as Garvinator correctly points out, article13.6 binds the arbiter here. My earlier post was a situation where a player makes an article6.10.a claim before flag fall, hence the need to adjust for errors.


Hopefully a decision with common sense, even though they are not gods.

Actually you don't have to reconstruct the whole game to determine there are discrepancies in the current game where White forgot two half-moves at move 12 and Black forgot two half-moves at move 26

(Yue-Carlsen, Medias 2010, but they didnt make the scoring mistakes, I invented them!)
White's Score Black's Score
1. d4 Nf6 --- 1. d4 Nf6
2. c4 g6 --- 2. c4 g6
... ...
11. Bxc4 Nd7 --- 11. Bxc4 Bxc4
12. Qd2 Qa5 --- 12. 0-0 Nd7
13. Rfd1 Rad8 --- 13. Qd2 Qa5
... ...
25. Nxc3 Ne5 --- 25. e6 Bxc3
26. Red1 Kf8 --- 26. Red1 Kf8
27. Rac1 Ke7 --- 27. Rac1 Ke7

It should be clear just from quickly skimming and comparing the two scores that they are not identical. A reconstruction would now be relevant, and would reveal that white missed 11...Bxc4 12.0-0 and black missed 26.Nxc3 Ne5, so the whole game is one move longer than both score sheets. The score sheets will only be identical at each point even though two half-moves had been omitted, if the players in fact made the same mistake at the same point. To verify all other situations simply compare the two score sheets, and if they are identical, we are 99% sure that no moves are lacking.

In my opinion flag falls are happening infrequently in a period that is not the last, and so the arbiter should be very sure to get it right. There must be no doubts, e.g. no scenario could reach the facts presented without the flag fall being in effect. Imagine you play a game and the arbiter decides you lose the game because you had only completed 39 moves, then you go home and put the game into the chess database and you discover that 40 moves had been completed. I would be very pissed in this situation, even if I technically could be said to not have fulfilled all obligations (of 8.1).

If the player had been noting down clock times in the scoresheet, such an issue would not arise and it would be easy for the arbiter to adjust an incorrectly frozen clock.

If I were the arbiter and you had observed your flag fall and claimed to me that it fell in error, a quick look at your scoresheet and the clocks move counter would immediately result in the scoresheet trumping the clock.
The same situation and result would occur if I, as arbiter, had observed your flag fall when it happened.

If flag fall was not observed until some indeterminate time after it happened, without clock times on the scoresheet, the move counter of the clock would trump the scoresheet because all moves played after the move where flag fall occurred are invalid. Even if it was discovered that a move was not recorded, making the position on the board at move39 actually move40, it means that the position on the board when the flag fell was that of move38 on the scoresheet. There is no evidence to prove missing clock clicks, only missed move recording on the scoresheet is detectable. If you were the player, you've just added an infraction of article8.1 to the list of mistakes.
I think as the player putting the game into the chess database, you would be thinking that you actually lost on time at move38 and actually played on till move40 (instead of move39) without noticing it.

Bill Gletsos
04-11-2011, 04:33 PM
The situation is simple.

The clock should only add the second time control once one clock has reached 0.

In fact this is exactly what the latest DGT 2010 model (Blue buttons) does.

It implements the FIDE time controls 90 minutes for 40 moves + 30 minutes with 30 seconds cumulative increment for each move starting from the first move and 100 minutes for 40 moves followed by 50 minutes for 20 moves, then 15 minutes for the remaining moves with 30 seconds cumulative increment for each move starting from first move as options 19 and 20 respectively.

Jesper Norgaard
06-11-2011, 04:54 PM
The situation is simple.

The clock should only add the second time control once one clock has reached 0.

In fact this is exactly what the latest DGT 2010 model (Blue buttons) does.

It implements the FIDE time controls 90 minutes for 40 moves + 30 minutes with 30 seconds cumulative increment for each move starting from the first move and 100 minutes for 40 moves followed by 50 minutes for 20 moves, then 15 minutes for the remaining moves with 30 seconds cumulative increment for each move starting from first move as options 19 and 20 respectively.
I completely agree.

I had a look on the manual for the DGT 2010 clock, and it states:
"In all Bonus options: If a player has used the last period playing time, and no time is added anymore, the clock stops, and also the clock of the opponent blocks and can not set to count down anymore. End of the game!"
I agree with that. In the last period where number of moves doesn't matter flag fall means end of game, unless no sequence of legal moves can lead to mate (for instance knight against queen).

The manual states specifically about options 19 and 20:
"In the preset options 19 and 20, each with two time periods, the second period starts when one player used all the first period time, and passes zero. Both players receive the second period Basic time added on that moment."
I don't know if this is a "clever" way to say that in options 19 and 20, the default is zero moves for the first period, but if it is the intention, it is quite clumsy.

It might be that options 19 and 20 preset will make 0 moves for the first period, but you can certainly change that to 40 moves quite easily if you want. However, this is only mentioned in the paragraph about 4 different time periods:
"For the periods 1 to 3, a move number can be programmed. If the move number is set to a non-zero number, the next period time is added when a player finished the programmed number of moves for that period. If the move number is programmed to the value 0 (zero), the transient to the next period takes place when a player passes zero time."
So they are clearly not saying here which one is default, and which one is preferable. A FIDE decree/rule of either using 0 moves or 40 moves for the first period(s), would clearly be beneficial, to stop all these discussions.

Then the manual goes on to state bombastically:
"If a player does not finish the programmed (non-zero) number of moves for a period, at passing zero time, the blinking flag is shown, and the DGT 2010 stops time counting for both players, indicating that the game has ended.
That player lost the game!"

That is not true in at least 4 different scenarios, and should therefore not have been stated like that:
(1) the clock had a malfunction (remote I admit)
(2) the moves on both score sheets show that the number of moves had been completed
(3) there is discrepancy of moves between the two score sheets
(4) there is no mating material

Situation (3) would require a reconstruction of the game to reach a
correct number of moves completed

In all situations, the indications of the clock of a fact (flag fall, number of moves clicked etc.) must be taken into consideration for an arbiter's decision, but cannot be the only source of information. It would be like when finding a certain persons DNA on the murder scene, and the jury thinking "he was the killer" even if this DNA had arrived (or could have arrived) before the murder took place.

If the clock shows 38 as move counter, you can conclude that 38 clicks on the clock had been made, you cannot include that 38 moves had been made alone from that fact, without consulting the score sheets.

Let's not pretend the clock can be the arbiter. And it sure would help if FIDE recommended 0 moves or x moves as a standard for several periods. (x could be 40 for first period, 20 for second etc.)

Jesper Norgaard
06-11-2011, 05:24 PM
And the move counter trumps the scoresheets in this case.
All moves after move38 are discounted as "occuring after flag fall".

Now the situation and decision would be different if flag fall was observed as it happened. Or the move counter discrepancy was detected before flag fall. Corrections can then be made.

In fact the FIDE rules show that the two situations are not different. Rule 6.8 states "A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid
claim to that effect." That means that if the flag fall was observed when the players had recorded 40 moves on the score sheet, then we can't say anything about when the missing clock presses happened, before or after the flag fall. The game needs to continue, and the clocks adjusted for the 2 missing clock presses.



I'm in favor of the move counter being displayed constantly. Makes errors detectable faster.

I completely agree with the recommendation and your reason. Perhaps it would have saved Nakamura of an embarrassing "fetch-orange-juice-flag-fall"? He should have known that he could not rely on the clock for that, or ask the arbiter. He needs to determine it based on his score sheet!

1. Information about TC activities since 2009
The main task of the Commission was testing of chess clocks. There are requests to include the check of
chess equipment to the duties of TC as well.
Mr. Mikko Markkula informed about his way to test chess clocks. He wants to have the time for the
second period added on a clock immediately when the preset number of 40 moves was completed for the
first period by a player playing white and then after of 40 moves completed by a player playing black.
Chairman Andrzej Filipowicz gave his view, that after move 40 the clock should add the time for the
following period. Furthermore he cannot see any reason why the clock should not show the number of
moves on its display.

I don't disagree with that 40 moves instead of 0 moves for the first period could become a FIDE standard, it just needs to be official. It is not official yet.

Keong Ang
07-11-2011, 11:18 AM
In fact the FIDE rules show that the two situations are not different. Rule 6.8 states "A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid
claim to that effect." That means that if the flag fall was observed when the players had recorded 40 moves on the score sheet, then we can't say anything about when the missing clock presses happened, before or after the flag fall. The game needs to continue, and the clocks adjusted for the 2 missing clock presses.

I was alluding to two different situations where the same rules are applied to different effect.

1.
Flag fall observed as it happened, and the player had 40 moves recorded on the score sheet.

2.
Flag fall discovered sometime after it happened, the player had 40 moves recorded on the score sheet but nobody knows exactly when the flag fell.

You correctly pointed out one article to apply, article6.8.
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

The other article to apply is article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

In situation1, the arbiter knows exactly when the flag fell and examining the scoresheets would show that 40 moves had been made. The move counter on the clock is wrong.

In situation2, the arbiter cannot know or find out when the flag fell. Logically the conclusion is that the players had played a number of extra moves after the flag fell. Article6.10.a. forces the arbiter to accept the chess clock's move counter as absolutely correct in this case.

Here we have 2 similar situations where the only difference is when the flag fall was observed, and the results are very different. In situation1, the game goes on and the clocks are adjusted, in situation2, the game has ended and the player whose flag fell has lost the game.

Over the past few months, I've been advised that my decision in cases like situation1 is too lenient and liberal in applying article6.10.a. Procedurally, if the clock is registering presses, the move counter on the clock must be accepted as correct. If the arbiter decides that the clock is defective, the clock MUST be replaced with another clock.

All this based on the interpretation that it is compulsory for players to press the clock. Forgetting to press the clock requires penalties to be imposed. Flag fall on the clock is essentially a convenient means to penalise the player for forgetting to press the clock. I'm not comfortable with this "hard line" interpretation, but the wording of the regulations seem to stipulate this.


I completely agree with the recommendation and your reason. Perhaps it would have saved Nakamura of an embarrassing "fetch-orange-juice-flag-fall"? He should have known that he could not rely on the clock for that, or ask the arbiter. He needs to determine it based on his score sheet!

1. Information about TC activities since 2009
The main task of the Commission was testing of chess clocks. There are requests to include the check of
chess equipment to the duties of TC as well.
Mr. Mikko Markkula informed about his way to test chess clocks. He wants to have the time for the
second period added on a clock immediately when the preset number of 40 moves was completed for the
first period by a player playing white and then after of 40 moves completed by a player playing black.
Chairman Andrzej Filipowicz gave his view, that after move 40 the clock should add the time for the
following period. Furthermore he cannot see any reason why the clock should not show the number of
moves on its display.

I don't disagree with that 40 moves instead of 0 moves for the first period could become a FIDE standard, it just needs to be official. It is not official yet.

Perhaps it is not official so that all types of clocks can be used. Clocks with move counters will function just as well if move counters are to be disabled. However, if FIDE makes it official that chess clock move counters must be used, we'll need to discard all clocks that are incapable of doing so, or use time controls that do not use number of moves. It is already an official requirement for the chess clocks to be able to regulate the rate of play used.

My view is that all this effort in checking the correct functioning of move counters when chess clocks are being certified means that we're supposed to be using them. We can choose not to because it is not compulsory, so we're quite free in this regard.

What I find puzzling is why should it be such a controversy?
During my FA training we were instructed in detail and depth about the chess clock's move counter and correct usage. Also, the FA exam answers on arbiter actions in various situations tend to require the words "check chess clock's move counter" or "adjust chess clock's move counter". In my opinion, it better be official enough since prospective arbiters could fail to obtain their FIDE titles over something as mundane as the move counter.

Anyway, I find it more practical to set the move counter. Saves work and makes tournament supervision easier.

Jesper Norgaard
07-11-2011, 12:55 PM
I was alluding to two different situations where the same rules are applied to different effect.

Everything indicates that we are looking at exactly the same two different situations. It is the interpretation that is different. In the current situation, there is no malfunction of the clock. It shows exactly what it recorded during the game. However, you interpret it to tell you that at move 38 the clock fell. What is the only correct interpretation is that at clock press 38 the clock fell. That gives a big difference in the different scenarios that could have happened.



In situation1, the arbiter knows exactly when the flag fell and examining the scoresheets would show that 40 moves had been made. The move counter on the clock is wrong.

Yes and that could have been the situation in situation 2 also, except that it fell a split second before the arbiter could notice it, so he doesn't know if it fell right now or 2 minutes ago. More likely it would be the players failing to make clock presses earlier than move 38 (on the board), perhaps as early as the first hurried theory moves played out. The main thing to observe is that according to rule 6.8, the flag fall is not considered to have happened until the moment that the arbiter noticed it, therefore when both players had made 40 moves on the score sheets, therefore no loss on time exists.



In situation2, the arbiter cannot know or find out when the flag fell. Logically the conclusion is that the players had played a number of extra moves after the flag fell. Article6.10.a. forces the arbiter to accept the chess clock's move counter as absolutely correct in this case.

The clock's move counter is absolutely correct, it's your conclusion that is wrong. The clock cannot tell you if moves were played after flag fall or not.



Over the past few months, I've been advised that my decision in cases like situation1 is too lenient and liberal in applying article6.10.a.

Sounds draconian, can you elaborate?



Procedurally, if the clock is registering presses, the move counter on the clock must be accepted as correct.

Yes in counting clock presses, not in counting moves. Your interpretation in situation 2 is not part of FIDE's laws of chess.



If the arbiter decides that the clock is defective, the clock MUST be replaced with another clock.

Correct.



All this based on the interpretation that it is compulsory for players to press the clock. Forgetting to press the clock requires penalties to be imposed.

This conclusion is not part of FIDE's laws of chess. Can you mention a single case of an IA that has penalised a player in an international tournament for not pressing the clock? One single case?



Flag fall on the clock is essentially a convenient means to penalise the player for forgetting to press the clock. I'm not comfortable with this "hard line" interpretation, but the wording of the regulations seem to stipulate this.

Wrong interpretation. It goes against common sense. You should only penalise for situations where there is no reasonable doubt as what has happened. For instance, if player 1 says his opponent touched a piece and moved it to a square and released it, and player 2 says he didn't release it, and there are no witnesses or video surveillance etc. then there is reasonable doubt as to who is right, and the arbiter can't insist on the move player 1 says was completed. Like the Kasparov-Polgar rapid game - incidentally this had a video surveillance that later showed that Kasparov had released the piece for a split second. But he was allowed to play another move with that piece because the video evidence was not ready yet. This is a common situation where in most cases, no video evidence is present. Even if you have a distinct feeling that player 2 is lying, you cannot judge on that alone.



Perhaps it is not official so that all types of clocks can be used. Clocks with move counters will function just as well if move counters are to be disabled.

The move counter is not disabled. The moves will be counted in both cases. But when the first period is set to 0 moves, then the move counter is not used to give time for period 2. The move counter is still active and should be used for correcting added time if claimed by a player.



However, if FIDE makes it official that chess clock move counters must be used, we'll need to discard all clocks that are incapable of doing so, or use time controls that do not use number of moves.

Probably a non-issue because all clocks that allow increment, also allow for zero or forty moves given to first period. Any counter-examples of clocks that cannot handle either mode?



My view is that all this effort in checking the correct functioning of move counters when chess clocks are being certified means that we're supposed to be using them. We can choose not to because it is not compulsory, so we're quite free in this regard.

I think you are right the arbiters are free in this regard, and I think it is a mistake of FIDE. FIDE should recommend one of them. It only leads to confusion among arbiters and players that it can be done differently.



What I find puzzling is why should it be such a controversy?
During my FA training we were instructed in detail and depth about the chess clock's move counter and correct usage. Also, the FA exam answers on arbiter actions in various situations tend to require the words "check chess clock's move counter" or "adjust chess clock's move counter". In my opinion, it better be official enough since prospective arbiters could fail to obtain their FIDE titles over something as mundane as the move counter.

Of course an arbiter must check and adjust the move counter in certain situations. If both players have time at move 35, and one player claims that only 33 moves have been pressed on the clock, the arbiter should adjust for the missing 2 clock presses on both clocks, and penalise nobody. After all, both players were making a mistake by not pressing the clock. But no penalty is prescribed for not pressing the clock in FIDE's laws of chess.



Anyway, I find it more practical to set the move counter. Saves work and makes tournament supervision easier.
What you mean is that you find it more practical to set 40 moves for the first period, so the clock will automatically add the time for the next period when 40 clock presses have been noticed. Why that would save work is not clear to me, after all in both cases when the time is added it happens automatically, and requires no intervention of the arbiter. What do you save?
In fact it is still your duty as arbiter to check that 40 moves have been played, even if the players by mistake put more clock presses in, and the clock therefore indicates that 40 moves have been played. Only the score sheet is the final evidence of 40 moves actually have been played.

Jesper Norgaard
07-11-2011, 08:38 PM
And the move counter trumps the scoresheets in this case.

Can you point out a single international tournament case where move counter on the official chess clock trumped the score sheets?



Article6.14 is formulated to allow players to get around article12.3.a when they cannot help noticing information displayed for spectators in the "playing hall". A very badly worded article.

I don't see the problem.

