CCT 11: Games so far...

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: CCT 11: Games so far...

Post by bob »

Damir wrote:Hi Peter

Should not this tournament be run more professionally, as I understand that Rybka crashed and lost the game in won position, without beeing able to resume the game.....
This kind of thing has been going on for quite some time now where losers were getting rewarded for another engine's bad luck.
As a TD it should not be impossible for you to change the outcome of the result....
Regards
Damir
When an engine _resigns_ the game is over, whether you like the result or not does not matter. ICC considers the game over. Engines lose because of bugs all the time. And it has nothing to do with ICC.
krazyken

Re: CCT 11: Games so far...

Post by krazyken »

this game was a lot of fun!

[Event "ICS Unrated Standard match"]
[Site "chessclub.com"]
[Date "2009.03.21"]
[Round "-"]
[White "zct"]
[Black "KtuluCCT11"]
[Result "0-1"]
[WhiteElo "2237"]
[BlackElo "2000"]
[TimeControl "3000+3"]

1. c4 (0:00) g5 (0:00) 2. Nc3 (1:47) Bg7 (4:11) 3. d4 (1:23) h6 (0:00) 4.
f4 (1:21) c5 (0:58) 5. dxc5 (0:20) gxf4 (1:29) 6. Bxf4 (1:18) Bxc3+ (0:21)
7. bxc3 (1:16) Qa5 (2:48) 8. Qd2 (0:00) Nf6 (0:45) 9. Qe3 (1:26)
{+0.04/16 1:25} Nc6 (1:40) 10. Nf3 (0:00) {+0.33/18} Ng4 (2:13) 11.
Qd2 (0:00) {+0.40/19} Nf6 (0:54) 12. Nd4 (1:12) {+0.72/17 1:11} Ne4 (1:57)
13. Nxc6 (1:09) {+0.50/17 1:09} dxc6 (0:37) 14. Qd4 (0:54) {+0.11/17 53}
f6 (1:25) 15. g3 (0:00) {-0.06/17} Qxc3+ (1:17) 16. Qxc3 (0:52)
{-0.04/18 52} Nxc3 (1:02) 17. Bd2 (0:00) {+0.03/19} Na4 (0:40) 18.
Bg2 (0:10) {+0.13/17 10} Nxc5 (0:24) 19. O-O (0:51) {+0.19/17 50}
Bg4 (0:47) 20. Rfb1 (0:49) {+0.01/17 49} O-O-O (0:30) 21. Be3 (0:17)
{+0.20/17 17} Nd7 (0:52) 22. Rb2 (0:00) {+0.09/18} Ne5 (1:09) 23.
Bxa7 (0:00) {+0.00/18} Rd7 (0:50) 24. c5 (0:48) {+0.04/16 48} Rhd8 (0:32)
25. h3 (0:26) {-0.44/18 25} Nc4 (0:09) 26. Rc2 (0:46) {-0.22/19 46}
Be6 (0:07) 27. Kf2 (0:45) {-0.25/18 44} Rd4 (0:09) 28. Rac1 (0:44)
{-0.12/16 43} Ne5 (0:26) 29. Bb6 (0:43) {-0.01/17 42} R8d7 (0:01) 30.
Rc3 (0:41) {-0.38/17 41} Ra4 (2:16) 31. a3 (0:00) {-0.48/18} Nc4 (1:57) 32.
g4 (0:00) {-0.53/19} Rd2 (0:27) 33. Bf3 (0:41) {-0.72/17 40} Ra2 (0:34) 34.
Rd3 (1:09) {-1.40/18 1:08} Nxb6 (0:03) 35. cxb6 (0:35) {-1.16/17 34}
R4xa3 (0:22) 36. Bxc6 (0:14) {-1.44/18 14} bxc6 (0:27) 37. Rxc6+ (0:09)
