Winboard Forum

On last Sunday I was playing a game in an internet tournament for blind players. Time control was 30 moves for 1 hour, and the moves were announced by voice.

Here is the game.

Code: Select all

[Event "For-the-people September-October swiss"]
[Site "?"]
[Date "2011.09.18"]
[Round "?"]
[White "Deville, Olivier"]
[Black "Ikramy, Ahmed"]
[Result "0-1"]
[ECO "E16"]
[PlyCount "62"]
[EventDate "2011.??.??"]

1. Nf3 Nf6 2. c4 e6 3. d4 Bb4+ 4. Nbd2 b6 5. g3 Bb7 6. Bg2 O-O 7. O-O d5 8.
cxd5 exd5 9. Ne5 Nbd7 10. Ndf3 Nxe5 11. Nxe5 Re8 12. Bf4 Bd6 13. Nd3 c5 14.
dxc5 bxc5 15. Bxd6 Qxd6 16. Rc1 c4 17. Nf4 g5 18. Nh3 Qe5 19. e3 Rab8 20. Qc2
Ba6 21. Rb1 Bc8 22. Qc3 Bf5 23. Rbd1 Rxb2 24. Qxe5 Rxe5 25. Nxg5 Bd3 26. Nf3
Rh5 27. Rfe1 Ne4 28. Bf1 Rf5 29. Bxd3 cxd3 30. Nh4 Rfxf2 31. Rxd3 Ng5 0-1


Tournament Rules say :

Code: Select all

19. Your moves should be the product of your own brain.
Help from a computer or other person is cheating and is prohibited.
Such help during an adjournment is also prohibited.
Make your relatively obvious moves quickly.
That way, time is saved, and you will not be suspected of
getting help from a computer.


Now when I check this game with a chess engine, I discover that Fruit 2.1 finds all black's moves from move 11 to move 31. I assume an opening book was used to generate moves 1 to 10.

Do you notice the same on your computer? And, if this is the case, do you think there is a probability that an amateur player can find exactly the same moves as Fruit does?

The tournament director did not acknoledge the cheating, and gave me a loss.

My opponent denies he cheated. He is a member of this forum and can answer this thread if he wants to.

You might dislike the way I am handling this issue. But I am strongly against cheating. I have exposed cloners more than once in the past. And we are on-topic since a chess engine was used (or so am I claiming).

Olivier

I can confirm that Fruit 2.1 predicts all the black moves from move 11 on.

Olivier Deville wrote:
Do you notice the same on your computer? And, if this is the case, do you think there is a probability that an amateur player can find exactly the same moves as Fruit does?

The probability of matching 20 not forced moves of Fruit 2.1 in a row is computable in 0

You might dislike the way I am handling this issue. But I am strongly against cheating. I have exposed cloners more than once in the past. And we are on-topic since a chess engine was used (or so am I claiming).

Olivier

Tolerance 0 to cheating

Another thread has been started on TalkChess : http://www.talkchess.com/forum/viewtopi ... 42&t=40490

Though I'd rather have comments posted on this forum, since I'm not sure my opponent has an account at TalkChess. Here at least he can answer if he feels like it.

Olivier

My opponent withdrew from all Skype-like chess groups for the blind.

This is the end of the story, as far as I am concerned.

Olivier

Moves 14-15, 18, 21-24 and 29-30 are forced. Moves 25 and 27 are obvious. Moves 20 and 31 are obvious once you spot them. Moves 12 and 13 are thematic (it's just a question of whether to further prepare c7-c5). Only one other move made sense on moves 16 and 28. Black's play is consistent with an aggressive style.

I could have been black in this game.

Teemu Pudas wrote:Moves 14-15, 18, 21-24 and 29-30 are forced. Moves 25 and 27 are obvious. Moves 20 and 31 are obvious once you spot them. Moves 12 and 13 are thematic (it's just a question of whether to further prepare c7-c5). Only one other move made sense on moves 16 and 28. Black's play is consistent with an aggressive style.

I could have been black in this game.

Yes you could, maybe I could too, or anyone playing random moves too (blindofld too?) ... but you could play other some millions of possible combinations too, even taking in account that there are forced moves.

The question is that Fruit 2.1 plays exactly the same sequence of moves, 20 in total. Other strong engines do not (Try Rybka 4 f.i, or Dirty 25AUG). Olivier himself, as white (and being a strong blind player) played few of the moves predicted by Fruit for white side.

EDIT : Fruit 2.1 is included in the WB JAWS package that black player was using....

There is also the context.

Here is what I wrote in my claim report to the TD :

Code: Select all

(...)I started to become suspicious when Ramy asked me to repeat my moves twice because Jaws was running in the background. Why would you need Jaws running all the time when you are playing a game of chess?(...)


...and here is what my opponent wrote when I accused him of cheating :

Code: Select all

(...)Olivier, i honestly did not cheat you. my God! i even did not use my
computer in our game(...)