For reference here is the full 6.14 article
"Screens, monitors, or demonstration boards showing the current position on the chessboard, the moves and the number of moves made, and clocks which also show the number of moves, are allowed in the playing hall. However, the player may not make a claim relying solely on information shown in this manner."



As an arbiter, the use of the word "clocks" with the "s" at the end of it definitely means that it is not referring to the chess clock that players are meant to click. This is in keeping with the meaning of the whole of article6.14. Anyway, it's a clause that applies to the player and not the arbiter."

Have a look at article 6.13:
"If an irregularity occurs and/or the pieces have to be restored to a previous position, the arbiter shall use his best judgement to determine the times to be shown on the clocks. He shall also, if necessary, adjust the clock’s move counter."
In the same paragraph clocks and the clock are used interchangeably! It is exactly the same case as a move - sometimes it means one single move from one player sometimes it means one move from each player. The reason why clocks is plural is because there are two separate time measuring devices (clocks) in a single chess clock.

My interpretation is that "Screens, monitors, or demonstration boards showing the current position on the chessboard, the moves and the number of moves made" does in fact cover all external sources of move information. The clocks that is mentioned thereafter must be the chess clock itself. The whole article should be interpreted the way, that if the move counters either on screens, or monitors, or demonstration boards, or the official chess clock, shows that 38 moves have been made, but a player's score sheet shows 40 moves have been made, he can't make a claim of flag fall. There is no way a flag fall that happened after the required number of moves have been made, can ever cause a player to lose on time. Adding move counters does not change that. And 6.8 specifies that the flag is considered to have fallen when it is observed by the arbiter or claimed by a player. Which part of that clause is it you don't understand? It's robotic nature should please you.

I agree that 6.14 only binds the player, not the arbiter. However, it would be strange indeed if information that the player can't use solely to start an arbiter case, can be used as the only and conclusive evidence for the arbiter himself. My common sense is hurting!

Bill Gletsos
07-11-2011, 08:55 PM
I was alluding to two different situations where the same rules are applied to different effect.

1.
Flag fall observed as it happened, and the player had 40 moves recorded on the score sheet.

2.
Flag fall discovered sometime after it happened, the player had 40 moves recorded on the score sheet but nobody knows exactly when the flag fell.

You correctly pointed out one article to apply, article6.8.
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

The other article to apply is article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

In situation1, the arbiter knows exactly when the flag fell and examining the scoresheets would show that 40 moves had been made. The move counter on the clock is wrong.

In situation2, the arbiter cannot know or find out when the flag fell. Logically the conclusion is that the players had played a number of extra moves after the flag fell. Article6.10.a. forces the arbiter to accept the chess clock's move counter as absolutely correct in this case.

Here we have 2 similar situations where the only difference is when the flag fall was observed, and the results are very different. In situation1, the game goes on and the clocks are adjusted, in situation2, the game has ended and the player whose flag fell has lost the game.Wrong.
In both situation 1 & situation 2 the game continues.


Over the past few months, I've been advised that my decision in cases like situation1 is too lenient and liberal in applying article6.10.a. Procedurally, if the clock is registering presses, the move counter on the clock must be accepted as correct.Wrong.

If the arbiter decides that the clock is defective, the clock MUST be replaced with another clock.Correct.


All this based on the interpretation that it is compulsory for players to press the clock.Wrong.

Forgetting to press the clock requires penalties to be imposed.Wrong.

Flag fall on the clock is essentially a convenient means to penalise the player for forgetting to press the clock. I'm not comfortable with this "hard line" interpretation, but the wording of the regulations seem to stipulate this.Wrong.


My view is that all this effort in checking the correct functioning of move counters when chess clocks are being certified means that we're supposed to be using them.Wrong.


What I find puzzling is why should it be such a controversy?
During my FA training we were instructed in detail and depth about the chess clock's move counter and correct usage. Also, the FA exam answers on arbiter actions in various situations tend to require the words "check chess clock's move counter" or "adjust chess clock's move counter". In my opinion, it better be official enough since prospective arbiters could fail to obtain their FIDE titles over something as mundane as the move counter.The need to check the move counter is to ensure that the clock is indeed correctly adding the increment time.


Anyway, I find it more practical to set the move counter. Saves work and makes tournament supervision easier.However this behaviour is incorrect.
The clock should only add the second time control once one clock reaches zero time.

ER
07-11-2011, 10:32 PM
Hey Keong Ang
You got just about one out of seven correct followed by a couple or three doubtful ones. :eek: :rolleyes: :lol:

Pepechuy
08-11-2011, 03:34 AM
Let's not pretend the clock can be the arbiter. And it sure would help if FIDE recommended 0 moves or x moves as a standard for several periods. (x could be 40 for first period, 20 for second etc.)

http://www.fide.com/component/handbook/?id=66&view=article
In particular:

1.3 Where a certain number of moves is specified in the first time control, it shall be 40 moves. Players benefit from uniformity here.

Jesper Norgaard
08-11-2011, 10:40 AM
http://www.fide.com/component/handbook/?id=66&view=article
In particular:

1.3 Where a certain number of moves is specified in the first time control, it shall be 40 moves. Players benefit from uniformity here.
Thanks for the input, but I believe it is not relevant. The issue being discussed here is whether the clock should be set to 0 moves or 40 moves in the event that there are 40 moves in the first time control period. The article simply states that it is better to have 40 moves for first time control (e.g. in tournament invitation) than 35 or 50 moves for the first period. It says nothing about how clocks should be set.

Jesper Norgaard
08-11-2011, 11:48 AM
Anyway, I find it more practical to set the move counter. Saves work and makes tournament supervision easier.



However this behaviour is incorrect.
The clock should only add the second time control once one clock reaches zero time.
Wrong. ;)

FIDE does not endorse any of the two ways to set the time. Bill Gletsos, please put forward any information that would indicate otherwise. I have found videos of rather old origin on youtube where instructions of settting a DGT clock mention that it must be set to zero rather than number of moves according to FIDE. But that is very far from being an official recommendation which should of course be viewable straight off the FIDE web site if it existed.

So apart from my tongue-in-cheek comment that this is wrong, actually both are correct.

The most damaging side-effect of using 40 moves set is that when there are too few clock presses from player mistakes, then the clock will freeze going into move 40 and if the players play on without noticing the clock freeze, all their actions in terms of moves and accumulated thinking time, is lost.

However, that is only a problem because clock manufacturers have gotten the idea that the clock should freeze - it only should in the last period. Why? Because too few or too many clock presses from the players is a common phenomena, so if these are present the clock will make wrong conclusions on when the next period has passed, or that a player overstepped the time limit. As far as I believe there is not a single clock on the market that does not freeze if first period is set to 40 moves rather than zero. If you could set it to not freeze in the not-last periods, then it would make more sense to me to use this mode, but this seems to be hardwired in all the clocks no exception.

antichrist
08-11-2011, 12:15 PM
Wrong. ;)

FIDE does not endorse any of the two ways to set the time. Bill Gletsos, please put forward any information that would indicate otherwise. I have found videos of rather old origin on youtube where instructions of settting a DGT clock mention that it must be set to zero rather than number of moves according to FIDE. But that is very far from being an official recommendation which should of course be viewable straight off the FIDE web site if it existed.

So apart from my tongue-in-cheek comment that this is wrong, actually both are correct.

The most damaging side-effect of using 40 moves set is that when there are too few clock presses from player mistakes, then the clock will freeze going into move 40 and if the players play on without noticing the clock freeze, all their actions in terms of moves and accumulated thinking time, is lost.

However, that is only a problem because clock manufacturers have gotten the idea that the clock should freeze - it only should in the last period. Why? Because too few or too many clock presses from the players is a common phenomena, so if these are present the clock will make wrong conclusions on when the next period has passed, or that a player overstepped the time limit. As far as I believe there is not a single clock on the market that does not freeze if first period is set to 40 moves rather than zero. If you could set it to not freeze in the not-last periods, then it would make more sense to me to use this mode, but this seems to be hardwired in all the clocks no exception.

so do you think there is still considerable value in using the old anologue clocks? no complications of that I can remember

Bill Gletsos
08-11-2011, 12:18 PM
Wrong. ;);)

FIDE does not endorse any of the two ways to set the time. Bill Gletsos, please put forward any information that would indicate otherwise. I have found videos of rather old origin on youtube where instructions of settting a DGT clock mention that it must be set to zero rather than number of moves according to FIDE. But that is very far from being an official recommendation which should of course be viewable straight off the FIDE web site if it existed.I cannot recall where but I remember reading it years ago.

Whether the prescibed number of moves have been made or not is determined when the situation in article 6.8 occurs and not when a players scoresheet indicates x number of moves have been made.

Once the situation is article 6.8 occurs, the score sheets are checked to determine if the prescribed number of moves have been made and if they have then any additional time is added at that point.

It is not case case that once x moves have been made that the additional time control is added.

This was the correct procedure when using manual clocks and was carried over when digital clocks were introduced.

antichrist
08-11-2011, 12:24 PM
;)
I cannot recall where but I remember reading it years ago.
............................................

This was the correct procedure when using manual clocks and was carried over when digital clocks were introduced.

well if it still necessary for the dops to get off their backside at 40 moves what advantage is with digital clocks? 500 headaches? as well the great percentage of chess population absolutely refuses to learn them and would happily consign them to the junk heap - okay mercy kill them

Bill Gletsos
08-11-2011, 12:31 PM
well if it still necessary for the dops to get off their backside at 40 moves what advantage is with digital clocks? 500 headaches? as well the great percentage of chess population absolutely refuses to learn them and would happily consign them to the junk heap - okay mercy kill themYou clearly do not understand.
With the digital clocks it can automatically add the second time control, however it should do so when one cock reaches zero and not after x moves have been made as indicated by the clocks move counter.

Jesper Norgaard
08-11-2011, 12:51 PM
so do you think there is still considerable value in using the old anologue clocks? no complications of that I can remember
No I cant see any value in using the old analogue clocks.
Here is the little list of horrors:

(1) No increment means no protection that some positions need many moves to win/draw, others few moves to win/draw. Often it is impossible to predict how many moves are needed, which would be a precondition to prevent it.

(2) Rule 10.2 makes it possible to get a draw, not a win when being low on time.

(3) Rule 10.2 cannot be invoked in many trivially non-losing or outright winning positions, because of "technicalities".

(4) Guillotine games can be won on time-administering issues alone, disregarding if chess skills would have lead to a win.

This does not mean that the increment solves all problems, and it adds new problems like "when is a flag fall actually a flag fall?". But the advantages of the increment far outweighs the disadvantages in my view.

ER
08-11-2011, 12:58 PM
Hey Keong Ang
You got just about one out of seven correct followed by a couple or three doubtful ones. :eek: :rolleyes: :lol:

LOL I did all the above as a test! Putting myself in an arbiter's role I didn't even get one decision correct! Thank God Bill wasn't my maths teacher! :P :lol: Oh well he wasn't even born then but still! :lol:

Keong Ang
08-11-2011, 03:05 PM
LOL I did all the above as a test! Putting myself in an arbiter's role I didn't even get one decision correct! Thank God Bill wasn't my maths teacher! :P :lol: Oh well he wasn't even born then but still! :lol:

:lol:

JaK, what I'd like to know is how were the clocks set when the rate of play has 2 time controls where the 2nd time control is entered after a prescribed number of moves. eg. 90minutes plus 30minutes at move40.

As a participant in tournaments outside Aus/NZ, did you notice how the clocks were set? All assuming you had played in tournaments with such time controls in the past year.
;)

ER
08-11-2011, 06:07 PM
:lol:

JaK, what I'd like to know is how were the clocks set when the rate of play has 2 time controls where the 2nd time control is entered after a prescribed number of moves. eg. 90minutes plus 30minutes at move40.

As a participant in tournaments outside Aus/NZ, did you notice how the clocks were set? All assuming you had played in tournaments with such time controls in the past year.
;)

LOL I haven't; usually I don't even care how the clocks are set in the first place. :lol:
If you asked me what the time control was before, during and after the tournament, I wouldn't even know! (*)
I don't even know how to start the clock properly since there are too many buttons these days.
Laurence Matheson and some other kids at Box Hill CC tried to show me once, I learnt how to set the blitz (was it on 5 or something?) then I forgot again.
I know how to put the clock off though (the little button at the back) so it doesn't spend battery life for nothing! :owned:
I 've noticed in tournaments (including a game b/n two IMs) the clocks went wrong and wouldn't add the time or something like that!
(*) Life is too short to lose a game on time! :owned:

antichrist
08-11-2011, 06:20 PM
LOL I haven't; usually I don't even care how the clocks are set in the first place. :lol:
If you asked me what the time control was before, during and after the tournament, I wouldn't even know! (*)
I don't even know how to start the clock properly since there are too many buttons these days.
Laurence Matheson and some other kids at Box Hill CC tried to show me once, I learnt how to set the blitz (was it on 5 or something?) then I forgot again.
I know how to put the clock off though (the little button at the back) so it doesn't spend battery life for nothing! :owned:
I 've noticed in tournaments (including a game b/n two IMs) the clocks went wrong and wouldn't add the time or something like that!
(*) Life is too short to lose a game on time! :owned:

AC
Exactly as I was saying: ............. as well the great percentage of chess population absolutely refuses to learn them and would happily consign them to the junk heap - okay mercy kill them

I dont even like how you cant see them clearly from an angle, you must lean right over to read them which is distracting and irritating, not just a quick squint out of the side of your eye like with anologue, you cant get out of a winning game to watch a great game on board one finish. Anyway getting back on topic - surely the clock must be added onto after 40 moves on the scoresheet regardless on how many clicks the clock says. And surely both players score lists would be identical.

antichrist
08-11-2011, 06:50 PM
No I cant see any value in using the old analogue clocks.
Here is the little list of horrors:

(1) No increment means no protection that some positions need many moves to win/draw, others few moves to win/draw. Often it is impossible to predict how many moves are needed, which would be a precondition to prevent it.

(2) Rule 10.2 makes it possible to get a draw, not a win when being low on time.

(3) Rule 10.2 cannot be invoked in many trivially non-losing or outright winning positions, because of "technicalities".

(4) Guillotine games can be won on time-administering issues alone, disregarding if chess skills would have lead to a win.

This does not mean that the increment solves all problems, and it adds new problems like "when is a flag fall actually a flag fall?". But the advantages of the increment far outweighs the disadvantages in my view.

I don't consider any of the above a disadvantage or a horror. If you lose on time then so be it, you must learn how to manage time better or lose to the superior player.

I have read that the results at top end of town are not changing due to time increment, so it seems that the stronger players before increment time also improved their game with the extra time or some other reason that I don't realise. Only the games became longer and more drawn out for no good reason, except maybe for the losers they dont appear to embarrassing lose out by getting behind in the time race that is part of any timed chess.

What is grossly unfair how quicker thinkers like myself lose a natural advantage to slow thinkers who should be out the door quick smart so I can either have a rest or watch the top boards.

NB I forgot an anology: digital clocks are like dyson vacuum cleaners, completely unnecessary, expensive,need a degree to run and strong shoulders and patience to achieve no better result

Keong Ang
08-11-2011, 08:16 PM
Wow, after months of inactivity, Bill Gletsos has succeeded in bumping this thread into activity! :cool:

Regarding it being compulsory for players to press the clock to complete their move (with some exceptions):
Laws of Chess, article6.7.a.
During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See the Articles 5.1.a, 5.2.a, 5.2.b, 5.2.c and 9.6)
The time between making the move on the chessboard and stopping his own clock and starting his opponent‘s clock is regarded as part of the time allotted to the player.
"During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock." is the sentence that makes it compulsory for the player to 'press the clock'.

Forgetting to press the clock is an infraction of the Laws of Chess. This leads on to article13.1
The arbiter shall see that the Laws of Chess are strictly observed.
That makes it compulsory for the arbiter to enforce the laws with the empowering option of applying penalties according to articles13.4.a-g where appropriate.

The laws state very clearly that it is compulsory for players to press the clock to complete their move, if it was not, where is the clause that says so? We're not talking about a move that ends the game here...


Regarding setting the clock's move counter to regulate the passing of each time control:

I have not found any official FIDE clause that says something like "...if digital clocks are used the move counter <must/must not> regulate the passage of time controls..." or words to that effect. Which is what this thread is all about!!

So, if someone can point us to the relevant clause, it would be a big help.

Meanwhile these seem to give clues to the direction FIDE is pointing us to:
FIDE Tournament Rules article10.c.1.
The display at all times should show the time available to complete a player’s next move.
FIDE Tournament Rules article10.c.4.
In case of passing a time control, a sign on the display must give clear signal which player passed the time limit first.
FIDE Tournament Rules article10.c.7.
Special attention should be given to the correct announcement of passing time controls.
FIDE Tournament Rules article10.c.8.
In case of accumulative or delay timing systems, the clock should not add any additional time if a player passed the last time control.
FIDE Tournament Rules article10.c.10.
It must be impossible to erase or change the data in display with a simple manipulation.

FIDE Tournament Rules article10.c.1. alone should be sufficient to require setting the move counter to 40 instead of 0.
"The display at all times should show the time available to complete a player’s next move."
If it is move 40+, the time displayed on the clock should already include the extra time added. When the player is supposed to have 32minutes remaining, the clock should show 32minutes remaining, not 2minutes and the player then has to figure out if another 30minutes would be added during zero time.