{-1.63/20 9} Kb7 (0:13) 38. Rxa3 (0:23) {-1.70/20 22} Kxc6 (0:10) 39.
Re3 (0:26) {-1.71/21 25} Kd7 (0:30) 40. Rd3+ (0:05) {-1.83/21 5} Ke8 (0:56)
41. Re3 (0:00) {-1.92/21} Bf7 (0:41) 42. Rb3 (0:35) {-2.11/18 35}
Ra8 (0:22) 43. Rb4 (0:12) {-2.29/17 12} Kd7 (0:42) 44. Kg1 (0:00)
{-2.36/16} Rb8 (0:45) 45. Kf2 (0:34) {-2.91/17 34} Kc6 (0:37) 46.
Re4 (0:00) {-2.87/17} e5 (0:28) 47. Ra4 (0:33) {-3.15/18 33} Kxb6 (0:24)
48. Ra1 (0:08) {-3.14/18 8} Bc4 (0:27) 49. Rb1+ (0:32) {-3.04/18 32}
Kc7 (0:13) 50. Rc1 (0:31) {-3.26/18 31} Rb4 (0:10) 51. Kg3 (0:20)
{-3.35/17 20} Kd6 (0:16) 52. e3 (0:14) {-3.57/17 13} Kd5 (2:04) 53.
Kf3 (0:00) {-4.34/21} Rb2 (0:12) 54. Rd1+ (0:17) {-4.34/19 17} Ke6 (0:23)
55. Kg3 (0:07) {-4.33/18 6} Rb3 (0:21) 56. Re1 (0:30) {-3.96/19 29}
f5 (0:22) 57. Kh4 (0:06) {-4.08/18 6} fxg4 (0:07) 58. hxg4 (0:29)
{-4.50/17 28} Kf6 (0:12) 59. Kh5 (0:49) {-5.32/19 48} Be6 (0:19) 60.
Rf1+ (0:27) {-5.48/19 26} Kg7 (0:15) 61. Re1 (0:11) {-5.56/21 10}
Rb2 (0:03) 62. Kh4 (0:26) {-5.60/20 25} Rg2 (0:24) 63. Kh3 (0:25)
{-5.61/17 25} Rxg4 (0:47) 64. Kh2 (0:00) {-5.79/19} Kf6 (0:37) 65.
Re2 (0:25) {-6.16/17 24} h5 (0:15) 66. Re1 (0:24) {-6.40/17 24} Bf5 (0:10)
67. Ra1 (0:13) {-6.37/17 13} Re4 (0:12) 68. Ra6+ (0:11) {-6.93/17 10}
Be6 (0:08) 69. Ra3 (0:29) {-7.14/17 28} h4 (0:13) 70. Kg1 (0:28)
{-8.31/17 27} Rg4+ (0:12) 71. Kh2 (0:23) {-8.48/18 23} Rb4 (0:07) 72.
Kg1 (0:21) {-8.28/16 21} Rb1+ (0:11) 73. Kf2 (0:34) {-10.89/20 34}
h3 (0:10) 74. Kg3 (0:20) {-10.60/19 19} Rg1+ (0:16) 75. Kh2 (0:17)
{-11.74/20 17} Rg2+ (0:11) 76. Kh1 (0:56) {-327.42/21 55} Re2 (0:08) 77.
Ra6 (0:17) {-13.23/16 17} Re1+ (0:09) 78. Kh2 (0:21) {-12.77/15 21}
Rxe3 (0:32) 79. Kg1 (0:00) {-13.82/14} Re1+ (0:11) 80. Kh2 (0:05)
{-14.26/15 5} Re2+ (0:10) 81. Kg1 (0:14) {-16.66/14 14} e4 (0:07) 82.
Ra5 (0:25) {-327.46/16 24} Rg2+ (0:53) 83. Kh1 (0:15) {-327.46/14 15}
Rc2 (0:42) 84. Rh5 (0:15) {-327.50/16 15} Kg6 (0:33) 85. Rh8 (0:15)
{-327.50/16 14} e3 (0:00) 86. Kg1 (0:14) {-327.52/19 14} h2+ (0:00) 87.
Rxh2 (0:14) {-327.52/14 14} Rxh2 (0:14) 88. Kxh2 (0:00) {-327.54/15}
e2 (0:25) 89. Kg2 (0:00) {-327.56/52} e1=Q (0:00) 90. Kf3 (0:07)
{-327.58/128 7} Kg5 (0:01) 91. Kg2 (0:06) {-327.60/128 6} Bd5+ (0:01) 92.
Kh2 (0:00) {-327.62/128} Qf2+ (0:21) 93. Kh3 (0:00) Be6# (0:04)
{zct checkmated} 0-1
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: CCT 11: Games so far...