Of course swearing does not help

If you take these data in account, the probability of guilt is close to 100%. Of course I have no problem if you think otherwise. The case is over anyway, my opponent left the field. If he is reading this, I just want to tell him that if you cheat, you get caught sooner or later, and I hope he will benefit from this painful experiment.

Olivier

In your initial post you wrote: "Do you notice the same on your computer? And, if this is the case, do you think there is a probability that an amateur player can find exactly the same moves as Fruit does?"
This suggested that you were interested in a discussion of the case. (I even had a look at the game to make some comments.) But now it seems that you are not interested in a rational discussion at all, but rather in slinging mud at your opponent, who may or may not have cheated.
Now that the topic is in the open, it should be fairly and openly discussed.
By the way, what is Jaws? The movie?

@Teemu: I agree with much of what you said.
However, I do not understand why you say that move 21 is forced. On the contrary, I think it is rather suspicious, unless black is a very strong player. 21...Bc8 would be quite natural (again, for a strong player), if there wasn't the very human and greedy alternative ...c3. The whole maneuver Rab8, Ba6, Bc8 leaves a very strong impression.
Also, you did not mention black's 26th move, Rh5. This move looks rather unnatural - what does the rook do on h5 - unless you see that after 27. Re1 Ne4, 28.Bf1 is the most tenacious defense, after which black now has 28...Rf5. A strong player can of course find Rh5, but it is far from trivial - I think most people would play Re8 without thinking (which wins as well).
Nevertheless, I totally agree that it is perfectly possible that a decent player finds all of these moves on his own.

Kappatoo wrote:In your initial post you wrote: "Do you notice the same on your computer? And, if this is the case, do you think there is a probability that an amateur player can find exactly the same moves as Fruit does?"
This suggested that you were interested in a discussion of the case. (I even had a look at the game to make some comments.) But now it seems that you are not interested in a rational discussion at all, but rather in slinging mud at your opponent, who may or may not have cheated.
Now that the topic is in the open, it should be fairly and openly discussed.
By the way, what is Jaws? The movie?

OK I take this back. If you want to discuss the topic, it is OK for me indeed.

Jaws is a screen reader used by blind people. You can get the Winboard for Jaws package here :

viewtopic.php?f=19&t=51528&start=1

Fruit 2.1 is part of the package.

Winboard now also works with NVDA, a free alternative to Jaws :

viewtopic.php?f=19&t=51961

My opponent posted in this thread.

Olivier

Kappatoo wrote:@Teemu: I agree with much of what you said.
However, I do not understand why you say that move 21 is forced. On the contrary, I think it is rather suspicious, unless black is a very strong player. 21...Bc8 would be quite natural (again, for a strong player), if there wasn't the very human and greedy alternative ...c3. The whole maneuver Rab8, Ba6, Bc8 leaves a very strong impression.
Also, you did not mention black's 26th move, Rh5. This move looks rather unnatural - what does the rook do on h5 - unless you see that after 27. Re1 Ne4, 28.Bf1 is the most tenacious defense, after which black now has 28...Rf5. A strong player can of course find Rh5, but it is far from trivial - I think most people would play Re8 without thinking (which wins as well).
Nevertheless, I totally agree that it is perfectly possible that a decent player finds all of these moves on his own.

For your information, Rh5 was played a tempo.

Olivier

Okay, thanks.
Like I said in my response to Teemu, I agree with him that many of those moves are very natural. Chess is anything but a random game, so you cannot just compute the probability of someone's move choice matching that of an engine. As it happens, I once played a game in which only one or two of the moves I played did not coincide with Rybka's preference (I do not recall which version, though). If you are interested, here is the game:
1. Nf3 g6 2. d4 Bg7 3. c4 Nf6 4. Nc3 O-O 5. e4 d6 6. Be2 e5 7. O-O Nc6 8. dxe5
dxe5 9. Bg5 Qxd1 10. Rfxd1 Bg4 11. h3 Bxf3 12. Bxf3 Nd4 13. Nd5 Nxd5 14. cxd5
c6 15. dxc6 bxc6 16. Rdc1 Rfb8 17. b3 a5 18. Rc4 Rb4 19. Rac1
Bf8 20. Bd1 Rab8 21. Be3 Rxc4 22. Rxc4 Rb4 23. Rc1 Bd6 24. Bd2 Rb8 25. Bxa5 Ra8
26. b4 Be7 27. a4 f6 28. f4 c5 29. fxe5 fxe5 30. Bc7 Rxa4 31. Bxa4 Ne2+ 32. Kf1
1-0
I was familiar with the variation up to move 13. You can check it for yourself.
I think I played a decent game here, but really nothing sensational. Sometimes, it just happens that there is a series of natural moves which are also best. (Needless to say, in other games, I play blunder after blunder.)

That said, black's play in your game was indeed very strong, and at least some of his moves raise suspicion. Do you have any information on your opponent's strength?