Articles10.c.4&7 supports this because the clock cannot display any of these if the move counter was not set to regulate the passing of each time control. If the rate of play announced said, 90minutes plus 30minutes at move40, the displays of the clock must show that the 2nd time control has been added at move40.

Articles 10.c.8&10 essentially preserves clock evidence and also mandates the "lock-up" or "freeze" function.

I believe the interpretation that we're meant to "set the move counter to 40 instead of 0" is correct in light of the testing procedure of clocks before they get FIDE endorsed. It seems in line with all the regulations.

When games are being broadcast live, we certainly need to set those DGT XL clocks move counters. If we use 0, the time remaining being broadcast would be wrong. Also, all the time stamps on the PGN files would be wrong as well.

In earlier posts, most recently by Jesper, the testing procedure before FIDE endorses a clock is quoted. Why should clocks be tested for proper passage of time controls at move40 with the move counter function if we are not meant to use it and simply leave it at zero?


Regarding clocks move counter versus scoresheet:
Laws of Chess article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.
is really stating the obvious. That's why all FIDE endorsed digital clocks lock-up. I know of clocks that did not lock-up and it was the lack of this function that caused them to not get FIDE endorsed.

Clock lock-up is to preserve evidence of the exact time the flag fell. The move counter should not register any extra moves and the other player's remaining time should be frozen. In cases where both flags could fall, the clock that locks-up also makes it possible to tell which flag fell first and therefore negates article6.11.

What the clock says trumps the scoresheet due to article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

If the clock says flag fell at move38, according to the Laws of Chess, that is conclusive. Players forgetting to press the clock is NOT an evident defect of the clock. The scoresheet can show more than 40moves, but the conclusive evidence that the clock presents is that the flag fell at move38 and all subsequent moves played after that, as recorded on the scoresheet occurred after flag fall.

Bill Gletsos
08-11-2011, 08:34 PM
Wow, after months of inactivity, Bill Gletsos has succeeded in bumping this thread into activity! :cool:

Regarding it being compulsory for players to press the clock to complete their move (with some exceptions):
Laws of Chess, article6.7.a.
During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See the Articles 5.1.a, 5.2.a, 5.2.b, 5.2.c and 9.6)
The time between making the move on the chessboard and stopping his own clock and starting his opponent‘s clock is regarded as part of the time allotted to the player.
"During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock." is the sentence that makes it compulsory for the player to 'press the clock'.

Forgetting to press the clock is an infraction of the Laws of Chess. This leads on to article13.1
The arbiter shall see that the Laws of Chess are strictly observed.
That makes it compulsory for the arbiter to enforce the laws with the empowering option of applying penalties according to articles13.4.a-g where appropriate.

The laws state very clearly that it is compulsory for players to press the clock to complete their move, if it was not, where is the clause that says so? We're not talking about a move that ends the game here...


Regarding setting the clock's move counter to regulate the passing of each time control:

I have not found any official FIDE clause that says something like "...if digital clocks are used the move counter <must/must not> regulate the passage of time controls..." or words to that effect. Which is what this thread is all about!!

So, if someone can point us to the relevant clause, it would be a big help.

Meanwhile these seem to give clues to the direction FIDE is pointing us to:
FIDE Tournament Rules article10.c.1.
The display at all times should show the time available to complete a player’s next move.
FIDE Tournament Rules article10.c.4.
In case of passing a time control, a sign on the display must give clear signal which player passed the time limit first.
FIDE Tournament Rules article10.c.7.
Special attention should be given to the correct announcement of passing time controls.
FIDE Tournament Rules article10.c.8.
In case of accumulative or delay timing systems, the clock should not add any additional time if a player passed the last time control.
FIDE Tournament Rules article10.c.10.
It must be impossible to erase or change the data in display with a simple manipulation.

FIDE Tournament Rules article10.c.1. alone should be sufficient to require setting the move counter to 40 instead of 0.
"The display at all times should show the time available to complete a player’s next move."
If it is move 40+, the time displayed on the clock should already include the extra time added. When the player is supposed to have 32minutes remaining, the clock should show 32minutes remaining, not 2minutes and the player then has to figure out if another 30minutes would be added during zero time.

Articles10.c.4&7 supports this because the clock cannot display any of these if the move counter was not set to regulate the passing of each time control. If the rate of play announced said, 90minutes plus 30minutes at move40, the displays of the clock must show that the 2nd time control has been added at move40.

Articles 10.c.8&10 essentially preserves clock evidence and also mandates the "lock-up" or "freeze" function.

I believe the interpretation that we're meant to "set the move counter to 40 instead of 0" is correct in light of the testing procedure of clocks before they get FIDE endorsed. It seems in line with all the regulations.

When games are being broadcast live, we certainly need to set those DGT XL clocks move counters. If we use 0, the time remaining being broadcast would be wrong. Also, all the time stamps on the PGN files would be wrong as well.

In earlier posts, most recently by Jesper, the testing procedure before FIDE endorses a clock is quoted. Why should clocks be tested for proper passage of time controls at move40 with the move counter function if we are not meant to use it and simply leave it at zero?


Regarding clocks move counter versus scoresheet:
Laws of Chess article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.
is really stating the obvious. That's why all FIDE endorsed digital clocks lock-up. I know of clocks that did not lock-up and it was the lack of this function that caused them to not get FIDE endorsed.

Clock lock-up is to preserve evidence of the exact time the flag fell. The move counter should not register any extra moves and the other player's remaining time should be frozen. In cases where both flags could fall, the clock that locks-up also makes it possible to tell which flag fell first and therefore negates article6.11.

What the clock says trumps the scoresheet due to article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

If the clock says flag fell at move38, according to the Laws of Chess, that is conclusive. Players forgetting to press the clock is NOT an evident defect of the clock. The scoresheet can show more than 40moves, but the conclusive evidence that the clock presents is that the flag fell at move38 and all subsequent moves played after that, as recorded on the scoresheet occurred after flag fall.What a load rubbish.

Keong Ang
08-11-2011, 08:56 PM
Yes and that could have been the situation in situation 2 also, except that it fell a split second before the arbiter could notice it, so he doesn't know if it fell right now or 2 minutes ago. More likely it would be the players failing to make clock presses earlier than move 38 (on the board), perhaps as early as the first hurried theory moves played out. The main thing to observe is that according to rule 6.8, the flag fall is not considered to have happened until the moment that the arbiter noticed it, therefore when both players had made 40 moves on the score sheets, therefore no loss on time exists.

Laws of Chess article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

You could be right to interpreting article6.8 as defining when the flag "legally fell".

But you must keep in mind what needs to be done upon flag fall, article6.3.
Immediately after a flag falls, the requirements of article 6.2 a. must be checked.

article6.2.a.
When using a chess clock, each player must make a minimum number of moves or all moves in an allotted period of time and/or may be allocated an additional amount of time with each move. All these must be specified in advance.

So procedurally the arbiter would need to check if the minimum number of moves were made. The scoresheet says "yes", so why did the clock register a flag fall?
Arbiter then checks the clock, and it says "flag fall at move38".

article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

So what the clock says must be conclusive.
If either player protests about clock presses not being registered, the arbiter should test the clock to see if it misses clock presses. If the presses are being counted (a couple of clicks would establish this), there is no evident defect.

The Laws of Chess give no excusable situation for players to forget to press their clock.
article6.7.a.
During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See the Articles 5.1.a, 5.2.a, 5.2.b, 5.2.c and 9.6)
The time between making the move on the chessboard and stopping his own clock and starting his opponent‘s clock is regarded as part of the time allotted to the player.

There is no explicit clause that says a scoresheet's information must overrule the clock. Instead we have a "clock reigns supreme" clause in article6.10.a.

So far, the only clause I can find that would overrule this loss by flag fall situation is,
article8.7.
At the conclusion of the game both players shall sign both scoresheets, indicating the result of the game. Even if incorrect, this result shall stand, unless the arbiter decides otherwise.

If both players sign both scoresheets with draw or swap the player who won, the result shall stand.
Even here, the arbiter could still overrule it.

Keong Ang
08-11-2011, 08:59 PM
What a load rubbish.
:eek: :eek:

You're saying that you would ignore all FIDE regulations in chess tournaments?
:eh:

Bill Gletsos
08-11-2011, 09:03 PM
Laws of Chess article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

You could be right to interpreting article6.8 as defining when the flag "legally fell".

But you must keep in mind what needs to be done upon flag fall, article6.3.
Immediately after a flag falls, the requirements of article 6.2 a. must be checked.

article6.2.a.
When using a chess clock, each player must make a minimum number of moves or all moves in an allotted period of time and/or may be allocated an additional amount of time with each move. All these must be specified in advance.

So procedurally the arbiter would need to check if the minimum number of moves were made. The scoresheet says "yes", so why did the clock register a flag fall?
Arbiter then checks the clock, and it says "flag fall at move38".

article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.

So what the clock says must be conclusive.
If either player protests about clock presses not being registered, the arbiter should test the clock to see if it misses clock presses. If the presses are being counted (a couple of clicks would establish this), there is no evident defect.Complete rubbish.

Whether the number of moves has been made or not is determined by the scoresheet and not the move counter.
The so called move counter is not a move counter at all. it is a press of the clock lever counter.

The Laws of Chess give no excusable situation for players to forget to press their clock.
article6.7.a.
During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See the Articles 5.1.a, 5.2.a, 5.2.b, 5.2.c and 9.6)
The time between making the move on the chessboard and stopping his own clock and starting his opponent‘s clock is regarded as part of the time allotted to the player.

There is no explicit clause that says a scoresheet's information must overrule the clock. Instead we have a "clock reigns supreme" clause in article6.10.a.However the clocks move counter does not reign at all.

If the players score sheets say there has been x moves and the clocks move counter shows less moves, then the problem is almost certainly with the clocks move counter and not the score sheets.

Keong Ang
08-11-2011, 09:15 PM
LOL I haven't; usually I don't even care how the clocks are set in the first place. :lol:
If you asked me what the time control was before, during and after the tournament, I wouldn't even know! (*)
I don't even know how to start the clock properly since there are too many buttons these days.
Laurence Matheson and some other kids at Box Hill CC tried to show me once, I learnt how to set the blitz (was it on 5 or something?) then I forgot again.
I know how to put the clock off though (the little button at the back) so it doesn't spend battery life for nothing! :owned:
I 've noticed in tournaments (including a game b/n two IMs) the clocks went wrong and wouldn't add the time or something like that!
(*) Life is too short to lose a game on time! :owned:

We could spend a long time deciding how to correctly set clocks at Queenstown Classic 2012. :doh:
Since I'm not the chief arbiter there, that will be the two IA's, Bob Gibbons and Shaun Press, problem. :owned:

I'm going to need a refresher when Bob and Shaun gives their FA seminar there. Noticed that the chess clock is a topic that is covered.
:hmm:

I still clearly recall and my FA seminar notes concur that we were trained to always make sure the chess clock move counter was correctly set to the number of moves (40moves) to regulate time controls. It is a compulsory requirement for both FAs and IAs.

Regulations for the Titles of Arbiters article3.4. and article4.5.
Skills to operate electronic chess clocks of different types and for different systems.

Keong Ang
08-11-2011, 09:39 PM
Complete rubbish.

Whether the number of moves has been made or not is determined by the scoresheet and not the move counter.
The so called move counter is not a move counter at all. it is a press of the clock lever counter.
However the clocks move counter does not reign at all.

When you are referring to a situation where article6.10.b. applies
If during a game it is found that the setting of either or both clocks was incorrect, either player or the arbiter shall stop the clocks immediately. The arbiter shall install the correct setting and adjust the times and move counter. He shall use his best judgement when determining the correct settings.
then obviously the scoresheets are used. No disagreement here.

We're discussing a flag fall situation where nobody saw the flag fall when it actually happened.
The first assumption the arbiter makes is that we have 2 players who could have played a number of moves, and recorded them on the scoresheet, AFTER the flag fell.

You seem to keep insisting that the players are not required to "press the clock lever" when the Laws of Chess makes it obligatory.
As long as the clock's move counter is registering every press of the clock lever, it is the move counter that regulates the passage of time controls and increments.


If the players score sheets say there has been x moves and the clocks move counter shows less moves, then the problem is almost certainly with the clocks move counter and not the score sheets.

Not if there's a flag fall and the clock has locked-up.
The score sheets merely show subsequent moves being played AFTER an unnoticed flag fall.

This is why the situation is different if the error was noticed before or during flag fall as we would now know that the score sheets are correct.

Another situation that occurs is when players have overclicked the clock. Would the arbiter make a player lose on time if subtracting incorrect increments or extra time cause flag fall?

Bill Gletsos
08-11-2011, 10:00 PM
When you are referring to a situation where article6.10.b. applies
If during a game it is found that the setting of either or both clocks was incorrect, either player or the arbiter shall stop the clocks immediately. The arbiter shall install the correct setting and adjust the times and move counter. He shall use his best judgement when determining the correct settings.
then obviously the scoresheets are used. No disagreement here.

We're discussing a flag fall situation where nobody saw the flag fall when it actually happened.
The first assumption the arbiter makes is that we have 2 players who could have played a number of moves, and recorded them on the scoresheet, AFTER the flag fell.

You seem to keep insisting that the players are not required to "press the clock lever" when the Laws of Chess makes it obligatory.There is nothing obligatory about it, nor is there any penalty for a player who does not do out, other than the wasting of their own time.

ER
08-11-2011, 10:03 PM
We could spend a long time deciding how to correctly set clocks at Queenstown Classic 2012. :doh:
Since I'm not the chief arbiter there, that will be the two IA's, Bob Gibbons and Shaun Press, problem. :owned:

I'm going to need a refresher when Bob and Shaun gives their FA seminar there. Noticed that the chess clock is a topic that is covered.
:hmm:

I still clearly recall and my FA seminar notes concur that we were trained to always make sure the chess clock move counter was correctly set to the number of moves (40moves) to regulate time controls. It is a compulsory requirement for both FAs and IAs.

Regulations for the Titles of Arbiters article3.4. and article4.5.
Skills to operate electronic chess clocks of different types and for different systems.

Thanks for the general information and the news that Bob and Shaun will hold an FA seminar there! Looking forward to meet you there and have a drink and a chat! :)

Bill Gletsos
08-11-2011, 10:18 PM
Not if there's a flag fall and the clock has locked-up.The clock only locks up if there is no subsequent time control.

The score sheets merely show subsequent moves being played AFTER an unnoticed flag fall.A flag is only considered to have fallen when a claim is made.
Therefore it is the number of moves made at the time the flag fall claim is made that is relevant and not the number of moves when the flag actually fell.

If at the time the claim is made the prescribed number of moves has been made then the game continues, the number of moves shown by the move counter is irrelevant.

Keong Ang
09-11-2011, 03:36 PM
The clock only locks up if there is no subsequent time control.
A flag is only considered to have fallen when a claim is made.
Therefore it is the number of moves made at the time the flag fall claim is made that is relevant and not the number of moves when the flag actually fell.

If at the time the claim is made the prescribed number of moves has been made then the game continues, the number of moves shown by the move counter is irrelevant.

You could be onto something here.
Article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

When interpreted to mean "official" flag fall occurred when the arbiter or either player finally sees the fallen flag, it changes the application of all subsequent articles.
:hmm:

Very interesting as it is fundamentally different to how I was trained.

Kerry Stead
09-11-2011, 10:43 PM
If a tree falls in the forest & there is no-one around to hear it, does it make a sound??

antichrist
09-11-2011, 11:12 PM
If a tree falls in the forest & there is no-one around to hear it, does it make a sound??

That is what I hate about digital clocks, when the flag falls the bloody alarm goes off and wakes up your opponent when you could have stolen a win on analogue clock - before or after 40 moves

analogy snipped

Kerry Stead
10-11-2011, 12:26 AM
That is what I hate about digital clocks, when the flag falls the bloody alarm goes off and wakes up your opponent when you could have stolen a win on analogue clock - before or after 40 moves

analogy snipped

I'm not sure what digital clocks you used, but when used at chess tournaments, digital clocks have any sound function switched off, so they don't make a noise.

One of the main advantages of digital clocks is that it reduces the 'stealing' of wins that may have occurred previously with analogue clocks, and makes the vast majority of games be decided by the actual chess on the board, rather than whether a person can make 10 decent moves in under 30 seconds, or something of that nature.

My comment was more in relation to the previously discussed issue of when does flagfall occur - when it actually happens, or when it is noticed, by the players and/or the arbiter?

Keong Ang
10-11-2011, 04:49 AM
I'm not sure what digital clocks you used, but when used at chess tournaments, digital clocks have any sound function switched off, so they don't make a noise.

Yes, digital clocks are set to function noiselessly during tournaments. The only noise that can be heard is "clicking" when the levers are physically pressed.


One of the main advantages of digital clocks is that it reduces the 'stealing' of wins that may have occurred previously with analogue clocks, and makes the vast majority of games be decided by the actual chess on the board, rather than whether a person can make 10 decent moves in under 30 seconds, or something of that nature.

That's also why the clocks lock-up feature is so good. With 30second increment players have to keep recording their moves, making it impossible to 'steal' a win if the flag fell during a past move.


My comment was more in relation to the previously discussed issue of when does flagfall occur - when it actually happens, or when it is noticed, by the players and/or the arbiter?

My opinion has always been that flag fall occurs when it actually happens.
Seems like common sense to me.

