Bug found: early draw based on 3-move-repetation!
-
- Posts: 4
- Joined: Tue Feb 19, 2002 7:02 am
Bug found: early draw based on 3-move-repetation!
Hi. There was some bug at this site concerning draw by 3-move-repetation some time ago, and interestingly, either this bug has returned or a new one has been found! Since in my game against amikhlin http://www.slowchess.com/viewgame.cgi?p1=g1104948220, I noticed draw was ruled, but because it struck me by real surprise, decided to look upon it, and to me, it seems the exact same position has been reached only twice! Since at the first time there was a pawn on h7, which is now on h5, and therefore different position, although black is to move on each move. Sometime ago I made same kind of post than this, and I didn´t know/remember that same 3 moves don´t have to happen exactly right after one another. But I think this time there is a flaw again in site´s software. I guess game is over anyway. I hope I´ve helped to improve this awesome site twice now.
Bug found: early draw based on 3-move-repetation!
I agree, it's a bug. The posistions after 34...h5, 37. Qc3 and 39. Kc1 are identical except for one thing; Who it is to move.
Nils
--
Consider donating some computer time to science!
Read more here: http://folding.stanford.edu
--
Consider donating some computer time to science!
Read more here: http://folding.stanford.edu
Re: Bug found: early draw based on 3-move-repetation!
[quote="energy"]I agree, it's a bug. The posistions after 34...h5, 37. Qc3 and 39. Kc1 are identical except for one thing; Who it is to move.[/quote]
In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.
The board position appeared for the third time when 39 Kc1 was played. I believe that is all that is necessary to satisfy the rule.
But don't take my word for it. Do one of two things either contact a USCF certified Tournament Director for his opinion or send the game in to Chess Life as I'm sure a much better explanation will follow. I believe it is an excellent example of how to salvage a draw.
A draw by repetition should not be "automatically" granted. It should be claimed and then the software should be "smart" enough to recognise the rule and grant or deny the claim.
Just as in over-the-board games a Chess clock may not make a beep or flash when the time control is finished. It is up to the player to notice and claim the win.
In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.
The board position appeared for the third time when 39 Kc1 was played. I believe that is all that is necessary to satisfy the rule.
But don't take my word for it. Do one of two things either contact a USCF certified Tournament Director for his opinion or send the game in to Chess Life as I'm sure a much better explanation will follow. I believe it is an excellent example of how to salvage a draw.
A draw by repetition should not be "automatically" granted. It should be claimed and then the software should be "smart" enough to recognise the rule and grant or deny the claim.
Just as in over-the-board games a Chess clock may not make a beep or flash when the time control is finished. It is up to the player to notice and claim the win.
Re: Bug found: early draw based on 3-move-repetation!
Not my software, in case you thought so.mateau wrote:In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.energy wrote:I agree, it's a bug. The posistions after 34...h5, 37. Qc3 and 39. Kc1 are identical except for one thing; Who it is to move.
You believe wrong. Read 9.2 here:mateau wrote: The board position appeared for the third time when 39 Kc1 was played. I believe that is all that is necessary to satisfy the rule.
http://www.fide.com/component/handbook/ ... ew=article
I'll leave most of your claims for Greg to deal with, but you seem to never have played in a serious "over-the-board" chess tournament? It's quite normal to have arbiters step in and declare a game lost on time in most of the tournaments I've played in.mateau wrote: Just as in over-the-board games a Chess clock may not make a beep or flash when the time control is finished. It is up to the player to notice and claim the win.
Nils
--
Consider donating some computer time to science!
Read more here: http://folding.stanford.edu
--
Consider donating some computer time to science!
Read more here: http://folding.stanford.edu
-
- Site Admin
- Posts: 1388
- Joined: Sun Mar 14, 1999 11:13 am
- Location: Jeffersonville, IN
- Contact:
Re: Bug found: early draw based on 3-move-repetation!
Being a certified USCF director, I do know that. But it's a trade off I made with making the programming easier, and making administration easier. In reality, it's up to the players to make valid moves, hit the clock, record their game, call mate and stalemate, call the flag, etc. If I had to get involved with all of that, it'd be a full time job. And forcing those duties on the player's makes things confusing. The USCF rules are primarily written for OTB chess, and they don't translate to playing chess by computer directly.mateau wrote:In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.
Funny thing: The USCF rule book points out that the clock should be as silent as possible, and not call any attention to a flag falling, etc. But the "Official Time Controls" document which is an addendum to the rules, commends clocks that do those things. And the Game Time II, the official clock of the USCF, both flashes and beeps by default. Not to mention the loud pop it makes when pressed normally.