Post by bob »

Damir wrote:If the game was unable to be resumed because Rybka operator got the error in his engine file, and was forced to resign the game, why not play the game from the beginning, instead of awarding the losing side by giving it a win... is totally uncalled for.
Regards
Damir
Where do you come up with these nonsensical suggestions? Do we then delay the next round by 2 hours so this game can be replayed? When an engine loses because of a bug in the evaluation, even though it is at +6, do we have to replay those games as well?

In CCT (and even in WCCC) events, "you run what you brung". Hopefully what you "brung" has been tested enough that it doesn't crash 2-3 times per game as Rybka is currently doing in each round. A lack of testing/debugging on the part of one team does not constitute a reason to replay every game that engine loses due to it crashing... Otherwise a single round would never end, since some sort of problem in an engine causes it to lose every game it loses, and since both sides can't win, one side always has a problem that needs attention.
bob
Posts: 20943
Joined: Mon Feb 27, 2006 7:30 pm
Location: Birmingham, AL

Re: CCT 11: Games so far...

Post by bob »

Damir wrote:If the game was unable to be resumed because Rybka operator got the error in his engine file, and was forced to resign the game, why not play the game from the beginning, instead of awarding the losing side by giving it a win... is totally uncalled for.
Regards
Damir
Where do you come up with these nonsensical suggestions? Do we then delay the next round by 2 hours so this game can be replayed? When an engine loses because of a bug in the evaluation, even though it is at +6, do we have to replay those games as well?

In CCT (and even in WCCC) events, "you run what you brung". Hopefully what you "brung" has been tested enough that it doesn't crash 2-3 times per game as Rybka is currently doing in each round. A lack of testing/debugging on the part of one team does not constitute a reason to replay every game that engine loses due to it crashing... Otherwise a single round would never end, since some sort of problem in an engine causes it to lose every game it loses, and since both sides can't win, one side always has a problem that needs attention.
glorfindel

Re: CCT 11: Games so far...

Post by glorfindel »

Damir wrote:You can create a nick there for 30 days, after that it will expire, and you would no longer be able to use it unless you renew it by buying a serial from Chessbase...
What are you talking about? The server being proprietary means that the protocol is undisclosed and you can only connect to it using chessbase's client. The client is only supported in one OS architecture. I have tried, but I never succeeded in playing a complete game at playchess, even as a guest.

In simple words: some people don't use Windows.
krazyken

Re: CCT 11: Games so far...

Post by krazyken »

This one ZCT got backed into a corner and din't make it out in time.

[Event "ICS unrated standard match"]
[Site "chessclub.com"]
[Date "2009.03.21"]
[Round "-"]
[White "zct"]
[Black "Buzz"]
[Result "0-1"]
[WhiteElo "2237"]
[BlackElo "2829"]
[TimeControl "3000+3"]