We can only see the logic around the articles in the Laws of Chess regarding what to do after flag fall if flag fall officially occurs when it actually happens, not when it is noticed. They do not make sense and are quite pointless if decisions are based on "flag fall occurs when it is noticed".

What does your training on this issue during your FA seminar say?

Bill Gletsos
10-11-2011, 08:39 AM
My opinion has always been that flag fall occurs when it actually happens.
Seems like common sense to me.Your "common sense" is in conflict with there laws of chess.


We can only see the logic around the articles in the Laws of Chess regarding what to do after flag fall if flag fall officially occurs when it actually happens, not when it is noticed.Rubbish.

They do not make sense and are quite pointless if decisions are based on "flag fall occurs when it is noticed".Rubbish.

Kerry Stead
10-11-2011, 10:38 AM
My opinion has always been that flag fall occurs when it actually happens.
Seems like common sense to me.

We can only see the logic around the articles in the Laws of Chess regarding what to do after flag fall if flag fall officially occurs when it actually happens, not when it is noticed. They do not make sense and are quite pointless if decisions are based on "flag fall occurs when it is noticed".

What does your training on this issue during your FA seminar say?
Think of it like an irregularity occurring during the game. Say a player now has two bishops on black squares (and has not promoted a pawn), or someone has left their king in check for multiple moves.
In those circumstances, you go back to the position before the irregularity occurred, based on scoresheets (if a recorded game), or what the players remember (if not recorded - rapid, etc), or as best as the arbiter can deduce/decide (if not recorded & players unsure).
Lets imagine a game is being played at 40/90 + 30 with a 30 second increment. Obviously there would need to be 'insufficient supervision' as a competent arbiter assigned to one board or match should notice any issue with flagfall immediately, so the idea of the arbiter noticing the flagfall is not possible in this case (say its a 300 player tournament). Lets imagine that after 39 moves, white's flag falls. Neither player notices this at the time. They both make a few moves reasonably quickly, then notice that after 41 moves, white's clock shows - 0:30 & they call the arbiter over.
What is the arbiter to rule in this case (lets assume no clock defects)? What evidence can the arbiter use in such a situation? There are scoresheets, but they show that the players have played 40 moves (and white has even played 41 moves). There is the clock, which shows that white's flag fell first & that he has been given his additional 30 minutes time, but that is what you would expect to occur if he had made his 40 moves in just under the allotted time & then recieved his extra 30 minutes. Even if the clock was set with the move counter, this would show white having played 41 moves (the clocks only lock up in the final period, so the time continues running after a flag fall), and with no other way to determine things, you must assume that the player has made the time control, so the game should continue.

As you pointed out, article 6.8 gives you the definition of flag fall ... and that is the one that is used in all cases. How could you be trained differently??

Jesper Norgaard
10-11-2011, 11:43 AM
Your "common sense" is in conflict with the laws of chess.

I think that arbiters definitely need common sense, but perhaps not Keong's weird variation. All the same, the worst fault I can possibly see in an arbiter is to insist in neglecting certain official rules, even if told to adhere to them, because he prefers a different set of rules.

antichrist
10-11-2011, 12:15 PM
I think that arbiters definitely need common sense, but perhaps not Keong's weird variation. All the same, the worst fault I can possibly see in an arbiter is to insist in neglecting certain official rules, even if told to adhere to them, because he prefers a different set of rules.

concerning the flagfall actually stopping the game regardless - on fairness grounds he is correct despite my saying otherwise in SB. 6.8 allows unfair results but maybe is necessary due to there being insufficient arbiters etc and totally aware players.

Kerry Stead
10-11-2011, 01:40 PM
concerning the flagfall actually stopping the game regardless - on fairness grounds he is correct despite my saying otherwise in SB. 6.8 allows unfair results but maybe is necessary due to there being insufficient arbiters etc and totally aware players.
You do realise that flagfall is the same whether you are using an analogue or a digital clock ...

antichrist
10-11-2011, 03:02 PM
You do realise that flagfall is the same whether you are using an analogue or a digital clock ...

If noticed when the flag falls and responded to they produce the same results, but if go unnoticed or unresponded to can't the repercussions be different, with digital the clock stops and an interpretation can be made, but with analogue the clock keeps on going to both flag falls with possibly different result, the player further behind may have only got there after his opponent's flag had already fallen.

I admit have hardly used digital at all so only know a little, I stopped comp due to them.

Keong Ang
10-11-2011, 11:52 PM
Think of it like an irregularity occurring during the game. Say a player now has two bishops on black squares (and has not promoted a pawn), or someone has left their king in check for multiple moves.
In those circumstances, you go back to the position before the irregularity occurred, based on scoresheets (if a recorded game), or what the players remember (if not recorded - rapid, etc), or as best as the arbiter can deduce/decide (if not recorded & players unsure).
Lets imagine a game is being played at 40/90 + 30 with a 30 second increment. Obviously there would need to be 'insufficient supervision' as a competent arbiter assigned to one board or match should notice any issue with flagfall immediately, so the idea of the arbiter noticing the flagfall is not possible in this case (say its a 300 player tournament). Lets imagine that after 39 moves, white's flag falls. Neither player notices this at the time. They both make a few moves reasonably quickly, then notice that after 41 moves, white's clock shows - 0:30 & they call the arbiter over.
What is the arbiter to rule in this case (lets assume no clock defects)? What evidence can the arbiter use in such a situation? There are scoresheets, but they show that the players have played 40 moves (and white has even played 41 moves). There is the clock, which shows that white's flag fell first & that he has been given his additional 30 minutes time, but that is what you would expect to occur if he had made his 40 moves in just under the allotted time & then recieved his extra 30 minutes. Even if the clock was set with the move counter, this would show white having played 41 moves (the clocks only lock up in the final period, so the time continues running after a flag fall), and with no other way to determine things, you must assume that the player has made the time control, so the game should continue.

As you pointed out, article 6.8 gives you the definition of flag fall ... and that is the one that is used in all cases. How could you be trained differently??

In 40moves/90minutes + 30minutes with 30seconds increment from move one rate of play, the clock is correctly set when the move counter is set at 40 for time control1, otherwise it cannot regulate the passage of time controls properly.
The DGT XL clocks at FIDE tournaments are set this way.
This is the way I was trained to set the clocks during my FA seminar.

When the number of clock presses reaches 40, 30minutes in time control2 is added. This is the only way to comply with FIDE Tournament Rules article10.c.1.
The display at all times should show the time available to complete a player’s next move.

If you set the move counter on the DGT XL to zero, you will find that the time displayed will not show the additional 30minutes after the 40th move is made until the flag falls. This is clearly in against the regulations.

When the move counter is set to 40, the situation you've described will never occur, the clock would have locked-up. The move counter will not register any extra moves and both players time would be frozen. The evidence is preserved by the clock.
Try it with a DGT XL.

With a DGT 2000, you must set the move counter to 40. If you do not, the clock will not add the extra 30minutes at all and flag fall will lock-up the clock and you will see the flashing minus sign.
Try it out on a DGT2000 by selecting option25 where you can set up to 4 time controls. Setting the move counter to zero tells the clock that this is the last time control with flag fall causing clock lock-up.

Using your described situation, with the procedure of dealing with it as an irregularity, the clock will tell you that white's flag fell at move39 and both players displayed time would have stayed frozen, with white's showing -0.00 (on a DGT2000).

As for interpreting article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.
My training was simple and straight forward.
The flag had already fallen, i.e. occurred in the past.
Obviously the flag fell before it was observed and it is now the arbiter's job to determine when flag fall occurred.

If flag fall is counted as when it is finally noticed. Article6.8 should read "Flag fall occurs when the arbiter..."

The procedure follows the Laws of Chess.

Article6.8 is invoked.
Apply article6.9
Except where one of the Articles: 5.1.a, 5.1.b, 5.2.a, 5.2.b, 5.2.c applies, if a player does not complete the prescribed number of moves in the allotted time, the game is lost by the player. However, the game is drawn, if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves.
Has the player completed 40moves? Scoresheet says 41moves so why did the flag fall?
Apply article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.
Checking chess clock shows that it is not defective. Move counter preserved evidence that flag fall occurred at move39.
Decision based on evidence mandated by article6.10.a is that player lost according to article6.9.

Why do you think my FA training is different? Isn't it the same as yours?:confused:

Bill Gletsos
11-11-2011, 01:57 PM
In 40moves/90minutes + 30minutes with 30seconds increment from move one rate of play, the clock is correctly set when the move counter is set at 40 for time control1, otherwise it cannot regulate the passage of time controls properly.Wrong.
The clock should only add the time when one clock hits zero.

The DGT XL clocks at FIDE tournaments are set this way.Quite strange given the latest model FIDE approved DGT clocks have a standard number for this time control and it does it when one clock hits zero.

This is the way I was trained to set the clocks during my FA seminar.That would seem to be a bit of a concern.


When the number of clock presses reaches 40, 30minutes in time control2 is added. This is the only way to comply with FIDE Tournament Rules article10.c.1.
The display at all times should show the time available to complete a player’s next move.

If you set the move counter on the DGT XL to zero, you will find that the time displayed will not show the additional 30minutes after the 40th move is made until the flag falls. This is clearly in against the regulations.However the latest FIDE approved 2010 DGT clocks have default options that do not follow this regulation.

When the move counter is set to 40, the situation you've described will never occur, the clock would have locked-up. The move counter will not register any extra moves and both players time would be frozen. The evidence is preserved by the clock.
Try it with a DGT XL.This appears to be in contradiction to FIDE regulations as the 40 move time control is not the last time control.

5.1.(i) In case of accumulative or delay timing systems, the clock should not add any additional time if a player passed the last time control.

With a DGT 2000, you must set the move counter to 40. If you do not, the clock will not add the extra 30minutes at all and flag fall will lock-up the clock and you will see the flashing minus sign.
Try it out on a DGT2000 by selecting option25 where you can set up to 4 time controls. Setting the move counter to zero tells the clock that this is the last time control with flag fall causing clock lock-up.This simply because DGT 2000 being the first model released does not allow for the setting of the second time control to start when one clock hits zero if there is a 30 second increment from move one.
This problem was rectified in later model DGT clocks (XL & 2010).

Using your described situation, with the procedure of dealing with it as an irregularity, the clock will tell you that white's flag fell at move39 and both players displayed time would have stayed frozen, with white's showing -0.00 (on a DGT2000).Irrelevant if at the time the flag fall claim is made the scoresheets show 40 moves have been made.

As for interpreting article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.
My training was simple and straight forward.
The flag had already fallen, i.e. occurred in the past.
Obviously the flag fell before it was observed and it is now the arbiter's job to determine when flag fall occurred.Incorrect.
The only time which flag fell first is relevant is if when a flag fall claim is actually made is;
a) if the score sheets of the players show less than 40 moves have been made or
b) in the final time control period if both flags have fallen.
If the score sheets show more than 40 moves have been made at the time a claim for a win on time is made then which flag fell first is irrelevant.

If flag fall is counted as when it is finally noticed. Article6.8 should read "Flag fall occurs when the arbiter..."Only in your opinion.
The meaning of the FIDE wording is quite clear to the rest of us.

The procedure follows the Laws of Chess.

Article6.8 is invoked.
Apply article6.9
Except where one of the Articles: 5.1.a, 5.1.b, 5.2.a, 5.2.b, 5.2.c applies, if a player does not complete the prescribed number of moves in the allotted time, the game is lost by the player. However, the game is drawn, if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves.
Has the player completed 40moves? Scoresheet says 41moves so why did the flag fall?
Apply article6.10.a.
Every indication given by the clocks is considered to be conclusive in the absence of any evident defect. A chess clock with an evident defect shall be replaced. The arbiter shall replace the clock and use his best judgment when determining the times to be shown on the replacement chess clocks.
Checking chess clock shows that it is not defective. Move counter preserved evidence that flag fall occurred at move39.
Decision based on evidence mandated by article6.10.a is that player lost according to article6.9.What total rubbish.
If at the time a claim for a win on time is made, the scoresheets say 41 moves have been made and the players agree and replaying through the scoresheets shows that indeed 41 moves were made then it is irrelevant what the clock says.
The only correct decision by the arbiter is to allow the game to continue.
For an arbiter to declare it lost based on the clock when all other evidence shows 41 moves have been made is clearly an incorrect decision.

Why do you think my FA training is different? Isn't it the same as yours?:confused:Because you views are clearly a concern.

Kerry Stead
11-11-2011, 10:37 PM
Why do you think my FA training is different? Isn't it the same as yours?:confused:
I'll leave the other questions, as Bill seems to have gone through them in detail.
Yes, our training on this matter was obviously different. At the seminar I attended, it was recommended to not use the move counter function of the clocks for a variety of reasons: avoids confusion, difficulties resolving problems if there is one, consitency in all situations, etc.
If you consider it another way, if a probelm occurs in a game (for whatever reason), it is far easier for the players to reconstruct the moves of the game (if determining if a time control has been reached for example), than it is to reconstruct the amount of time taken for each decision. Players can remember a sequence of moves (whether written down or not) much easier than they can remember 9if they know at all) how long a decision took, for example.
I think the non-use of the move counter function avoids problems in large swiss events (which are by far the most common in Australia) where there are a range of skill and experience levels, but this may not be such an issue if you are talking about a top-level international event with experienced players.
Having said that, there was recently an issue with a clock & move counters - the Kramnik-Radjabov playoff match in the Candidates tournament ... so these issues with clocks do happen & I find it is easier to set up circumstances so that these potential issues can be minimised before they even become a problem.

antichrist
12-11-2011, 05:02 PM
If noticed when the flag falls and responded to they produce the same results, but if go unnoticed or unresponded to can't the repercussions be different, with digital the clock stops and an interpretation can be made, but with analogue the clock keeps on going to both flag falls with possibly different result, the player further behind may have only got there after his opponent's flag had already fallen.

I admit have hardly used digital at all so only know a little, I stopped comp due to them.
----------------------------------------------------------------

Am I totally misconstructing everything about this issue - I would not mind knowing? thanks

Keong Ang
12-11-2011, 05:55 PM
Wrong.
The clock should only add the time when one clock hits zero.
Quite strange given the latest model FIDE approved DGT clocks have a standard number for this time control and it does it when one clock hits zero.

Then what would be the reason for FIDE to test clocks to correctly move to the next time control on the 40th clock press?
The proper test should be that the next time control starts when a clock hits zero.

Is there a good reason for FIDE approved clocks to have a lock-up feature for time controls that are not the last when the move counter is set?

Anyway, I am convinced that you are correct that clocks should have the move counter set to zero.

To my mind, the only FIDE Regulation that would makes setting the move counter to zero wrong is FIDE Tournament Rules article10.c.1.
"The display at all times should show the time available to complete a player’s next move."

This is practically impossible to achieve when the move counter is set to zero. However, the world "should" in article10.c.1. (highlighted by me), means it is recommended but not compulsory.

Also, you have also correctly pointed out that the clock's "move counter" is merely a clock press register. When we have to correct clock move counter errors due to too many clock presses, when the scoresheets and position on the board proves that there were not that many moves, it logically follows that errors due to too few clock presses also need to be corrected. This would strongly support a "scoresheet trumps the clock" argument.

When the clock's move counter is not set to zero, the clock would lock-up if the set number of moves are not made when flag falls. This would result in the clock destroying evidence. It would make it impossible for the arbiter to then correct the error accurately.

Thank you Bill for pointing out the latest DGT2010 (with blue buttons). I have since completed checking the manual. I believe you have already read this note in page6.
If a player does not finish the programmed (non-zero) number of moves for a particular period, when the display indicates 0:00, the blinking flag is shown and the DGT 2010 stops time counting for both players, indicating that the game has ended, with the player in question having lost the game on time. This means that the players have to stop the clock correctly after each move. Therefore the FIDE does not encourage this method of transition to the next period. The transition on zero time of one player is the preferred method.
This indicates that there is an official FIDE recommendation/preference for setting the move counter to zero. Frustratingly, it does not say where FIDE said so.


That would seem to be a bit of a concern.
Yes, it is a concern! I'll have to follow this up. A player could have lost a norm or worse.

This appears to be in contradiction to FIDE regulations as the 40 move time control is not the last time control.

5.1.(i) In case of accumulative or delay timing systems, the clock should not add any additional time if a player passed the last time control.
There is no contradiction to FIDE regulations here. The regulation you quoted only says that clock lock-up should happen at the last time control. It does not say that clock lock-up must not happen in a time control that is not the last.

This is probably why clocks that lock-up when the move counter is set can still get FIDE approval.
An irrelevant issue when the clock's move counter is set to zero.

As for interpreting Laws of Chess article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

My conclusion is that I was wrong with my original opinion that "flag fall officially occurs when it actually happens". The arbiter would have a much easier task if "flag fall officially occurs when it is noticed" since it would make all sorts of forensic investigation unnecessary. It also shifts primary responsibility of noticing flag fall onto the players.

The "...when either player has made a valid claim to that effect." part of article6.8 implies that a player notices the passing of a time control and thinks insufficient moves have been made. Necessitating the pausing of the clock and making a flag fall claim that is checked for validity. The arbiter checking the flag fall claim would then use the scoresheets to confirm the validity of the claim. The application of other articles in the Laws of Chess would still follow smoothly.