Greg Miller
Re: Bug found: early draw based on 3-move-repetation!
gmiller wrote:Being a certified USCF director, I do know that. But it's a trade off I made with making the programming easier, and making administration easier. In reality, it's up to the players to make valid moves, hit the clock, record their game, call mate and stalemate, call the flag, etc. If I had to get involved with all of that, it'd be a full time job. And forcing those duties on the player's makes things confusing. The USCF rules are primarily written for OTB chess, and they don't translate to playing chess by computer directly.mateau wrote:In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.
I wasn't at all suggesting that you get involved. I would make the observation that you simply have a toggle that "Claims or Offers" the draw. It is up to the player to mind his game if he misses a draw then that just as crucial to that game as overlooking a hanging piece.
Silent as possible means turn off the alarm. The official clock may buzz and beep but not during a tournament. Who wants to hear 700 clocks at the World Open all beeping.
The rules are specific about the use of any electronic device being used to determine the outcome of a game. TD's that I know run their tournament with silent clocks. But hey that's not you.
Funny thing: The USCF rule book points out that the clock should be as silent as possible, and not call any attention to a flag falling, etc. But the "Official Time Controls" document which is an addendum to the rules, commends clocks that do those things. And the Game Time II, the official clock of the USCF, both flashes and beeps by default. Not to mention the loud pop it makes when pressed normally.
Re: Bug found: early draw based on 3-move-repetation!
No Arbiter will step in and declare a game lost on time unless a claim has been made that the game was won on time. I don't know of too many people who would declare for a loss. Arbiters do not stand around and wait for the flag to drop. But again we were not discussing whether or not Arbiters declare wins. Just programs declaring draws.energy wrote:Not my software, in case you thought so.mateau wrote:In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.energy wrote:I agree, it's a bug. The posistions after 34...h5, 37. Qc3 and 39. Kc1 are identical except for one thing; Who it is to move.
Did that really require a response?? Does this??
You believe wrong. Read 9.2 here:mateau wrote: The board position appeared for the third time when 39 Kc1 was played. I believe that is all that is necessary to satisfy the rule.
http://www.fide.com/component/handbook/ ... ew=article
I'll leave most of your claims for Greg to deal with, but you seem to never have played in a serious "over-the-board" chess tournament? It's quite normal to have arbiters step in and declare a game lost on time in most of the tournaments I've played in.mateau wrote: Just as in over-the-board games a Chess clock may not make a beep or flash when the time control is finished. It is up to the player to notice and claim the win.
Your guess about my not playing "serious" chess says quite a lot about you. Are you trying to use logical fallacies to add weight to your argument? "Ad Hominen", "Circumstantial Ad Hominem" and "An Apeal to Authority" springs to mind.
In fact I first joined USCF in 1972 and have played in many Tournaments since then. So what was your point?
FIDE 9.3 States: The game is drawn, upon a correct claim by the player having the move......... Hmmmm "Upon a correct claim". That is the key phrase I believe.
We again were not discussing an Arbitration for a win but rather a declaration or decision for an unclaimed and probably unwanted draw. And Arbitration does not "NORMALLY" determine the outcome of a game. It is used sparingly and with extreme discretion.
And please lets not get into that okay.
I did not post to argue. I posted because I had an opinion regarding whether or not the programming should automatically declare a game drawn. In my opinion a simple toggle that either offers or claims a draw is not that difficult for someone with the skills of Mr. Miller. In fact it would make his life a bit easier NOT having to deal with annoying posts like this one I'm sure. I am of course referring to myself.
The only draw that should be automatic is when both players have inadequate material for a checkmate or the game is stalemated. Some players just do not know when the can not win for lack of material. The rest of it is all up to the players to claim. If two players want to chase one another around the chess board for 300 hundred moves then so what. Yes of course the game is drawn by the rules but only if one player decides to claim it. Maybe they both believe that they can win. Before you get on about the arbiter again yes he can declare the draw if the game interferese with the running and scheduling of the tournament. And that is the distinction.
The game in question probably would not have ended in a draw if the software had not butted in. I am sure that Mr. Miller wants to improve the play on his site and that would mean eveloving the software. The Chessbase site had a huge bug involving draw offers that had to be answered and unscrupulous players would constantly offer draws during very short time controls to try and affect the outcome of a game. That was fixed.
Re: Bug found: early draw based on 3-move-repetation!