1. d4 (0:00) Nf6 (0:00) 2. c4 (0:00) e6 (0:00) 3. Nf3 (0:00) b6 (0:00) 4.
g3 (0:00) Bb7 (0:00) 5. Bg2 (0:00) Bb4+ (0:00) 6. Bd2 (0:00) Bxd2+ (0:00)
7. Qxd2 (0:00) O-O (0:00) 8. Nc3 (0:00) d6 (0:00) 9. O-O (2:31)
{+0.81/16 2:31} Ne4 (0:00) 10. Nxe4 (1:22) {+0.62/17 1:22} Bxe4 (0:00) 11.
Rac1 (1:20) {+0.54/16 1:19} Nd7 (1:44) 12. Qe3 (1:18) {+0.45/16 1:18}
Nf6 (1:41) 13. Rfd1 (0:00) {+0.48/16} Qe8 (1:37) 14. g4 (1:16)
{+0.57/16 1:16} h6 (1:34) 15. b3 (1:33) {+0.97/16 1:32} Qc6 (1:31) 16.
g5 (1:11) {+0.94/17 1:11} hxg5 (0:17) 17. Qxg5 (0:53) {+0.97/17 52}
Rad8 (0:35) 18. Kh1 (1:25) {+1.51/16 1:25} Nh7 (1:27) 19. Qh4 (1:06)
{+1.42/17 1:06} Qb7 (1:24) 20. Rg1 (1:04) {+1.47/16 1:04} c5 (0:17) 21.
dxc5 (1:14) {+1.39/17 1:14} dxc5 (1:21) 22. Ng5 (0:00) {+1.20/18}
Nxg5 (1:18) 23. Qxg5 (0:00) {+1.13/19} f6 (1:15) 24. Qg4 (0:01) {+0.99/18}
f5 (1:15) 25. Qg5 (1:16) {+1.33/17 1:15} Rd4 (1:11) 26. f3 (0:00)
{+1.10/17} Bc6 (1:09) 27. Rcd1 (0:04) {+1.14/16 4} e5 (1:02) 28.
Rdf1 (0:59) {+1.06/16 58} Rf6 (1:04) 29. Bh3 (0:57) {+1.22/15 57}
Qf7 (0:05) 30. Rg3 (0:50) {+0.98/16 50} Rdd6 (1:02) 31. Rfg1 (1:08)
{+1.00/17 1:07} g6 (1:00) 32. Bf1 (0:00) {+0.91/17} Qg7 (0:58) 33.
Rh3 (0:53) {+0.83/16 52} Rfe6 (0:04) 34. a3 (0:52) {+0.80/17 51} e4 (0:56)
35. Bg2 (0:50) {+0.79/16 49} e3 (0:55) 36. Rh6 (1:01) {+1.03/15 1:00}
Qf6 (0:53) 37. Qf4 (0:47) {+0.90/15 47} Kg7 (0:04) 38. Rh3 (0:41)
{+0.75/16 41} Rd4 (0:51) 39. Qh6+ (0:00) {+0.57/17} Kf7 (0:49) 40.
Re1 (0:00) {+0.51/16} f4 (0:48) 41. Rh4 (0:56) {+0.71/17 56} g5 (0:00) 42.
Qxf6+ (1:16) {+0.30/19 1:15} Kxf6 (0:00) 43. Rh6+ (0:41) {+0.10/19 41}
Ke5 (0:47) 44. Rh5 (0:00) {+0.00/20} Rg6 (0:45) 45. h3 (0:00) {+0.00/19}
a5 (0:44) 46. a4 (0:51) {-0.42/18 50} Bd7 (0:43) 47. h4 (0:35)
{-1.26/19 35} Bf5 (0:41) 48. Rh8 (0:00) {-1.06/18} Bc2 (0:40) 49. h5 (0:38)
{-1.50/16 38} Rgd6 (0:02) 50. Re8+ (1:03) {-2.13/18 1:03} Kf6 (0:39) 51.
Bh3 (0:00) {-2.23/18} Bxb3 (0:38) 52. h6 (0:43) {-2.83/16 43} Kg6 (0:36)
53. Re7 (0:00) {-2.99/16} Kxh6 (0:35) 54. Bf5 (0:00) {-3.21/14} g4 (0:34)
55. Kg2 (0:01) {-3.67/13 0.2} Kg5 (0:34) 56. fxg4 (0:52) {-3.39/14 52}
Bxc4 (0:00) 57. Rg7+ (0:33) {-3.71/14 33} Kf6 (0:00) 58. Rg6+ (0:32)
{-4.18/15 32} Ke5 (0:32) 59. Rg5 (0:48) {-3.62/14 47} Bd5+ (0:00) 60.
Kf1 (1:46) {-6.29/17 1:46} f3 (0:32) 61. Bc2+ (0:00) {-6.39/15} Kf4 (0:31)
62. Rf5+ (0:59) {-8.98/17 58} Kg3 (0:00) 63. exf3 (0:26) {-9.20/16 25}
Rd2 (0:04) 64. Rxd5 (1:16) {-327.46/17 1:16} Rf2+ (0:00) 65. Kg1 (0:23)
{-327.48/22 23} Rxd5 (0:06) 66. Rxe3 (0:16) {-327.50/22 16} Rg2+ (0:13) 67.
Kh1 (0:09) {-327.52/25 8} Rxc2 (0:29) 68. f4+ (0:00) {-327.54/30}
Kf2 (0:28) 69. Kh2 (0:00) {-327.56/40} Kxe3+ (0:00) 70. Kg3 (0:16)
{-327.58/128 16} Rd1 (0:00) 71. f5 (0:06) {-327.60/128 5} Rh1 (0:00) 72.
g5 (0:00) {-327.62/128} Rhh2 (0:00) 73. f6 (0:00) {-327.64/128}
Rcg2# (0:00)
{zct checkmated} 0-1
User avatar
Zach Wegner
Posts: 1922
Joined: Thu Mar 09, 2006 12:51 am
Location: Earth