Keong Ang
12-11-2011, 06:29 PM
I'll leave the other questions, as Bill seems to have gone through them in detail.
Yes, our training on this matter was obviously different. At the seminar I attended, it was recommended to not use the move counter function of the clocks for a variety of reasons: avoids confusion, difficulties resolving problems if there is one, consitency in all situations, etc.
I have experimented with different clock move counter settings and time controls, and looked at how the regulations would be applied to various scenarios. In all cases, I'd prefer to be the arbiter when the clock move counter is set to zero.
The worst that could happen when the move counter is set to zero is that a player "escapes being flagged" by playing sufficient moves to enter the next time control. The opponent is the one who shoulders the responsibility to notice and claim flag fall in a timely manner. If the flag fall goes unnoticed, it really is no problem as the game would continue.

From the arbiter's point of view, there will be less likelihood of disputes if the move counter was not set. A much easier job.


If you consider it another way, if a probelm occurs in a game (for whatever reason), it is far easier for the players to reconstruct the moves of the game (if determining if a time control has been reached for example), than it is to reconstruct the amount of time taken for each decision. Players can remember a sequence of moves (whether written down or not) much easier than they can remember 9if they know at all) how long a decision took, for example.
I think the non-use of the move counter function avoids problems in large swiss events (which are by far the most common in Australia) where there are a range of skill and experience levels, but this may not be such an issue if you are talking about a top-level international event with experienced players.
Having said that, there was recently an issue with a clock & move counters - the Kramnik-Radjabov playoff match in the Candidates tournament ... so these issues with clocks do happen & I find it is easier to set up circumstances so that these potential issues can be minimised before they even become a problem.
I have now come to think that the move counter should only be set for games that are being broadcast. Even then, it is really done for the spectators benefit.

As an arbiter of tournaments where the clock move counter is set, there is the constant job of having to monitor the clocks move counters and making sure that the scoresheets match up. This makes constant patrolling of the tournament playing area necessary. You can imagine that this tends to distract players.

A good arbiter should be always present, but invisible. Which means the arbiter's actions during a tournament should be one that does not distract the players while at the same time giving confidence that the tournament is being properly supervised.

I'm trying to find out why my FA seminar notes seem to be so at odds with your FA seminar's. I'm wondering if being jet-lagged and sleep-deprived caused me to mishear the lessons. I'd better to clear this up with FIDE.
:hmm:

Keong Ang
12-11-2011, 06:31 PM
----------------------------------------------------------------

Am I totally misconstructing everything about this issue - I would not mind knowing? thanks

When a titled arbiter can be wrong about such issues, you can be excused!!
:lol: :lol:

Jesper Norgaard
13-11-2011, 04:57 AM
What is grossly unfair how quicker thinkers like myself lose a natural advantage to slow thinkers who should be out the door quick smart so I can either have a rest or watch the top boards.
What is grossly unfair is that "quicker thinkers" like yourself are entitled to win games without showing results on the chessboard.



NB I forgot an analogy: digital clocks are like Dyson vacuum cleaners, completely unnecessary, expensive, need a degree to run and strong shoulders and patience to achieve no better result
Except that increments *do* achieve better results. Actually I feel that it is even more unfair to the chess games being cut a logical solution (than unfair for the players to lose on time), and thereby devaluating the pool of classical games being added to, by terminating some games by lottery.

antichrist
13-11-2011, 05:45 AM
What is grossly unfair is that "quicker thinkers" like yourself are entitled to win games without showing results on the chessboard.


Except that increments *do* achieve better results. Actually I feel that it is even more unfair to the chess games being cut a logical solution (than unfair for the players to lose on time), and thereby devaluating the pool of classical games being added to, by terminating some games by lottery.

Well I read ages ago that in top chess the increment time made no different to results, the same people still won. That means a few things: the quicker players can adjust to increment and still win, therefore increment is just a waste of precious time that stops me watching top board games or resting, and prevents an extra round of 60 mins guillotine in a day.

Since clocks were invented that favoured the quicker player just because
they are the better player (even if that means only have studied their openings better).

The pool of classical games I guess is severely mostly added onto by the top players and as they know their openings perfectly I dont think it would make much difference to their standard.

With post mortems on games and Fritz programs there is plenty of opportunity to read what if had happened in annotations. That has been happening for a hundred years.

When I played grade I cant ever remember being in time trouble but also never won any game because of time advantage - not that I realised it was because of that anyway.

In tourneys against much higher-rated players I have lost in games that time pressure probably contributed to a quicker loss but that was only a handful of games - for me to deserve to win consistently at that level I would have to study a lot more and then presumably I would not frequently be in time pressure.

As well I just hate being at the board longer with people I consider dopes coz they cannot work out the move in reasonably time - that I consider due to insufficient study of the game or insufficient natural talent.

After finding out what is the point of this thread I agree that time should be added onto after 40 moves. As it seems obvious that clickers must mistakes in clicking, too many or too few or not registering, the 40 moves must be established by the scoresheet. To prevent cheating this should be observed by the DOP before resetting the clock - so the DOP still gets his exercise. So there is one "advantage" of digital clocks nullifed.

Jesper Norgaard
13-11-2011, 06:00 AM
Is there a good reason for FIDE approved clocks to have a lock-up feature for time controls that are not the last when the move counter is set?
Yes that FIDE has continuously neglected to make up it's mind in this area. As a result, the clock producers quite naturally implemented both ways to handle time periods since there was no decision. Also, in games like scrabble or Go, they may make other decisions, so having a versatile clock seems a good idea.



Anyway, I am convinced that you are correct that clocks should have the move counter set to zero.

Bravo! As you can see yourself I set out this thread with the conviction that setting the first period to 40 moves could be even better or equal to setting it to zero. May I congratulate you on achieving a more mature view on this. The clock can never be the arbiter, it can only be a passive assistant, although an essential one.



To my mind, the only FIDE Regulation that would makes setting the move counter to zero wrong is FIDE Tournament Rules article10.c.1.
"The display at all times should show the time available to complete a player’s next move."

However when both players handle the clock presses accurately, this rule is complied with, and the clock could never know the moves on the board, so the common sense of 10.c.1 is really not violated. I think this rule is violated much worse when you use 40 moves for the first period, and too many clock presses are entered, and the clock insists that a player has 31 minutes to complete his next move, while in reality he has only 1 minute.



Also, you have also correctly pointed out that the clock's "move counter" is merely a clock press register. When we have to correct clock move counter errors due to too many clock presses, when the scoresheets and position on the board proves that there were not that many moves, it logically follows that errors due to too few clock presses also need to be corrected. This would strongly support a "scoresheet trumps the clock" argument.

Exactly! By the way, it has always been like that, even before digital clocks.



When the clock's move counter is not set to zero, the clock would lock-up if the set number of moves are not made when flag falls. This would result in the clock destroying evidence. It would make it impossible for the arbiter to then correct the error accurately.

Exactly!



Thank you Bill for pointing out the latest DGT2010 (with blue buttons). I have since completed checking the manual. I believe you have already read this note in page6.
If a player does not finish the programmed (non-zero) number of moves for a particular period, when the display indicates 0:00, the blinking flag is shown and the DGT 2010 stops time counting for both players, indicating that the game has ended, with the player in question having lost the game on time. This means that the players have to stop the clock correctly after each move. Therefore the FIDE does not encourage this method of transition to the next period. The transition on zero time of one player is the preferred method.
This indicates that there is an official FIDE recommendation/preference for setting the move counter to zero. Frustratingly, it does not say where FIDE said so.

Since DGT has always been somehow the semi-official clock manufacturer for FIDE, this recommendation actually makes some weight for me. The recommendation ought to be on the FIDE web site.



Yes, it is a concern! I'll have to follow this up. A player could have lost a norm or worse.

Exactly! Very courageous of you to take this up in such a thoughtful manner.




As for interpreting Laws of Chess article6.8
A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.

My conclusion is that I was wrong with my original opinion that "flag fall officially occurs when it actually happens". The arbiter would have a much easier task if "flag fall officially occurs when it is noticed" since it would make all sorts of forensic investigation unnecessary. It also shifts primary responsibility of noticing flag fall onto the players.

Exactly! It has always been the primary responsibility of the players, to both manage their own time to avoid flag fall, and monitor the opponents time and flag fall. It is also the responsibility for the arbiter to do this and stop the game for investigation when he notices that a flag falls happen, but it can only be primary responsibility when there is one arbiter for each game.



The "...when either player has made a valid claim to that effect." part of article6.8 implies that a player notices the passing of a time control and thinks insufficient moves have been made. Necessitating the pausing of the clock and making a flag fall claim that is checked for validity. The arbiter checking the flag fall claim would then use the scoresheets to confirm the validity of the claim. The application of other articles in the Laws of Chess would still follow smoothly.
Exactly!

I hope it feels better to get 10 "Exactly" from Jesper than 10 "Rubbish" from Bill :clap:
That does not guarantee that I am correct in all of the above :uhoh:

As Captain Underpants would say, carry on!

Garvinator
14-11-2011, 01:33 AM
So are we all now in agreement that the move counter is not to be used in multi period time controls?

Jesper Norgaard
14-11-2011, 04:45 AM
So are we all now in agreement that the move counter is not to be used in multi period time controls?
I think that question needs a poll to be meaningful, but - yes

I would like to rephrase it though to avoid confusion. In multi period time controls, the non-last periods should be set to zero moves. The move counter is still running, and should be monitored and corrected by the arbiter when appropriate, and can be claimed by players to be wrong.

Nothing new in this, but it just seems confusing to me to claim that the move counter is not used, when it clearly is active and used. It is actually the number of moves per period that is not used.

Garvinator
16-11-2011, 08:24 PM
I think that question needs a poll to be meaningful, but - yesWhy? This is a rules debate/discussion, not a democratic election.

How would you stop people who have no idea from voting?

Denis_Jessop
16-11-2011, 08:35 PM
Why? This is a rules debate/discussion, not a democratic election.

How would you stop people who have no idea from voting?

But that's what happens in our federal elections etc :) and voting is compulsory upon pain of a fine if you don't :hmm:

DJ

Garvinator
16-11-2011, 08:48 PM
But that's what happens in our federal elections etc :) and voting is compulsory upon pain of a fine if you don't :hmm:
Wrong, voting is not compulsory. How many times does that need to be repeated in regards to Australian voting. You should know better Denis than to claim that voting is compulsory.

Why did you reply to my comment and not to Jesper's?

Keong Ang
17-11-2011, 09:12 AM
I think that question needs a poll to be meaningful, but - yes

I would like to rephrase it though to avoid confusion. In multi period time controls, the non-last periods should be set to zero moves. The move counter is still running, and should be monitored and corrected by the arbiter when appropriate, and can be claimed by players to be wrong.

Nothing new in this, but it just seems confusing to me to claim that the move counter is not used, when it clearly is active and used. It is actually the number of moves per period that is not used.

Once the time controls are passed through flag falls instead of number of clock presses, the clock's move counter is relatively redundant. The players scoresheets would tell the number of moves made.

You're correct that the move counter is still active but it is practically of limited use to regulate the passage of time controls. The move counter would be used to check that the 30second increments have been added correctly.

How do we set the DGT2000 correctly? They will lock up at the end of the 1st time control if the move counter was not set. Is there a way around this?
If we are supposed to set the move counter to zero, the DGT2000 seems to be incapable of handling multiple time controls based on number of moves that also have increments from move1.

Keong Ang
17-11-2011, 09:16 AM
So are we all now in agreement that the move counter is not to be used in multi period time controls?

I think the only holdout in this thread regarding setting the move counter has capitulated. ;) :lol:

It would take some higher up instruction or interpretation from FIDE to change my mind.
I'm still making follow up inquiries with FIDE.

Jesper Norgaard
17-11-2011, 11:21 AM
Why? This is a rules debate/discussion, not a democratic election.

How would you stop people who have no idea from voting?

Actually my comment had an extremely modest scope, I just wanted to hint that if all arbiters on the forum answered we might get a flood of posts. Still my intention was not to suggest a poll, much less obliging everyone to answer, or exclude those who are not able to. It was actually meant to be extremely inoffensive, and non-suggestive.

I didn't even want it to be a criticism, in fact - I would like to retract it although I know I can't :doh:

Denis_Jessop
17-11-2011, 12:37 PM
Wrong, voting is not compulsory. How many times does that need to be repeated in regards to Australian voting. You should know better Denis than to claim that voting is compulsory.

Why did you reply to my comment and not to Jesper's?

Garvin, you have me puzzled. Section 245 of the Commonwealth Electoral Act 1918 provides in some detail that voting is compulsory. Perhaps you are talking about something else.

My response was to your comment


How would you stop people who have no idea from voting?

the implication being that many people voting in federal elections have "no idea". This also applies to State elections.

DJ

Kevin Bonham
17-11-2011, 08:55 PM
Garvin, you have me puzzled. Section 245 of the Commonwealth Electoral Act 1918 provides in some detail that voting is compulsory. Perhaps you are talking about something else.

He's raising the standard quibble that it is actually only attending the polling booth, getting marked off and receiving a ballot paper that is compulsory. The quibble is actually false as concerns the letter of the law, but true in effect since there is no way to force a person to exercise a valid vote without invading the privacy of the ballot.

Pepechuy
18-11-2011, 09:17 AM
I do not know if the following is easy or difficult to implement, but would be helpful.
There is agreement that the clock does not count moves, only clock presses.
I wonder if it is possible for the clock to keep record of the exact moment of each press, and show it when the arbiter requires it.

Jesper Norgaard
18-11-2011, 10:31 AM
I do not know if the following is easy or difficult to implement, but would be helpful.
There is agreement that the clock does not count moves, only clock presses.
I wonder if it is possible for the clock to keep record of the exact moment of each press, and show it when the arbiter requires it.
Sure it might be technically possible, but would it help anything? Remember rule 6.8 still indicates that flag falls in the past are null and void, it is the situation at the moment an arbiter observes or a player claims a flag fall, that counts. Please explain how it would help.

Keong Ang
18-11-2011, 02:08 PM
I do not know if the following is easy or difficult to implement, but would be helpful.
There is agreement that the clock does not count moves, only clock presses.
I wonder if it is possible for the clock to keep record of the exact moment of each press, and show it when the arbiter requires it.

In a system where DGT XL clock is linked to DGT eboard, clock times can be recorded against each move.

Keong Ang
18-11-2011, 02:32 PM
Sure it might be technically possible, but would it help anything? Remember rule 6.8 still indicates that flag falls in the past are null and void, it is the situation at the moment an arbiter observes or a player claims a flag fall, that counts. Please explain how it would help.

There's the complication of a player claiming flag fall when extra 30 minutes added at eg. move40.

When the move counter is not used to regulate passage of time controls, the next time control starts upon flag fall. The new blue button DGT2010 even displays the flag for up to 5 minutes. The clock is trying to get player/arbiter to check and take action.

What if there were too many clock presses? The move counter (i.e. clock press counter) would be showing more moves than the scoresheets.

The arbiter would have to remove the extra moves AND subtract the extra 30second increments. This could result in loss on time because the player wouldn't have made the 40move time control if these extra 30second increments were not added in error.

I think such a situation was discussed earlier in this thread (most probably by you!). It is rare and in most cases, a player would have got away with it if nobody noticed the clock error in time. Practically such claims need to be made at move40 just as the flag fell.

Denis_Jessop
18-11-2011, 03:45 PM
He's raising the standard quibble that it is actually only attending the polling booth, getting marked off and receiving a ballot paper that is compulsory. The quibble is actually false as concerns the letter of the law, but true in effect since there is no way to force a person to exercise a valid vote without invading the privacy of the ballot.

It is indeed a quibble when Subs. 245(15) provides that


(15) An elector is guilty of an offence if the elector fails to vote at an election.

to which strict liability applies.

Whether the offence could be successfully prosecuted is another matter.

Also subs.245(1) provides


(1) It shall be the duty of every elector to vote at each election.


DJ

Pepechuy
19-11-2011, 08:53 AM
Sure it might be technically possible, but would it help anything? Remember rule 6.8 still indicates that flag falls in the past are null and void, it is the situation at the moment an arbiter observes or a player claims a flag fall, that counts. Please explain how it would help.

Yes, it would help in some cases.
Imagine, 39 moves have been made. All the moves have been correctly recorded by both players and the clock has been properly pressed by each one after every move: the move counter also shows 39 moves.
White oversteps the time limit, and half second later makes the 40th move. The clock automatically adds the time for the second control (whether it is as time expired or at completing move 40, well, that is what you are discussing in this thread).
Black did not really had the chance to claim a win on time before White moved, even if it was noticed as happened.
With the old analog clocks, Black should only avoid touching the clock and claim a win, it was clear that White's time expired before completing the 40th move.
I guess there are other cases where it might be helpful.

Kerry Stead
19-11-2011, 10:35 AM
Yes, it would help in some cases.
Imagine, 39 moves have been made. All the moves have been correctly recorded by both players and the clock has been properly pressed by each one after every move: the move counter also shows 39 moves.
White oversteps the time limit, and half second later makes the 40th move. The clock automatically adds the time for the second control (whether it is as time expired or at completing move 40, well, that is what you are discussing in this thread).
Black did not really had the chance to claim a win on time before White moved, even if it was noticed as happened.
With the old analog clocks, Black should only avoid touching the clock and claim a win, it was clear that White's time expired before completing the 40th move.
I guess there are other cases where it might be helpful.