Of course. Accuracy matters.mateau wrote:Did that really require a response?? Does this??energy wrote:Not my software, in case you thought so.mateau wrote: In actual fact the real bug is that your software determined the outcome of a game. It is up to the player to claim the draw.energy wrote:You believe wrong. Read 9.2 here:mateau wrote: The board position appeared for the third time when 39 Kc1 was played. I believe that is all that is necessary to satisfy the rule.
http://www.fide.com/component/handbook/ ... ew=article
Wrong again. Read 6.9 here:mateau wrote:No Arbiter will step in and declare a game lost on time unless a claim has been made that the game was won on time.energy wrote:I'll leave most of your claims for Greg to deal with, but you seem to never have played in a serious "over-the-board" chess tournament? It's quite normal to have arbiters step in and declare a game lost on time in most of the tournaments I've played in.mateau wrote: Just as in over-the-board games a Chess clock may not make a beep or flash when the time control is finished. It is up to the player to notice and claim the win.
http://www.fide.com/component/handbook/ ... ew=article
Yes, that was part of your discussion. Mine was to point out errors in a few of your arguments. Not only does such errors tend to make a discussion less valuable in itself, but if they are not corrected they tend to spread. And accuracy matters.mateau wrote:Arbiters do not stand around and wait for the flag to drop. But again we were not discussing whether or not Arbiters declare wins. Just programs declaring draws.
No, just giving you the benefit of doubt. My next best explanation would be that since we're from different continents, our tournament experiences may differ. Here in Europe a serious tournament is almost always an international one, where the FIDE rules are followed.mateau wrote:Your guess about my not playing "serious" chess says quite a lot about you. Are you trying to use logical fallacies to add weight to your argument? "Ad Hominen", "Circumstantial Ad Hominem" and "An Apeal to Authority" springs to mind.
The point was that some of your statements were wrong. Unfortunately it seems you keep adding errors to this thread.mateau wrote:In fact I first joined USCF in 1972 and have played in many Tournaments since then. So what was your point?
9.3 is about the 50 moves rule. Accuracy counts.mateau wrote:FIDE 9.3 States: The game is drawn, upon a correct claim by the player having the move......... Hmmmm "Upon a correct claim". That is the key phrase I believe.
9.2 states several conditions for a draw, one of them being (as in 9.3) "a correct claim by the player having the move". This means you must declare a draw only when it's your move. This part says nothing about the position on the board. That's described in the rest of 9.2, which it seems you didn't bother to read?
What I am discussing is not your call to make, though of course you're free to ignore my contribution if you so wish.mateau wrote:We again were not discussing an Arbitration for a win but rather a declaration or decision for an unclaimed and probably unwanted draw.
And then you go on discussing it, anyway. Seems what you really mean is you don't want me to reply to your statements.mateau wrote:And Arbitration does not "NORMALLY" determine the outcome of a game. It is used sparingly and with extreme discretion.
In the international tournaments I participate (and participated) in, it's not only normal, it was and is always so.
Sorry, but no. It matters.mateau wrote:And please lets not get into that okay.
You posted to argue all right, you just don't seem to want others to argue back at you.mateau wrote:I did not post to argue.
And though I was pretty sure what the answer would be, I didn't reply to that part of your statements. Greg answered as I expected, and I have no problem with his views on that issue. His server, his rules.mateau wrote:I posted because I had an opinion regarding whether or not the programming should automatically declare a game drawn.
Can you support this claim with a reference to relevant regulations?mateau wrote:Before you get on about the arbiter again yes he can declare the draw if the game interferese with the running and scheduling of the tournament. And that is the distinction.
Nils
--
Consider donating some computer time to science!
Read more here: http://folding.stanford.edu
--
Consider donating some computer time to science!
Read more here: http://folding.stanford.edu
Wow!
Now that's what I call a lively debate!
(Edge to Energy so far)
(Edge to Energy so far)
Re: Bug found: early draw based on 3-move-repetation!
YOU HAVE A SERIOUS PROBLEM DUDE YOU REALLY NEED TO GET OFF THE PIPE.energy wrote:Of course. Accuracy matters.mateau wrote:Did that really require a response?? Does this??energy wrote: Not my software, in case you thought so.energy wrote: You believe wrong. Read 9.2 here:
http://www.fide.com/component/handbook/ ... ew=article
Wrong again. Read 6.9 here:mateau wrote:No Arbiter will step in and declare a game lost on time unless a claim has been made that the game was won on time.energy wrote: I'll leave most of your claims for Greg to deal with, but you seem to never have played in a serious "over-the-board" chess tournament? It's quite normal to have arbiters step in and declare a game lost on time in most of the tournaments I've played in.
http://www.fide.com/component/handbook/ ... ew=article
Yes, that was part of your discussion. Mine was to point out errors in a few of your arguments. Not only does such errors tend to make a discussion less valuable in itself, but if they are not corrected they tend to spread. And accuracy matters.mateau wrote:Arbiters do not stand around and wait for the flag to drop. But again we were not discussing whether or not Arbiters declare wins. Just programs declaring draws.
No, just giving you the benefit of doubt. My next best explanation would be that since we're from different continents, our tournament experiences may differ. Here in Europe a serious tournament is almost always an international one, where the FIDE rules are followed.mateau wrote:Your guess about my not playing "serious" chess says quite a lot about you. Are you trying to use logical fallacies to add weight to your argument? "Ad Hominen", "Circumstantial Ad Hominem" and "An Apeal to Authority" springs to mind.