Re: CCT 11: Games so far...

Post by Zach Wegner »

Tough one. Well played by Dirty. ZCT definitely needs some work in the eval department. And SMP department too...
Nick C

Re: CCT 11: Games so far...

Post by Nick C »

Hi Damir, it's Nick Carlin. I understand that you and also a lot of good Rybka supporters are very upset and I apologize, I can't tell you how bad I felt.

If I may explain, I normally use Winboard on ICC and am very familiar with it, now when you get an engine crash in Winboard you can just hit the X button (top right) to close the Winboard GUI and you will be disconnected from ICC without resigning the game, it will just be adjourned for later continuation.

Now this time around I am using the Chess Assistant GUI, because it now can support Aquarium books, and I am trying out that format, because the book making tools are very good there. During the game we got an engine crash, when that happens you have to shutdown and reload your GUI asap. to get your engine back. So my Winboard mentality kicked in and I clicked on the X top right, this actually caused a pop-up window saying something like "Do you really want to do this and RESIGN (in upper case letters) the game" .. I didn't read it and just hit enter, so the CA GUI happily sent a resign to ICC.

Now, the ICC doesn't have the ability to resume resigned games, if it did I am sure we would not be having this discussion, as when I resigned we had a TOTALLY AND ABSOLUTELY WON position. It was because of this fact that I felt an award of 1/2-1/2 would have been more appropriate, it would have penalized the team that made the error, whilst at the same time not awarded a full point to an entrant who completely lost the game on the board.

But the decision was made and we must stand by it. There is no point in criticizing the TD, he did what he thought correct and generally I think Peter is an excellent and fair TD, I made the mistake, and unfortunately all Rybka fans must pay for it so my apologies.

We'll do our best tomorrow, chin up!
User avatar
Graham Banks
Posts: 41455
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: CCT 11: Games so far...

Post by Graham Banks »

Nick C wrote:I think Peter is an excellent and fair TD
And with a lot of experience too.
Such a situation as occurred puts a TD in a no win situation (excuse the pun). He'd have been criticised no matter what decision was made.
gbanksnz at gmail.com
User avatar
hgm
Posts: 27809
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: CCT 11: Games so far...

Post by hgm »

Nick C wrote:..., if it did I am sure we would not be having this discussion,
I am sure we would. This is not a technical problem, it would be no problem at all to resume a game from where you resigned by simply pasting in the FEN of the position where you resigned in a loadable game, and play from there.

Point is that the FIDE rules, by which we are playing, specify that a game is finished after you resign. That you did not intend to resign is immaterial, just as it is immaterial in any games you win that the opponent did not intend to be checkmated.

CCT is not a contest between programs, it is a match between teams consisting of machines, programs, programmers and operators. If one team includes a blundering operator that cost them the game, too bad for them. Apparently the operator is the Achilles heel of the Rybka team. Other teams have to deal with, e.g., far inferior hardware. We don't 'correct' any wins of Rybka against those teams to 1/2-1/2 results because they did not intend their hardware to be inferior either.