The situation is exactly the same as the analogue clock & the player loses on time in the situation you described.

Lets assume the time control is 40/90+30 with a 30 second increment from move 1 to make the example easy to explain.

In the situation you describe, the players have made their first 39 moves & as the player with the white pieces plays his 40th move, he oversteps the time limit, then presses the clock. The clock then adds the second time control (move counter set to 0 for first period). In this case the display on the clock for white would be -0:30 showing that his flag has fallen for the first time period (and fallen first amongst the players) & that he has 30 minutes remaining. Black simply calls the arbiter over, they check the score sheets & white loses on time!

Lets imagine the same situation, but in this case, white makes the 40th move with 1 second to spare on the clock (and therefore does not overstep the time control). The clock display for white would then be 0.31, with no flag shown (he has not overstepped the time control), and his 30 second increment has been added, so the clock shows him as having 31 seconds remaining. If he then thinks for more than 31 seconds on this 41st move, his flag would 'fall', but he would have already completed 40 moves by this stage, and therefore simply continue the game in the second time period.

Jesper Norgaard
19-11-2011, 12:10 PM
What if there were too many clock presses? The move counter (i.e. clock press counter) would be showing more moves than the scoresheets.

The arbiter would have to remove the extra moves AND subtract the extra 30second increments. This could result in loss on time because the player wouldn't have made the 40move time control if these extra 30second increments were not added in error.

I think such a situation was discussed earlier in this thread (most probably by you!). It is rare and in most cases, a player would have got away with it if nobody noticed the clock error in time. Practically such claims need to be made at move40 just as the flag fell.

The cases above discussed by Pepechuy and Kerry Stead in fact results in simple flag falls or simple passing to next period without flag fall. This would not be any different with the new clock. Instead I think Keong is on to something here. Let me try and construct such an example where there would be a difference from using a normal DGT clock and one where each exact clock press is remembered.

Suppose player A has completed 36 moves, and notices he is short on time, in fact he only has 31 seconds left. He plays out 3 moves quickly with an average of 5 seconds per move, and now has earned 1:30 - 0:15 = 1:15, so now has 1:46 left on the clock, to make his 40.th. move.

Now his opponent discovers that there is a discrepancy of the move counter showing 41 when in fact it should say 39. It is clear that both players have made 2 clock presses too much earlier in the game, so he calls the arbiter over.

Now adjusting straight away the arbiter can withdraw 2 from the move counter to set it to 39, and withdraw 1 minute from both clocks, so player B's clock now shows only 0:46. Everybody happy. But what about the situation at move 36? In fact player A who had 31 seconds left on the clock, actually had lost on time because both players had 1 minute too much on the clock at that moment too. It would only be apparent if you know his last three moves only took 15 seconds.

With a clock that knows all exact times that the clock was pressed (in fact better, just the distance between each clock press, because that is all that matters) it would be possible to calculate that player A had lost on time.

Two comments. First of all I don't think that this is relevant to make an arbiter calculate that player A had in fact lost on time. It would be better to think of a new generation of clocks where if you adjust the move counter down, the clock will automatically check for this kind of flag fall, and as a result of decreasing move counter by two, would have returned the clock to the state at move 36 where white's flag falls, and blinking.

Second of all, I have my doubts to whether these types of flag falls should be actually carried out. The reason is that it goes against the passus that the clock should always show the available time for the next move for each player, and you could say this case would be a contradiction. Probably player A calculated much earlier he had enough time, relying on the clock display.

The other reason is how easy it is to swindle with this. To few clock presses is actually not that simple to obtain, it requires both players to press too few times. Too many is abundantly easy to obtain, for instance a spectator drops a chess book on the clock, and inadvertently press the clock, but then clicks back without realizing the move counter went ahead. Likewise if player B with malice quickly clicks the clock a couple of extra times while player A is on the toilet. Likewise a competitor C in the tournament with malice, who sees both player A and B away, and benefits if player A loses - he comes over and perform a few extra clock presses knowing that player A normally is the one to go into time trouble. When player C sees player A has little time left after move 39, he discreetly informs player B that the move counter is apparently wrong.

So, perhaps these kind of flag falls should not be punished. The punishment could be just that player A and B get less time for period 2. Otherwise it is too easy to swindle.

Keong Ang
19-11-2011, 03:39 PM
I've got a reply from FIDE regarding some issues. Quite surprised that IA Panagiotis Nikolopoulos, Chairman, FIDE Arbiters' Commission actually managed to reply to my question when he's currently the Chief Arbiter of the Woman's World Chess Championship (that incidentally uses the same 40/90+30 with a 30 second increment from move 1 that this thread is about).

I'll quote my questions and the received answers for everyone's benefit.



On Fri, Nov 11, 2011, Keong Ang wrote:


Question 1.

With a rate of play of 90 minutes for 40 moves + 30 minutes with 30
seconds cumulative increment for each move starting from the first move,
what is the correct way to set the move counter on DGT XL clocks?

Should the move counter be set at 40 moves for time control 1 and 0
moves for time control 2?
Or should the move counter be set at 0 moves for all time controls.

If the move counter was set at 40 moves for time control 1, the DGT XL
would lock-up if flag fall occurs.
If the move counter was set at 0 moves for all time controls, the DGT XL
would enter time control 2 when the allotted time in time control 1 is
used up.

FIDE Tournament Rules article10.c.
"If electronic chess clocks are used, they must function in full
accordance with the FIDE Laws."
FIDE Tournament Rules article10.c.1.
"The display at all times should show the time available to complete a
player’s next move."
FIDE Tournament Rules article10.c.4.
"In case of passing a time control, a sign on the display must give
clear signal which player passed the time limit first."
FIDE Tournament Rules article10.c.7.
"Special attention should be given to the correct announcement of
passing time controls."
FIDE Tournament Rules article10.c.8.
"In case of accumulative or delay timing systems, the clock should not
add any additional time if a player passed the last time control."

Based on the above clauses from FIDE Tournament Rules, especially to
comply with article10.c.1, the DGT XL clock would need to have the move
counter set to 40 moves for time control 1 and 0 moves for remaining
time control.

What is the correct way to set up the DGT XL move counter to regulate
the passage of time controls?


Question 2.

What is the correct interpretation of Laws of Chess article6.8
"A flag is considered to have fallen when the arbiter observes the fact
or when either player has made a valid claim to that effect."

When does flag fall occur - when it actually happens, or when it is
noticed by the players and/or the arbiter?


Question 3.

Rate of play of 90 minutes for 40 moves + 30 minutes with 30 seconds
cumulative increment for each move starting from the first move.
DGT XL had 40 moves set on move counter for time control 1.

Flag fell at move 38, but neither the arbiter nor players notice until
move 41 is played, what should be the arbiter's decision?
The DGT XL has locked-up at move 38 but the players scoresheet shows
move 41.
Correct interpretation and application of Laws of Chess articles 6.8,
6.9, 6.10.a, 6.13 and 6.14 would be required.

If DGT XL had move counter set to 0 moves for all time controls, it will
not lock-up at flag fall at move 38. Additional 30 minutes would be
added and the move counter would keep registering moves. By move 41,
would there be any flag fall to notice? Would this be in compliance with
FIDE regulations?


Thank you.

Regards,
FA Keong Ang.
FIDE ID: 4303660 (NZL)


On Fri, 2011-11-18 at 14:37 +0200, FIDE Arbiters wrote:

Dear Mr. Ang,

Actually there are 2 ways to handle the rate of play you mention.
(a). the player goes to the second period ( 30 minutes are added to his time ) after he completes 40 moves.
(b). the player goes to the second period ( 30 minutes are added to his time ) after he completes 2 hours of game.

Question 1

The correct way to set DGT XL on the rate of play you mention and for (a) case, is by using the 00 mode (manually). You have to put 40 moves for the first period 1 and 00 moves for the second period 2. If there is a flag fall, the clock locks up and gives an indication "-". The game shall stop.
For the (b) case you put in the same mode 00 for the moves in all periods. If the player will not play 40 moves in 2 hours, then there is a flag indication but the clock does not lock up. The Arbiter has to be present and to check if the player has completed 40 moves. If not, the Arbiter has to stop the game.

In my opinion the correct way to set the clock is to put 40 moves in the move counter for the period 1 ( it means you will have the above case (a)) and 00 moves for the period 2.
This is what it is used in all top tournaments and world FIDE and Continental events.


Question 2

According to 6.8 the flag fall occures when either player or the Arbiter observes (calls) it.
So the right interpretation of 6.8 for games with standard time control ( not rapid, not blitz) is that the Arbiter has to interfere immediately with the game, when we have a flag fall ( i.e. when a player didn't complete the required number of moves in the allotted time and the game shall be ended). Because in fact the flag fall has to be observed (called) at the same time that it happened. And it is mainly an Arbiter's duty.
For rapid and blitz the Arbiter does not interfere, so the players have to observe (indicate) the flag fall.

Question 3

Since the clock has locked up, it means that the game have finished by flag fall and it was the Arbiters' mistake that he didn't indicate this.
If the counter was set to 00 the clock would continue, but a flag indication is shown. The Arbiter should be present to check if both players completed the 40 moves before the next period will be added.
In any case the game shall be ended.
DGT XL has been manufactured in compliance with the Laws of Chess. It is the Arbiters' duty to apply the regulations and the Laws in any case.

With best regards
Panagiotis Nikolopoulos
Chairman
FIDE Arbiters' Commission

It looks like setting the clock's move counter to 40moves or 0moves are both allowed. However, setting the move counter to 40moves is the "more correct" way and it is used for all top FIDE and Continental events.

Regarding article6.8, IA Nikolopoulos seems to be saying that the arbiter has to act as if he intervened when flag fall occurred. This view seems to be followed up in his answer to Question3.

I think the FIDE answer seems to confirm that the move counter has to be set at 40moves and if the clock locks up due to insufficient clock presses, the game ends. :eek:
Seems like a clock trumps the scoresheet view.

This is probably the worst answer for me to receive from FIDE after being convinced that setting the move counter is wrong. :doh: :wall: :wall:
What do you all think??? :confused:
At this point it looks like I'm going to be forced to change my position on setting the move counter! Just as I thought we've all come to some consensus on this issue. :rolleyes:

Kerry Stead
19-11-2011, 11:49 PM
One thing to keep in mind when reading the response from IA Nikolopoulos is that he seems to be referring to 'all top tournaments and world FIDE and Continental events' when making his comments. In such circumstances, there are either a limited number of boards, or multiple arbiters to supervise the playing area & the large number of boards in action.

If you compare his suggested responses to the practicalities of running a large FIDE-rated Swiss event (say 100+ players, with only 1 or 2 arbiters), then it is physically impossible for an arbiter to notice all/any flag falls when they are required to be in charge of so many boards (in his response to question 2, he notes that it is the arbiter's duty to observe all flag falls & immediately step in to stop the game & award the result accordingly). It is these type of situations where I feel that NOT using the move counter is far more beneficial, as it is much easier to resolve a dispute when a clock is still functioning than to try to resolve an issue when a clock has locked up.

Of course in an ideal world, you would be able to have 1 arbiter for every 5 boards (or thereabouts), but this is not always possible, particularly for events that are not of the 'top tournaments and world FIDE and Continental events' level.

Keong Ang
20-11-2011, 07:03 AM
One thing to keep in mind when reading the response from IA Nikolopoulos is that he seems to be referring to 'all top tournaments and world FIDE and Continental events' when making his comments. In such circumstances, there are either a limited number of boards, or multiple arbiters to supervise the playing area & the large number of boards in action.

If you compare his suggested responses to the practicalities of running a large FIDE-rated Swiss event (say 100+ players, with only 1 or 2 arbiters), then it is physically impossible for an arbiter to notice all/any flag falls when they are required to be in charge of so many boards (in his response to question 2, he notes that it is the arbiter's duty to observe all flag falls & immediately step in to stop the game & award the result accordingly). It is these type of situations where I feel that NOT using the move counter is far more beneficial, as it is much easier to resolve a dispute when a clock is still functioning than to try to resolve an issue when a clock has locked up.
Another part of IA Nikolopoulos reply that I noted was that it was his opinion that the 40moves be set. He even started his answer by stating that there were two ways to set the clock. So we are quite free in this regard.


Of course in an ideal world, you would be able to have 1 arbiter for every 5 boards (or thereabouts), but this is not always possible, particularly for events that are not of the 'top tournaments and world FIDE and Continental events' level.

There certainly weren't 1 arbiter for every 5 boards at WYCC2010. There were 3 tournament halls in 2 buildings with around 1300 players. From what I was told, it was the largest annual FIDE event. I doubt anything we organise in Australia or NZ is larger.

I hesitate to suggest that it is impractical to use the move counter in large events. During FIDE WYCC2010 all the clocks had their move counters set. A number of players lost when their clocks locked up. IA Nikolopoulos was the CA there (and the lecturer of my FA seminar) so I'm not surprised at how the clocks were set.

I presume IA Nikolopoulos has the opinion that flag fall occurs when it actually happens. Otherwise it would be impossible to say that "In any case the game shall be ended." in his answer to Question3.
If flag fall occurs when it is observed, the answer should be that the game continues.

When flag fall occurs when it actually happens, it would be better to set the clock to lock up upon flag fall, especially when you have inadequate supervision due to insufficient arbiters. If the clock does not lock up, the game could continue until the next time control looks valid. IA Nikolopoulos says that the game must end in both cases, and I can see that he views it to be easier for the arbiter to end the game if the clock locked up as it would be difficult for a player to argue against a clock that locked up at move38, i.e. evidence is that the player did not make the time control.

My guess is that IA Nikolopoulos probably thinks that the arbiter should extract a result as soon as possible. When you have more than a thousand players and coaches waiting for the next round's pairings, not much effort would be put into finding reasons to continue a game. When the idea of a resolved dispute is game ended with a result, then a locked up clock is great. The arbiter simply tells the player who got flagged that the game is over and he lost (or draw if opponent cannot checkmate). Very few players are willing to pay 100 euros to appeal the arbiters decision when the appeals committee is unlikely to want to let the game continue as it would interrupt the progress of the tournament.

Jesper Norgaard
20-11-2011, 09:28 PM
I think that the Chairman of the Rules commission (IA Geurt Gijssen) and the Chairman of the Arbiters commission (IA Panagiotis Nikolopoulos) should take a good wrestle to settle the dispute. Obviously they don't agree. I must admit I find the interpretation of rule 6.8 of Nikolopoulos particularly disturbing, sounds like he never understood that rule.

Here are the full excerpts from 2010 April and May items of Geurt Gijssen's column "An arbiter's notebook" that are relevant to the questions raised:

==== April 2010 column ====== geurt144.pdf =========
Question In a tournament, a game ended with the following problem:
The rate of play was 90 minutes for 40 moves, then 30 minutes for the remaining moves with an increment of 30 seconds from move one. I was called to arbitrate a dispute by a flagged player. It seemed that both players had played forty moves (according to their scoresheets), while the clock displayed thirty-nine moves for either player. It is almost clear that they missed a clock press, so that they could ask for 30 seconds more if the game could be restarted. But one of the player's flag displayed, so after checking that the clock was functioning properly (it was), I declared the game lost on time by the flagged player. (I had no way of knowing when he played his fortieth move, before or after the flag fall, or if he would have flagged or not in case he had been awarded the missing 30 seconds.)
He wanted me to add the 30 seconds of increment for the fortieth move and restart the game, which I rejected, saying that he was responsible for pressing the clock after each move, which he failed to perform correctly. The reason may be a light pressing of the clock, though I couldn't make it happen myself.
What is your impression of this event? Pierre Becker (France)

Answer I remember only one tournament in which we had the same kind of problem: it was the European Championship in Ohrid 2001. I was called upon several times because of flag falls, although forty moves were played in the first period. And every time the scoresheets showed forty moves and the move counter of the clock only thirty-nine moves. When we checked the scoresheets, the figure forty was correct. Article 6.7.a is relevant to answer your question:
During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent's clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See the Articles 5.1.a, 5.2.a, 5.2.b, 5.2.c and 9.6.)

The first part of the last sentence is essential. A move is completed after it is made and the player's own clock was stopped and the opponent's started. This means that the last move before the time control, in your case the fortieth, is considered to be completed after the clock has been stopped. If both clocks showed thirty-nine moves, then the last move was played and also completed by the player of the black pieces. In that case only the flag of the player of the white pieces could fall when he was thinking about his fortyfirst move. I assume that they made forty moves as both scoresheets indicated. The game should continue and the arbiter has to adjust the clocks. I am afraid that I have to inform you that your decision was wrong.

...