The point was that some of your statements were wrong. Unfortunately it seems you keep adding errors to this thread.mateau wrote:In fact I first joined USCF in 1972 and have played in many Tournaments since then. So what was your point?
9.3 is about the 50 moves rule. Accuracy counts.mateau wrote:FIDE 9.3 States: The game is drawn, upon a correct claim by the player having the move......... Hmmmm "Upon a correct claim". That is the key phrase I believe.
9.2 states several conditions for a draw, one of them being (as in 9.3) "a correct claim by the player having the move". This means you must declare a draw only when it's your move. This part says nothing about the position on the board. That's described in the rest of 9.2, which it seems you didn't bother to read?
What I am discussing is not your call to make, though of course you're free to ignore my contribution if you so wish.mateau wrote:We again were not discussing an Arbitration for a win but rather a declaration or decision for an unclaimed and probably unwanted draw.
And then you go on discussing it, anyway. Seems what you really mean is you don't want me to reply to your statements.mateau wrote:And Arbitration does not "NORMALLY" determine the outcome of a game. It is used sparingly and with extreme discretion.
In the international tournaments I participate (and participated) in, it's not only normal, it was and is always so.
Sorry, but no. It matters.mateau wrote:And please lets not get into that okay.
You posted to argue all right, you just don't seem to want others to argue back at you.mateau wrote:I did not post to argue.
And though I was pretty sure what the answer would be, I didn't reply to that part of your statements. Greg answered as I expected, and I have no problem with his views on that issue. His server, his rules.mateau wrote:I posted because I had an opinion regarding whether or not the programming should automatically declare a game drawn.
Can you support this claim with a reference to relevant regulations?mateau wrote:Before you get on about the arbiter again yes he can declare the draw if the game interferese with the running and scheduling of the tournament. And that is the distinction.
YOU ARE DETERMINED TO MAKE THIS INTO AN ARGUMENT AND FOR ME IT'S NOT EVEN A DEBATE. IT'S JUST THAT I HAPPEN TO BE BORED AND UNFORTUNATELY FOR ME I DECIDED TO REPLY TO A FORUM POST. THANK YOU FOR REMINDING ME JUST HOW TEDIOUS AND UNIMPORTANT FORUMS IN GENERAL REALLY ARE.
YOU WILL OF COURSE GO FOR THE LAST WORD. IT'S ALL YOURS. THERE IS ABSOLUTELY NOTHING HERE WORTH DEBATING.
AGAIN DESPITE WHAT YOU THINK MY MOTIVES WERE, OR MY HISTORY IS I DID NOT COME HERE TO ARGUE I HAD AN OPINION REGARDING DRAWS AND YOU SEEM NOW TO NEED TO MAKE THIS ABOUT ARBITRATED WINS, YOUR GLORIOUS INTERNATIONAL TOURNAMENT EXPERIENCE AND WHAT SEEMS TO BE FIDE-SNOB REMARK ABOUT USCF.
AT THE END OF THE DAY OR WEEK WHO REALLY CARES.
SO YOU WIN OKAY!!!!! YOU NEED IT MORE THAN I DO!!!!
SO FEEL FREE TO RANT AD NAUSEUM.
IT'S ALL YOURS!!!!
END OF THREAD.
-
- Site Admin
- Posts: 1388
- Joined: Sun Mar 14, 1999 11:13 am
- Location: Jeffersonville, IN
- Contact:
Re: Bug found: early draw based on 3-move-repetation!
mateau wrote:I wasn't at all suggesting that you get involved. I would make the observation that you simply have a toggle that "Claims or Offers" the draw. It is up to the player to mind his game if he misses a draw then that just as crucial to that game as overlooking a hanging piece.
Yea, that's how it is on ICC. If you have a three move rep, you have to press the draw button to actually draw the game. I was playing a game on there once, had a major advantage in material, but my opponent found a 3 move rep opportunity and went for it. We moved pieces back and forth about 30 times, he obviously wanted the draw, I obviously didn't. His only mistake was not knowing how to make the claim using the software at hand. In a OTB game, the player would have certainly got the TD involved, and drawn the game, but in games played over the Internet, that's not so easily done. In this particular situation, the opponent's play earned him a draw, but the user interface robbed him of it.
I think my implementation strikes a good balance. Three move reps are extremely rare, the opponents normally agree to a draw when they see the opponent's ability for force one. And when they do happen, it's a near certainty that one of the players will want to make the claim.
I'll agree that it is possible, and has probably happened, where there was a three move rep, and neither player wanted the draw, but the software here would have forced it on to them. But it's a rare enough occurrence that I'm not too worried about it. I've seen many more games where the players wanted to play on after a flag fell, let their opponent take a move back, or make invalid moves.
Greg Miller