Kappatoo wrote:Okay, thanks.
Like I said in my response to Teemu, I agree with him that many of those moves are very natural. Chess is anything but a random game, so you cannot just compute the probability of someone's move choice matching that of an engine. As it happens, I once played a game in which only one or two of the moves I played did not coincide with Rybka's preference (I do not recall which version, though). If you are interested, here is the game:
1. Nf3 g6 2. d4 Bg7 3. c4 Nf6 4. Nc3 O-O 5. e4 d6 6. Be2 e5 7. O-O Nc6 8. dxe5
dxe5 9. Bg5 Qxd1 10. Rfxd1 Bg4 11. h3 Bxf3 12. Bxf3 Nd4 13. Nd5 Nxd5 14. cxd5
c6 15. dxc6 bxc6 16. Rdc1 Rfb8 17. b3 a5 18. Rc4 Rb4 19. Rac1
Bf8 20. Bd1 Rab8 21. Be3 Rxc4 22. Rxc4 Rb4 23. Rc1 Bd6 24. Bd2 Rb8 25. Bxa5 Ra8
26. b4 Be7 27. a4 f6 28. f4 c5 29. fxe5 fxe5 30. Bc7 Rxa4 31. Bxa4 Ne2+ 32. Kf1
1-0
I was familiar with the variation up to move 13. You can check it for yourself.
I think I played a decent game here, but really nothing sensational. Sometimes, it just happens that there is a series of natural moves which are also best. (Needless to say, in other games, I play blunder after blunder.)

That said, black's play in your game was indeed very strong, and at least some of his moves raise suspicion. Do you have any information on your opponent's strength?

Thanks, I reckon you genuinely want to discuss the topic in an open mind.

I don't know my opponent's rating, he is not in the FIDE list. He told me he came 3rd in this years' overall Egyptian championship, but gave no link to the standings.

He can post in this thread if he wants to, he is not banned from here or whatever.

The strongest blind player at the moment (and probably the strongest of all times) is IM Sergey Krylov, currently rated 2352.

Olivier

Did he play any other games in that tournament?
Of course it would be good if your opponent responded here. I nevertheless think that to say that the probability of his guilt is close to 100% is vastly exaggerated. I agree that it is clearly greater than 50%, though.
(What is the probability that I cheated in the game I cited above?)

Kappatoo wrote:Did he play any other games in that tournament?
Of course it would be good if your opponent responded here. I nevertheless think that to say that the probability of his guilt is close to 100% is vastly exaggerated. I agree that it is clearly greater than 50%, though.
(What is the probability that I cheated in the game I cited above?)

Here is another game he played in the French league for blind people.

Code: Select all

[Event "Tournoi Skype 007"]
[Site "Skype"]
[Date "????.??.??"]
[Round "1.6"]
[White "Gerbaux, Regis"]
[Black "Ikramy, Ahmed"]
[Result "0-1"]
[ECO "C68"]
[WhiteElo "1370"]
[BlackElo "1799"]
[PlyCount "58"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Bxc6 dxc6 5. O-O f6 6. c3 Qd3 7. Re1 Bc5 8.
Nxe5 fxe5 9. Qh5+ Kf8 10. Qxe5 Bb6 11. Qf4+ Ke8 12. Qe5+ Ne7 13. Qg3 Qxg3 14.
hxg3 Rf8 15. Rf1 c5 16. c4 Nc6 17. Nc3 Ne5 18. d3 Nxd3 19. b3 Ba5 20. Nd5 c6
21. Bd2 cxd5 22. Bxa5 dxe4 23. Rad1 b5 24. f3 Bb7 25. Rd2 bxc4 26. bxc4 Ne5 27.
fxe4 Rxf1+ 28. Kxf1 Nxc4 29. Rd3 Nxa5 0-1


1799 was a provisional rating I gave him for making the pairings.

Moves from 6...Qd3 to 13...Qxg3 are Fruit moves. One might say he let play the computer until he got a winning position, and then felt secure enough to play himself. but of course this is only speculative. Let's add several moves were obvious ones. So this game alone proves nothing.

I am not saying he is cheating all the time. Possibly he is capable of playing good chess. I strongly believe he cheated against me.

I checked the results of the Egyptian championships on the FIDE website. He is not there.

Olivier

Ahmed, the owner of http://www.chess2u.com/ who is egyptian, made a quick inquiry.

Code: Select all

 [Olivier] -  any news?
 [Ahmed] -   I talked by phone with friends in Chess Federation
 [Ahmed] -  and no one know this guy
 [Ahmed] -  he is lying
 [Olivier] -  ok thanks
 [Olivier] -  do you allow me to forward this to the winboard forum thread?
 [Ahmed] -  Of course
 [Olivier] -  thanks again for your help
 [Ahmed] -  you are welcome
 [Ahmed] -  Top Ten player ( 2011 )
 [Ahmed] -  http://chess-results.com/tnr53811.aspx?lan=1


Olivier