Question Dear, Mr. Gijssen. I am suspicious about a recent decision, because I don't understand all the conclusions made. Here is the situation: In Upper Austria, we play team matches without an arbiter and with DGT clocks. Normal play is 40 move 100 min, 50 min for the rest of the game, additional 30 seconds each move.
The following situation occurred: I (White) made my forty-first move, got up from the table to get some coffee, looked back at the board, and saw the clock of my opponent signaling -0.00 / blinking. So I claimed victory of the game. A member of the other team assayed that after the fortieth move "it is too late" to claim a time violation, because the second part of the time control has started and he will get an additional 50 min. I answered that my clock was
showing 1h 27min, so his move forty must have been out of time. I just didn't realize it immediately, but the DGT stops counting time after -0.00 is registered for one side. The other team later protested the acceptance of this time violation. There reasoning was that a) it was too lateť b) maybe the clock malfunctioned, but no evidence was found of this, or c) maybe less than forty moves were made. (I
argued that calling the time violation a move earlier would still result with a -0.00 on the clock.) In the end, weeks later the judgement was 0-1. I lost because I "didn't continue the game."
Article 6 states, "the flag is considered to have fallen when a player has made a valid claim to that effect." However, I did not have success. Here are my questions:
. What is a valid claim?
. How do you view this situation?
. Must one claim a time violation immediately after the fortieth move? Is it necessary to look at the clock on every move to instantly claim a time violation?
Many thanks in advance. Harald Eder (Austria)

Answer I think that this situation is very clear. You made your fortieth move and pressed the clock. Fifty minutes were added to your time and the opponent's clock started to run. Your opponent started to think about his fortieth move. Then, before he pressed his clock, his flag fell. It is irrelevant whether his flag fell before he made his move or after he made his move. It is clear that he overstepped the time limit. He should lose the game, because he didn't complete the required number of moves (in this case forty) in the allotted time. The big advantage of the DGT clock is that the clock shows the
overstepping of the time control. For a claim to be valid, the clock of the opponent must have fallen (provided that the clock was not malfunctioning) and the required number of moves has not been completed. As far as I can see, your claim fulfilled all these requirements. It is not necessary to watch the opponent's clock all the time. But it is probably wise to do so around the fortieth move, especially in time trouble. It is possible to claim a flag fall beyond the fortieth move (in cases where the DGT clock is used), but it always creates controversy. By the way, with an analogue clock, a flag claim after some moves have been played is not possible in my opinion, because there is no evidence that the flag fell before the player completed his fortieth move.
===================

==== May 2010 column ===== geurt145.pdf ==========
(A correction to the previous answer to Harald Eder):

Question Dear, Mr. Gijssen. I refer to the April 2010 column. I am the arbiter who made the decision in this case. You believe that this situation is very straightforward. I disagree. I think it is not so clear and my assessment of the case follows:
There are no doubts about the following:
1. It is a team championship match without an arbiter.
2. The playing time is 100 minutes for 40 moves, 50 minutes for the rest of the game and an additional 30 seconds for each move from the start of the game.
3. The move counter is activated.
4. White made his forty-first move, stopped his own clock and started Black's clock. White saw that the opponent's clock was "0.00" and claimed "flag fall." At this point, White's forty-first move had been completed.
5. We don't know the number of moves taken into account by the move counter.
There are two possibilities as to what may have happened:
1. 1. Black's flag fell during his fortieth move, but White saw this only
after White made his forty-first move.
2. 2. Black completed all forty moves in the allotted time, but the move counter was not identical with the number of moves played. (Perhaps the player(s) forgot to press the clock). Therefore, it is possible that the player was on the second time period of his game having completed his fortieth move.
At this point, it is not possible to say which one of the above occurred. If point one occurred, then, yes, it is possible to claim a flag fall beyond the fortieth move in cases where a digital clock is used. I feel two conditions must be fulfilled to justify this. Firstly, the digital clock does not stop after a flag fall and secondly, the number of moves on the move counter is identical with the number of moves played on the chessboard. If not, a digital clock acts as
an analogue clock. As you say in the case of an analogue clock, White
destroyed the proof of the flag fall with his forty-first completed move. There is no evidence that the flag fell before the player completed his fortieth move. The claim is too late.
So why must the move counter be identical to the played moves on the chessboard? This brings me to point number two above. As an arbiter I have often experienced that players forget to press the clock or White starts the clock after making his first move. After a flag fall, there is often a misunderstanding, as the players fail to see the reason why the flag has fallen. They cannot accept that they may have forgotten to press the clock. An arbiter can normally resolve such a situation easily. Article 6.3 says that immediately
after a flag fall, the requirements of Article 6.2 a. must be checked. In our case this means that both players had made the minimum number of moves in the first time period. Accordingly, my decision was that the claim of the white player was too late. It was not possible to confirm that Black had overstepped the time limit. Both
cases mentioned above are possible and the game should have continued. White lost the game because he left the tournament area. I will not explain here how to handle a protest in Upper Austria if there is no arbiter. As I said, the players sometimes forget to press the clock, which can cause problems after a flag fall. Therefore, I suggest that when the game is played with an increment of thirty seconds or more with each move, the clock shows the move counter. Also the players must record their previous move before
making the next move. This would make it apparent to the Arbiter and the players if the move counter is not identical with the recorded moves. I look forward to your response. Best regards, IA Mitterhuemer Günter (Upper Austria)

Answer The "mistake" the players made, after the flag fall was noticed, was that they did not check the number of moves in the move counter. When the move counter shows the same figures as the score sheets, then there are no doubts. I admit the fact that the flag fall was noticed after move forty-one created some doubts, because it was really possible that both players didn't press the clocks. By the way, in 2006 there was a FIDE decision that the move
counter should not be activated, but, as far as I can see, in many events, including the match Anand – Topalov, the move counter was activated. I revise my conclusion in the previous Notebook. It is not completely certain that the player of the black pieces overstepped the time. Therefore, he should have the benefit of the doubt and the game should be continued with fifty minutes for the second period.
Another point is whether it is correct to forfeit the game for the player of the white pieces. I can understand that he was of the opinion that he was completely right. Perhaps there is a reason to change this decision. One consideration might be to nominate arbiters (for instance, the captains) for team matches.
===================

antichrist
20-11-2011, 09:43 PM
on the old paper pad scoresheets there was no line printed after the 40th move (that I can remember) yet I cant remember any problems with the 40th move being recorded and acted upon for manual clock adjustment

Keong Ang
21-11-2011, 02:32 PM
I think that the Chairman of the Rules commission (IA Geurt Gijssen) and the Chairman of the Arbiters commission (IA Panagiotis Nikolopoulos) should take a good wrestle to settle the dispute. Obviously they don't agree. I must admit I find the interpretation of rule 6.8 of Nikolopoulos particularly disturbing, sounds like he never understood that rule.
Thanks for your information. I have previously read IA Geurt Gijssen's April&May2010 articles when you've mentioned them earlier in the thread. Nice to reread them again.

I don't think IA Geurt Gijssen and IA Panagiotis Nikolopoulos disagree on the interpretation of article6.8 and I believe they both understand it very clearly. When I read the quoted articles, all the actions and discussion can only happen if everyone understood that flag fall occurs when it happened, rather than when it is observed. If flag fall occurs when it is observed, there would be no need to even debate when the flag fell etc.

Interpreting article6.8 as "flag fall occurs when it happened" seems to be IA Geurt Gijssen's point of view even for analog clocks. I quote from his January2001 article .



Question Dear Mr. Gijssen, I became a regional arbiter (first step in the ladder in my country) in
October and was immediately thrown to the wolves. Here is what happened the first time I had to
witness a quick-play finish (in Turin, at the end of November). Black had seemingly very little
time on the clock. White was with about five minutes. They played a few moves, then White, who
was clearly winning but was starting to run short of time, made the mistake of taking Black's last
rook, not realizing there was a stalemate.

Game over? Well... In a matter of very few seconds (no more than five) I realized that Black has
overstepped his time (his clock indicated 6.01), but due to a mechanical fault, his flag was still
"up" (and, by the way, there was no way to make it fall no matter how hard we tried - apart from
"opening" the clock, of course). What to do? I had no idea, so I reported the whole matter to my
chief arbiter and he made the final decision: White wins.

Was it the right decision? After the fact I talked with other arbiters but we didn't come to an
agreement. The majority says that stalemate means draw no matter the surroundings. My opinion,
though, is different and agrees with the decision taken by my chief arbiter. Obviously nobody
can't tell what would have happened if the flag had fallen as it should have, but as White was
paying attention at Black's clock and I was at the table, I can easily figure out that either of us
would have noticed a fallen flag. If my more experienced colleagues are right, the game should
have been a draw. I am perplexed, though.

Please, let me allow muddling the picture a little bit. Forget the stalemate. Let's suppose that
White's flag has fallen and the same sequence of event follows; that is the arbiter realizes in a
matter of five seconds that Black's time expired about one minute before, but, because of a
mechanical fault, his flag didn't fall. Should Black be declared a winner? My colleagues say so,
but I don't feel this can be a right decision. Of course, if I had realized earlier that there was a
mechanical fault, things would have been a lot smoother. But it was difficult anyway and, as I said
above, it was my first tournament and I didn't imagine that a flag "up" couldn’t always be trusted.
Roberto Ricca (Italy)

Answer Article 6.10 talks about a defective chess clock: Every indication given by the clocks is
considered to be conclusive in the absence of any evident defect. A chess clock with an evident
defect shall be replaced. The arbiter shall use his best judgement when determining the times to
be shown on the replacement chess clock.

Article 6.9 says that if a player does not complete the prescribed number of moves in the allotted
time, the game is lost by the player.

And finally Article 6.8 says: A flag is considered to have fallen when the arbiter observes the fact
or when a valid claim to that effect has been made by either player.

It is clear there was a mechanical defect. The clock therefore should be replaced, but in this case
there is no need to replace the clock because the game was over. I understand that you are very
sure that at the very moment the stalemate appeared on the board, White had already overstepped
the time limit. There is therefore no doubt that the game is lost by White according to Article 6.9.

But what about Article 6.8? In my opinion Article 6.8 is pre-empted by Article 6.10.

The above question by now FA Roberto Ricca, who has made the JaVaFo pairings engine, shows a situation where the game result was changed due to concluding when flag fall happened. Not only that, but IA Geurt Gijssen even gives his opinion that article6.10 pre-empts article6.8. A statement that makes sense only if flag fall occurs when it happened, as in this situation, neither arbiter, nor players noticed flag fall (the flag was stuck on "up").

Regarding the topic of setting the move counter, I think we now know what this mysterious "....., in 2006 there was a FIDE decision that the move
counter should not be activated,....." is. Credit should go to IA Shaun Press for pointing the location of this decision out to me.

Minutes 77th FIDE Congress 27May-5June2006, Turin, Italy. Section D Technical Commission:

Mr. Stubenvoll said that the Technical Commission decided not to allow the use of any kind of move counters in any FIDE events. From 1 January 2008 all chess clocks producers should ensure that this move counter could be disabled.
FIDE GA approved this report.

I noticed from the Technical Commission minutes in annex57 that both IA Gijssen and IA Nikolopoulos were present. Also, the vote to not use move counters were unanimous with one abstention. So far I haven't found any decision that changed things since 2006. So, why do all FIDE events currently have the move counter set? Seemingly in contravention of the Turin decision?

Looks like something I have to follow up with IA Nikolopoulos. ;)

Jesper Norgaard
21-11-2011, 05:07 PM
Thanks for your information. I have previously read IA Geurt Gijssen's April&May2010 articles when you've mentioned them earlier in the thread. Nice to reread them again.;)
You are welcome ;)


I don't think IA Geurt Gijssen and IA Panagiotis Nikolopoulos disagree on the interpretation of article6.8 and I believe they both understand it very clearly. When I read the quoted articles, all the actions and discussion can only happen if everyone understood that flag fall occurs when it happened, rather than when it is observed. If flag fall occurs when it is observed, there would be no need to even debate when the flag fell etc.
Firstly I disagree with you here - I don't see any evidence of that everyone understands that flag fall occurs when it happens. Try again with other cases! ;)

IA Nikolopoulos allegedly wrote: "So the right interpretation of 6.8 for games with standard time control (not rapid, not blitz) is that the Arbiter has to interfere immediately with the game, when we have a flag fall (i.e. when a player didn't complete the required number of moves in the allotted time and the game shall be ended). Because in fact the flag fall has to be observed (called) at the same time that it happened. And it is mainly an Arbiter's duty."
It's the arbiter's duty all right. But if the arbiter always interferes immediately, the "considered" clause is in fact never invoked. Nikolopoulos seems to argue that the police should be right there to make the arrest when the burglar breaks in. That is definitely NOT the point of rule 6.8.

The January 2001 example is interesting but hardly convincing since as Geurt says article 6.10 pre-empts article 6.8, the wording of 6.8 thus never came into consideration. I agree with that by the way. Nonetheless I also disagree with Geurt and the chief arbiter in the decision. To me the articles 5.1(a) [checkmate], 5.1(b) [resignation], 5.2(a) [stalemate}, 5.2(b) [dead position], 5.2(c) [draw by agreement] all trumphs 6.10 unless the 6.10 claim came first. In this case, stalemate happened at the board, which literally ends the game with a draw acording to the rules. The players were also responsible to check that the clock was working, and their actions on the board trumphs any flag falls or clock anomalities.
Ricco talks about muddling the water with white flag fall even though black's flag would have happened a minute before had it not been for the defective clock. In that case I consider the flag falls being of equal value and should per 6.8 be considered to have happened at exactly the same time - when it was discovered by the arbiter - but the case should be handled as a normal double flag fall in the last period, where a winner must be found, and the clock shows which flag fell first (the unnoticed flag fall) so black wins.



Regarding the topic of setting the move counter, I think we now know what this mysterious "....., in 2006 there was a FIDE decision that the move counter should not be activated,....." is. Credit should go to IA Shaun Press for pointing the location of this decision out to me.

Minutes 77th FIDE Congress 27May-5June2006, Turin, Italy. Section D Technical Commission:

FIDE GA approved this report.

I noticed from the Technical Commission minutes in annex57 that both IA Gijssen and IA Nikolopoulos were present. Also, the vote to not use move counters were unanimous with one abstention. So far I haven't found any decision that changed things since 2006. So, why do all FIDE events currently have the move counter set? Seemingly in contravention of the Turin decision?

Indeed. Thank you and especially Shaun Press for this information, that was long awaited! The info is in fact available online on the FIDE web site:
http://www.fide.com/images/stories/NEWS/download/annexes/ga2006/annex57.pdf



Looks like something I have to follow up with IA Nikolopoulos. ;)
Absolutely, and indeed something to bring up with Geurt. In my opinion this information belongs directly in the laws of chess, so that maximum awareness is secured - we can see how this has drifted from being an unanimous decision in an relatively hidden technical report to being essentially forgotten by key International Arbiters. Either it should be in the laws of chess or in C.02.5.1 of the Handbook on the FIDE web site.

The key point is also the justification in the report: "Except for the last time period a clock should not get into a “frozen state” when the time in a period is entirely used".

Keong Ang
21-11-2011, 11:06 PM
Firstly I disagree with you here - I don't see any evidence of that everyone understands that flag fall occurs when it happens. Try again with other cases! ;)

IA Nikolopoulos allegedly wrote: "So the right interpretation of 6.8 for games with standard time control (not rapid, not blitz) is that the Arbiter has to interfere immediately with the game, when we have a flag fall (i.e. when a player didn't complete the required number of moves in the allotted time and the game shall be ended). Because in fact the flag fall has to be observed (called) at the same time that it happened. And it is mainly an Arbiter's duty."
It's the arbiter's duty all right. But if the arbiter always interferes immediately, the "considered" clause is in fact never invoked. Nikolopoulos seems to argue that the police should be right there to make the arrest when the burglar breaks in. That is definitely NOT the point of rule 6.8.
We've disagreed on article6.8 before. I thought that perhaps my jet-lagged memory failed me during my FA seminar. From IA Nikolopoulos's reply he's essentially confirmed what I remembered. Flag fall is considered to have fallen in the past when the arbiter or either player finally notices it.

Maybe IA Nikolopoulos is wondering how a FA trained by him could still be asking such a basic question when he was answering my question. Most of the scenarios on correct handling of flag fall during the FA seminar would be nonsense since it would be pointless for an arbiter to need to determine when flag fall occurred if it officially occurs when noticed.


The January 2001 example is interesting but hardly convincing since as Geurt says article 6.10 pre-empts article 6.8, the wording of 6.8 thus never came into consideration. I agree with that by the way. Nonetheless I also disagree with Geurt and the chief arbiter in the decision. To me the articles 5.1(a) [checkmate], 5.1(b) [resignation], 5.2(a) [stalemate}, 5.2(b) [dead position], 5.2(c) [draw by agreement] all trumphs 6.10 unless the 6.10 claim came first. In this case, stalemate happened at the board, which literally ends the game with a draw acording to the rules. The players were also responsible to check that the clock was working, and their actions on the board trumphs any flag falls or clock anomalities.
Ricco talks about muddling the water with white flag fall even though black's flag would have happened a minute before had it not been for the defective clock. In that case I consider the flag falls being of equal value and should per 6.8 be considered to have happened at exactly the same time - when it was discovered by the arbiter - but the case should be handled as a normal double flag fall in the last period, where a winner must be found, and the clock shows which flag fell first (the unnoticed flag fall) so black wins.
Your opinion rests on interpreting article6.8 as flag fall officially occurs when it is noticed. However, the decision IA Gijssen arrives at depends on interpreting article6.8 as flag fall occurs when it happened.

Even in Geurt's articles that you've quoted, I'm wondering why Geurt would even bother with the situation at the time the flag fell since it is irrelevant if "flag falls when it is noticed" instead of "flag falls when it happened".

I think it would be much simpler to be the arbiter if "flag falls when it is noticed" since there will be no need for further investigation on when the flag actually fell.


Indeed. Thank you and especially Shaun Press for this information, that was long awaited! The info is in fact available online on the FIDE web site:
http://www.fide.com/images/stories/NEWS/download/annexes/ga2006/annex57.pdf


Absolutely, and indeed something to bring up with Geurt. In my opinion this information belongs directly in the laws of chess, so that maximum awareness is secured - we can see how this has drifted from being an unanimous decision in an relatively hidden technical report to being essentially forgotten by key International Arbiters. Either it should be in the laws of chess or in C.02.5.1 of the Handbook on the FIDE web site.

The key point is also the justification in the report: "Except for the last time period a clock should not get into a “frozen state” when the time in a period is entirely used".

Regarding not setting the move counter, my suspicion is that the situation has changed and this has drifted into another irrelevance, watered down to a FIDE recommendation.
Note that this very same Technical Commission now tests clocks so that they only get FIDE endorsed if the next time control is entered on the 40th press of the clock lever. For all intents and purposes, instead of implementing the 2006 ruling, the exact opposite has occurred. I'm wondering if there was a more recent ruling that superseeded the 2006 one.

In any case, arbiters don't really bother about obscure minutes, a ruling needs to be in the regulations to be applicable. The regulations that are relevant to arbiters are listed in the Arbiters Commission website, they are actually subsets of the FIDE Handbook.

I can understand why IA Nikolopoulos says the correct way to set the clock is to set the move counter to 40moves. Then reinforces it by saying it is set for all top tournaments and world FIDE and Continental events. That would be the only way to comply with C.02.5.1.b. that is the same as C.06.10.c.1. "The display at all times should show the time available to complete a player’s next move."
So, if a player had 30seconds left on the clock before entering the next time control, the clock should now display 31minutes left (30seconds left + 30minutes in next time control + 30seconds increment).
If the move counter was set to zero, the clock would display 1minute (30seconds left + 30seconds increment). A situation where the time available to complete the player's next move (31minutes) is not being displayed, instead the erroneous 1minute is displayed.

Regarding when clock lock-up should happen, the 2006 minutes record a decision to only allow lock-up in the final time control. What we see is that far from implementing the decision in the regulations, we instead have clauses that implement clock lock-up in the final time control WITHOUT disallowing it for earlier time controls. A clock that can lock-up upon flag fall in any time control that is not the last is also fully compliant with the regulations.

I'll have to take some time to think about all these issues before following up with IA Nikolopoulos again. Don't want him to revoke my FA!! :uhoh: :whistle:

Santa
22-11-2011, 10:46 AM
The procedure is for the arbiter to check that the scoresheets show sufficient moves when the clock adds the 2nd control's time. Insufficient moves recorded in the scoresheet means the players have been clicking the clock repeatedly without making moves. Definately incorrect procedure by the players.



So my opponent makes a move, presses the clock button, then knocks over pieces on the board.

In a previous era, I'd simply restart my opponent's clock so my time would not be spent while I couldn't use the board. What should I do now?

If I proceed now as formerly, we'd bot get an increment (which isn't especially unfair to us but might be to others), and if the clock's counting moves, we'd "reach time control" early, again not unfair to us, except maybe if one of is in immediate risk of loss on time and this averts the hazard.

Jesper Norgaard
22-11-2011, 03:28 PM
So my opponent makes a move, presses the clock button, then knocks over pieces on the board.

In a previous era, I'd simply restart my opponent's clock so my time would not be spent while I couldn't use the board. What should I do now?

If I proceed now as formerly, we'd bot get an increment (which isn't especially unfair to us but might be to others), and if the clock's counting moves, we'd "reach time control" early, again not unfair to us, except maybe if one of is in immediate risk of loss on time and this averts the hazard.
You're right it is incorrect to just press back. You are supposed to stop the clock and call the arbiter, to explain the offense of your opponent of correcting pieces in your time. The result will most likely be a warning to your opponent, possibly a time compensation to you for the disturbance (1-2 minutes), and possibly a reduction in time for your opponent (1-2 minutes).

ER
22-11-2011, 03:56 PM
Hey you two, are you still going? :eek: :doh: :lol: good luck anyway! :clap:

Jesper Norgaard
22-11-2011, 04:05 PM
Most of the scenarios on correct handling of flag fall during the FA seminar would be nonsense since it would be pointless for an arbiter to need to determine when flag fall occurred if it officially occurs when noticed.
Any examples of that?



Your opinion rests on interpreting article6.8 as flag fall officially occurs when it is noticed. However, the decision IA Gijssen arrives at depends on interpreting article6.8 as flag fall occurs when it happened.
Wrong. His interpretation depended on giving higher priority to 6.10 than 6.8.



I think it would be much simpler to be the arbiter if "flag falls when it is noticed" since there will be no need for further investigation on when the flag actually fell.

Yes but it would be against the laws, specifically 6.8. You're supposed to use the laws as they are written.



I'll have to take some time to think about all these issues before following up with IA Nikolopoulos again. Don't want him to revoke my FA!! :uhoh: :whistle:
I was thinking more in the lines of revoking his IA :uhoh: :whistle:, but I guess there is not a snowballs chance in Hell that will happen. Still it would be interesting to know why there has been a 180° drift from the 2006 Turin decision of not using the number of moves per period setting. Certainly I have no evidence that it would be his fault. Indeed if a new decision exists from FIDE reversing the 2006 decision, it might have been perfectly "legal". On the other hand, living up to own decisions is not something FIDE has shown a great performance in, just look at the botched WCh cycles, so I actually doubt there was ever a conscious decision.

Garvinator
22-11-2011, 04:12 PM
Indeed if a new decision exists from FIDE reversing the 2006 decision, it might have been perfectly "legal". On the other hand, living up to own decisions is not something FIDE has shown a great performance in, just look at the botched WCh cycles, so I actually doubt there was ever a conscious decision.
I think in these type of circumstances where two opposite actions (whether to set the move counter or not) are both regarded as legal under the fide laws of chess and permissable by the type of clock, then the best course of action is to use the setting that is less likely to cause problems.

Which still in my opinion is to not use the move counter.

Jesper Norgaard
22-11-2011, 04:13 PM
Hey you two, are you still going? :eek: :doh: :lol:
Shhh, Jak, don't disturb the geniuses, we are trying to make FIDE history here, ... or perhaps settle a dog fight, I don't know :hmm:



good luck anyway! :clap:
Thanks!

Jesper Norgaard
22-11-2011, 04:23 PM
I think in these type of circumstances where two opposite actions (whether to set the move counter or not) are both regarded as legal under the fide laws of chess and permissable by the type of clock, then the best course of action is to use the setting that is less likely to cause problems.

Which still in my opinion is to not use the move counter.
Indeed, and in fact corresponds to the apparently latest official decision in Turin 2006. That is why the following quote is puzzling:



In my opinion the correct way to set the clock is to put 40 moves in the move counter for the period 1 and 00 moves for the period 2. This is what it is used in all top tournaments and world FIDE and Continental events.

Go figure.

Keong Ang
22-11-2011, 05:49 PM
So my opponent makes a move, presses the clock button, then knocks over pieces on the board.

In a previous era, I'd simply restart my opponent's clock so my time would not be spent while I couldn't use the board. What should I do now?

If I proceed now as formerly, we'd bot get an increment (which isn't especially unfair to us but might be to others), and if the clock's counting moves, we'd "reach time control" early, again not unfair to us, except maybe if one of is in immediate risk of loss on time and this averts the hazard.

You're right it is incorrect to just press back. You are supposed to stop the clock and call the arbiter, to explain the offense of your opponent of correcting pieces in your time. The result will most likely be a warning to your opponent, possibly a time compensation to you for the disturbance (1-2 minutes), and possibly a reduction in time for your opponent (1-2 minutes).

Jesper Norgaard has the correct course of action.
Just in case, "stop the clock" means pause the clock.

Keong Ang
22-11-2011, 06:48 PM
Any examples of that?
I better not divulge further details here as the examples were later used as the basis of the FA exam questions. IA Nikolopoulos reply to my questions confirms to me that my recollection is more or less correct. This means that the interpretation of article6.8 as "flag fell when it happened" is correct. Anyone who had tried to answer the exam questions using a "flag fell when it is noticed/called" would have got their decision making logic wrong and made the wrong decision. Result is that they would have got sufficient answers wrong to fail the FA exam.


Wrong. His interpretation depended on giving higher priority to 6.10 than 6.8.
Exactly what I meant by Geurt interpreting article6.8 as "flag fell when it happened". To give higher priority to article6.10 demonstrates that he places greater priority on the evidence given by the clock. The situation was one where flag fall was not noticed/called because the flag was stuck. However the arbiter knows the first flag fell before the draw position occurred, and that overrode the draw.


Yes but it would be against the laws, specifically 6.8. You're supposed to use the laws as they are written.
I recall one FA candidate who kept insisting on interpreting article6.8 as "flag falls when it is noticed" being corrected by IA Nikolopoulos. Essentially the "considered to have fallen" in the past tense means it fell when it happened. Interpreting it as "flag fell when it is noticed/called" was dismissed as illogical and only possible if going off on a legalistic tangent. That FA candidate did not pass the exam.

The other IAs who were assisting during that FA seminar also used article6.11 that starts "If both flags have fallen and it is impossible to establish which flag fell first then:" to demonstrate that "flag fell when it happened" is the only interpretation that would let article6.8 work in harmony with the other articles. Article6.11 explicitly requires the arbiter to investigate when flag fall happened. If it is possible to know which flag fell first, the game ends.


I was thinking more in the lines of revoking his IA :uhoh: :whistle:, but I guess there is not a snowballs chance in Hell that will happen. Still it would be interesting to know why there has been a 180° drift from the 2006 Turin decision of not using the number of moves per period setting. Certainly I have no evidence that it would be his fault. Indeed if a new decision exists from FIDE reversing the 2006 decision, it might have been perfectly "legal". On the other hand, living up to own decisions is not something FIDE has shown a great performance in, just look at the botched WCh cycles, so I actually doubt there was ever a conscious decision.
I'm not even contemplating such futile thoughts. :lol:

On the face of it, it looks like the decision of the Technical Commission was simply ignored by subsequent General Assemblies, hence the seeming 180° drift. In any case, the current regulations that are in effect do not prohibit the use of the move counter. In fact, they seem to mandate it's use for full compliance.

There is also article6.13 "If an irregularity occurs and/or the pieces have to be restored to a previous position, the arbiter shall use his best judgement to determine the times to be shown on the clocks. He shall also, if necessary, adjust the clock’s move counter." Which is something that is unnecessary if the move counter does not regulate the passage of time controls. During my FA seminar, we were trained in the correct procedure in applying article6.13. Resetting the move counter was required if it happened before move40, but unnecessary in the last time control. It was assumed that the move counter would always be set to 40moves to regulate the passage of time controls.

Keong Ang
22-11-2011, 07:09 PM
I think in these type of circumstances where two opposite actions (whether to set the move counter or not) are both regarded as legal under the fide laws of chess and permissable by the type of clock, then the best course of action is to use the setting that is less likely to cause problems.

Which still in my opinion is to not use the move counter.
You're right here. Both actions (whether to set the move counter or not) are legal under FIDE regulations.

Actually there are 2 ways to handle the rate of play you mention.
(a). the player goes to the second period ( 30 minutes are added to his time ) after he completes 40 moves.
(b). the player goes to the second period ( 30 minutes are added to his time ) after he completes 2 hours of game.
My opinion is still that it would benefit the players if the move counter is not used. It increases the chances of the game continuing and a result obtained over the board.

In any case, I'd defer to whatever setting the Chief Arbiter of the tournament decides on.

If I was the Chief Arbiter:eek: :eek: , I'll set the move counter if the tournament falls under "all top tournaments and world FIDE and Continental events." Simply because that was how I was trained and informed.:wall: :wall:
If the tournament is something else, I'd set the move counter to what benefits the players the most. :whistle:

I don't think either setting would have less problems than the other. It really depends on what kind of problems you'd prefer to deal with. :uhoh:

Keong Ang
22-11-2011, 07:15 PM
Shhh, Jak, don't disturb the geniuses, we are trying to make FIDE history here, ... or perhaps settle a dog fight, I don't know :hmm:
Dog fight??? :eh:
I feel like a dog chasing his tail!! :rolleyes:

I think the chances of us making FIDE history are less than the chance of being made history by FIDE. :hand:

Thanks!
Yes JaK, we'll need all the luck we can get!

ER
22-11-2011, 07:34 PM
Look guys, joking aside, seeing people taking so seriously the arbiter's responsibilities and approaches to real tournament questions, is encouraging. In Chess we need responsible people like yourselves and no devil may care characters who just decide on matters on the spirit of the moment! Keep up the good work! :clap:

Santa
23-11-2011, 11:18 AM
You're right it is incorrect to just press back. You are supposed to stop the clock and call the arbiter, to explain the offense of your opponent of correcting pieces in your time. The result will most likely be a warning to your opponent, possibly a time compensation to you for the disturbance (1-2 minutes), and possibly a reduction in time for your opponent (1-2 minutes).

In the tournaments I'm likely to play in, I think processing as in the past would be more expedient. I don't think a penalty of a minute or two would adequately compensate for the emotional disturbance of stopping clocks, finding an arbiter etc etc. Especially as the arbiter is likely to be busy with his own game.

If it's something we don't agree with, then yes we need an arbiter.

In the days of BHB clocks, it was quite usual for one of the players to adjust the clocks when necessary, showing to opponent so as to get confirmation that it's done adequately, and then carry on.

Santa
23-11-2011, 11:29 AM
Look guys, joking aside, seeing people taking so seriously the arbiter's responsibilities and approaches to real tournament questions, is encouraging. In Chess we need responsible people like yourselfs and no devil may care characters who just decide on matters on the spirit of the moment! Keep up the good work! :clap:

I agree with you Jak, it's good to see arbiters so keen to get it right.

However, deferring to His Holiness the Cheif Arbiter is always sensible so that all the assistance and deputies do it the same way.

And if you are His Holiness, persumably you have the experience to choose a course that isn't contentious, and can say, "OKay Chaps[1], we will do this and that.."

1. Chaps includes Chicks as approriate.

ER
23-11-2011, 11:40 AM
I agree with you Jak, it's good to see arbiters so keen to get it right.

However, deferring to His Holiness the Cheif Arbiter is always sensible so that all the assistance and deputies do it the same way.

And if you are His Holiness, persumably you have the experience to choose a course that isn't contentious, and can say, "OKay Chaps[1], we will do this and that.."

1. Chaps includes Chicks as approriate.

:lol:Also maybe a prayer (on our knees) to his Holiness before the start of each round could be of some (divine) help! :lol:

Jesper Norgaard
23-11-2011, 12:04 PM
I agree with you Jak, it's good to see arbiters so keen to get it right.

Thanks.



However, deferring to His Holiness the Cheif Arbiter is always sensible so that all the assistance and deputies do it the same way.

And if you are His Holiness, presumably you have the experience to choose a course that isn't contentious, and can say, "Okay Chaps[1], we will do this and that.."

1. Chaps includes Chicks as approriate.

Santa, I'm really lost. What on earth are you talking about. This is even more cryptic than AC. I have once in a while seen some chicks coming to chess tournaments, but Chaps?
(a) chapters
(b) guys
(c) assistant arbiters
(d) Australian aborigin tribe members
(e) chapels

Ehhh, I give up, and what is that about the contentious course. I'm clueless. Is it a combined arbiter course and tournament? Is there a dictionary for this kind of speak? :confused:

Keong Ang
20-01-2012, 01:52 PM
A plus for setting the move counter occurred during round5 of Queenstown 2012, at board5.

Rate of play is "100 minutes for 40 moves followed by 50 minutes for 20 moves, then 15 minutes for the remaining moves with 30 seconds cumulative increment for each move starting from first move"
FIDE's longest rate of play with increments approved for titles.

At move 37, black played Qd7 and got flagged just before pressing the clock.
In this case no question about black losing on time arose because the clock locked-up and since the game was broadcast on a DGT eboard, a lot of data was captured, including clock times.

If the move counter was not set, the black player would still lose on time, but it would have required arbiter intervention.
Both scoresheets recorded black's 37th move, data capture shows that move37 was made but not completed in time. So no dispute can arise in this case.

If the move counter was not set, another 50minutes would have been added (2nd time control started) and a 30second increment would also have been added. The clock would let the game continue and it has become the arbiter's duty to stop the game.
The 2nd time control cannot apply at move37.

Practically, from an arbiter's perspective, it is much simpler to have a game ended with a result if the clock locks up upon expiration of time. It would be more disruptive and perplexing to both players and spectators if the clock now displays that there's 50minutes remaining for black and the arbiter says game-over.

As I am currently sitting at the arbiter's desk looking at the playing area, clocks that display the actual remaining time for each player are less confusing to everyone concerned than those that require cross checking with scoresheets.
Most applicable when players make 60moves and hit the 3rd time control to get an additional 15minutes.

There are 148 players on 74boards here and by setting the clock's move counter to regulate the passage of time controls, it is possible for a single arbiter to supervise the tournament.
A quick glance at the players scoresheets and clock display is sufficient to pick up on most possible errors.

Most players here look at the clock adding 50minutes when they complete the move40 and adding 15minutes when they complete move60. It is more comprehensible to the players than to publish a time control and then field queries from all the players when the clock did not add time at move40.