Fruit does not see draw - and resigns!

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

Moderator: Ras

ASchwartmann

Fruit does not see draw - and resigns!

Post by ASchwartmann »

Take a look at

http://www.andreas-schwartmann.de/ccrl1/

Fruit does not see the draw in this position - and resigns!

Strange, isn't it?
Last edited by ASchwartmann on Fri Aug 03, 2007 3:06 pm, edited 1 time in total.
User avatar
mclane
Posts: 18901
Joined: Thu Mar 09, 2006 6:40 pm
Location: US of Europe, germany
Full name: Thorsten Czub

Re: Fruit does not see draw - and resigns!

Post by mclane »

sometimes search, especially null move search cannot replace knowledge.

CSTAL DOS sees this after you set up the positions WIHTOUT using the search, just with the static evaluations.
User avatar
M ANSARI
Posts: 3721
Joined: Thu Mar 16, 2006 7:10 pm

Re: Fruit does not see draw - and resigns!

Post by M ANSARI »

I think many engines have problems with wrong bishop position. I remember Shredder showing a +9 score in a similar position. Obviously some code needs to be added for such endings. Not sure if Rybka has that code ... would be interesting to see how Rybka evaluates this.
Tord Romstad
Posts: 1808
Joined: Wed Mar 08, 2006 9:19 pm
Location: Oslo, Norway

Re: Fruit does not see draw - and resigns!

Post by Tord Romstad »

[d]k7/P4p2/P7/1KB2p2/8/8/8/8 b - -
ASchwartmann wrote:Fruit does not see the draw in this position - and resigns!

Strange, isn't it?
Not really. Drawn endgames with a doubled rook pawn and a bishop of the wrong color are extremely rare in practice. Implementing specific evaluation rules for all sorts of obscure endgames is boring and tedious, and does not really help playing strength. Most programmers tend to concentrate on more general chess knowledge, which can be applied in a wider variety of positions.

My program actually does evaluate the given position as a draw, but I doubt that it has any effect on playing strength whatsoever. It might save half a point once every 10000 games or so, but on the other hand it also makes the program a tiny bit slower in some endgames, which probably costs a tiny fraction of an Elo point. The net effect is totally insignificant. The only reason I have implemented this evaluation rule at all is that I hate to see my program lose positions which even I know how to draw.

Tord
User avatar
Roman Hartmann
Posts: 295
Joined: Wed Mar 08, 2006 8:29 pm

Re: Fruit does not see draw - and resigns!

Post by Roman Hartmann »

ASchwartmann wrote:Take a look at

http://www.andreas-schwartmann.de/ccrl1/

Fruit does not see the draw in this position - and resigns!

Strange, isn't it?
Fruit didn't resign but rather the GUI (Fruit cannot resign as it's an UCI-engine). Always a bit dangerous to let the GUI decide.

Roman
BBauer
Posts: 658
Joined: Wed Mar 08, 2006 8:58 pm

Re: Fruit does not see draw - and resigns!

Post by BBauer »

No need to add knowledge or in some other cases stupitidy.
Perhaps implemented knowlege causes such a hiegh score.
It would be enough not to resign even if the evaluation is -10 :roll:
regards
Bernhard
Uri Blass
Posts: 10822
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Fruit does not see draw - and resigns!

Post by Uri Blass »

Fruit knowsand even old versions know

New game
k7/P4p2/P7/1KB2p2/8/8/8/8 b - - 0 1

Analysis by Fruit 2.0:

1...f5-f4
= (0.00) Depth: 1/1 00:00:00
1...f5-f4 2.Kb5-a4
= (0.00) Depth: 2/2 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3
= (0.00) Depth: 3/5 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3
= (0.00) Depth: 4/5 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2
= (0.00) Depth: 5/8 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2
= (0.00) Depth: 6/9 00:00:00 4kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6
= (0.00) Depth: 7/10 00:00:00 9kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4
= (0.00) Depth: 8/11 00:00:00 25kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5
= (0.00) Depth: 9/14 00:00:01 54kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3
= (0.00) Depth: 10/15 00:00:01 117kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4
= (0.00) Depth: 11/15 00:00:01 222kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4
= (0.00) Depth: 12/18 00:00:01 414kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3
= (0.00) Depth: 13/20 00:00:01 720kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 14/22 00:00:01 1193kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 15/22 00:00:02 1882kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 16/22 00:00:02 2817kN

(, 03.08.2007)


New game
k7/P4p2/P7/1KB2p2/8/8/8/8 b - - 0 1

Analysis by Fruit 2.2:

1...f5-f4
= (0.00) Depth: 1/1 00:00:00
1...f5-f4 2.Kb5-a4
= (0.00) Depth: 2/2 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3
= (0.00) Depth: 3/5 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3
= (0.00) Depth: 4/5 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2
= (0.00) Depth: 5/8 00:00:00
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2
= (0.00) Depth: 6/9 00:00:00 4kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6
= (0.00) Depth: 7/10 00:00:00 9kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4
= (0.00) Depth: 8/11 00:00:00 24kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5
= (0.00) Depth: 9/14 00:00:01 51kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3
= (0.00) Depth: 10/15 00:00:01 112kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4
= (0.00) Depth: 11/15 00:00:01 208kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4
= (0.00) Depth: 12/18 00:00:01 380kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3
= (0.00) Depth: 13/20 00:00:01 650kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 14/20 00:00:01 1022kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 15/22 00:00:02 1293kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 16/24 00:00:02 1811kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 17/26 00:00:02 2515kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-a3 f5-f4 7.Ka3-a4 f4-f3 8.Ka4-a3
= (0.00) Depth: 18/26 00:00:02 3119kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8
= (0.00) Depth: 19/28 00:00:02 3623kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4
= (0.00) Depth: 20/28 00:00:03 4575kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7
= (0.00) Depth: 21/28 00:00:03 5265kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3
= (0.00) Depth: 22/32 00:00:03 6292kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8
= (0.00) Depth: 23/32 00:00:03 7150kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8 14.Ka3-a4
= (0.00) Depth: 24/32 00:00:04 8410kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8 14.Ka3-a4
= (0.00) Depth: 25/34 00:00:04 9675kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8 14.Ka3-a2 Ka8-b7 15.Ka2-b1
= (0.00) Depth: 26/37 00:00:05 11589kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8 14.Ka3-a2 Ka8-b7 15.Ka2-b1 Kb7-a8
= (0.00) Depth: 27/37 00:00:06 13196kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8 14.Ka3-a2 Ka8-b7 15.Ka2-b1 Kb7-a8 16.Kb1-a1
= (0.00) Depth: 28/44 00:00:07 15243kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-a3 Kb7-a8 14.Ka3-a2 Ka8-b7 15.Ka2-b1 Kb7-a8 16.Kb1-a1 Ka8-b7
= (0.00) Depth: 29/44 00:00:08 17502kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-b5 Kb7-a8 14.Kb5-a4
= (0.00) Depth: 30/44 00:00:09 19715kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-b5 Kb7-a8 14.Kb5-a4
= (0.00) Depth: 31/49 00:00:10 22480kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-b5 Kb7-a8 14.Kb5-a4
= (0.00) Depth: 32/49 00:00:11 25724kN
1...f5-f4 2.Kb5-a4 f4-f3 3.Ka4-a3 f3-f2 4.Bc5xf2 f7-f6 5.Ka3-a4 f6-f5 6.Ka4-b5 f5-f4 7.Kb5-b6 f4-f3 8.Kb6-c5 Ka8xa7 9.Kc5-b5+ Ka7-a8 10.a6-a7 Ka8-b7 11.Kb5-a5 Kb7-a8 12.Ka5-a4 Ka8-b7 13.Ka4-b5 Kb7-a8 14.Kb5-a4
= (0.00) Depth: 33/54 00:00:13 29347kN

(, 03.08.2007)
Last edited by Uri Blass on Fri Aug 03, 2007 4:07 pm, edited 1 time in total.
User avatar
Matthias Gemuh
Posts: 3245
Joined: Thu Mar 09, 2006 9:10 am

Re: Fruit does not see draw - and resigns!

Post by Matthias Gemuh »

ASchwartmann wrote:Take a look at

http://www.andreas-schwartmann.de/ccrl1/

Fruit does not see the draw in this position - and resigns!

Strange, isn't it?


Yes, very strange indeed. I mean the resign.
AFAIK, the UCI protocol has no resign. Is Fruit breaking the protocol ?

Matthias.
My engine was quite strong till I added knowledge to it.
http://www.chess.hylogic.de
Uri Blass
Posts: 10822
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Fruit does not see draw - and resigns!

Post by Uri Blass »

My fruit can see the draw easily


,a - Fruit 2.3, Friend mode
k7/3r1p2/PP6/1KB2p2/8/8/8/8 b - - 0 1

Analysis by Fruit 2.3.1:

55...f5-f4 56.Kb5-c4
= (-0.18) Depth: 1 00:00:11
55...f5-f4 56.Kb5-c4 f4-f3
= (-0.24) Depth: 2 00:00:11
55...f5-f4 56.Kb5-c6 Rd7-a7 57.b6-b7+ Ra7xb7 58.a6xb7+ Ka8-b8
+- (2.89) Depth: 3 00:00:11
55...Rd7-a7 56.Bc5-d6 Ra7-d7 57.Kb5-c6
² (0.46) Depth: 3 00:00:12
55...Rd7-a7 56.Bc5-d6 f7-f6 57.Bd6-f4 Ra7-d7
² (0.41) Depth: 4 00:00:12
55...Rd7-a7 56.Bc5-d6 f7-f6 57.Bd6-c7 f5-f4 58.Bc7xf4
± (1.17) Depth: 6 00:00:12
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 Ra7-d7 58.Kb5-c5 f7-f5
± (1.02) Depth: 6 00:00:12 7kN
55...Rd7-a7 56.Bc5-d6 f7-f6 57.Bd6-c7 f5-f4 58.Bc7xf4 Ra7-d7 59.Kb5-c5
± (0.95) Depth: 7 00:00:13 10kN, tb=60
55...Rd7-a7 56.Bc5-d6 f7-f6 57.Bd6-c7 f5-f4 58.Bc7xf4 Ra7-d7 59.Kb5-c5 f6-f5
± (1.02) Depth: 8 00:00:13 13kN, tb=78
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.Be5-d4
= (0.19) Depth: 9 00:00:13 24kN, tb=84
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 10 00:00:14 38kN, tb=125
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 11 00:00:14 59kN, tb=273
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 12 00:00:14 82kN, tb=296
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 13 00:00:14 114kN, tb=322
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 14 00:00:14 170kN, tb=325
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 15 00:00:14 265kN, tb=334
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 16 00:00:15 417kN, tb=350
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 17 00:00:15 608kN, tb=460
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 18 00:00:16 834kN, tb=557
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 19 00:00:17 1145kN, tb=668
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 20 00:00:18 1726kN, tb=2102
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 21 00:00:20 2213kN, tb=2455
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 22 00:00:22 2918kN, tb=3673
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 23 00:00:24 3694kN, tb=5939
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 24 00:00:27 4747kN, tb=9097
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 25 00:00:29 5934kN, tb=15116
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 26 00:00:33 7470kN, tb=20009
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 27 00:00:36 9430kN, tb=26581
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 28 00:00:40 11582kN, tb=35767
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 29 00:00:45 14327kN, tb=47466
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 30 00:00:49 17716kN, tb=61272
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 31 00:00:56 22615kN, tb=72223
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 32 00:01:05 27846kN, tb=95541
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 33 00:01:13 33508kN, tb=110479
55...Rd7-a7 56.Bc5-d6 f5-f4 57.Bd6xf4 f7-f5 58.Bf4-c7 f5-f4 59.Bc7-e5 f4-f3 60.b6-b7+ Ra7xb7+ 61.a6xb7+
= (0.00) Depth: 34 00:01:26 40829kN, tb=137770

(, 03.08.2007)
User avatar
mclane
Posts: 18901
Joined: Thu Mar 09, 2006 6:40 pm
Location: US of Europe, germany
Full name: Thorsten Czub

Re: Fruit does not see draw - and resigns!

Post by mclane »

i have to disagree - strongly !!

[Event "15th WMCC"]
[Site "Paris"]
[Date "1997.10.30"]
[Round "8"]
[White "Chess System Tal / P2-300"]
[Black "Dark Thought / Alpha 767"]
[Result "1/2-1/2"]
[ECO "B01"]
[PlyCount "210"]
[EventDate "1997.10.26"]

1. e4 d5 2. exd5 Qxd5 3. Nc3 Qa5 4. d4 Nf6 5. Nf3 c6 6. Bc4 Bg4 7. h3 Bh5 8.
Bd2 e6 9. Qe2 Bb4 10. g4 Bg6 11. O-O-O Nbd7 12. Kb1 O-O-O 13. a3 Bxc3 14. Bxc3
Qc7 15. Bd2 Nb6 16. Ba5 c5 17. Rhe1 cxd4 18. Ba2 Qf4 19. Bxb6 axb6 20. Ne5 Kb8
21. Qb5 Rd6 22. Qb4 Rhd8 23. f3 Nd5 24. Bxd5 Rxd5 25. Nxg6 hxg6 26. Qxb6 g5 27.
Rd3 Qh2 28. Red1 e5 29. Rb3 R5d7 30. Qa5 f6 31. Rd2 Qh1+ 32. Ka2 Rc8 33. Qa4
Rdc7 34. Qb4 Qf1 35. Qd6 Ka8 36. Qb6 Qa6 37. Qxa6+ bxa6 38. Kb1 Ka7 39. Kc1 Rh8
40. Rh2 a5 41. Kd2 Rhc8 42. Kd3 Ka6 43. a4 Rc4 44. Ra3 Rb4 45. c3 Rb7 46. Rc2
Rd8 47. cxd4 Rxd4+ 48. Ke3 Rdb4 49. Rc6+ Ka7 50. Rc5 Kb6 51. Rc2 Rd7 52. Ke2
Rc7 53. Rd2 Rc1 54. Rad3 Rxa4 55. Rd6+ Kb7 56. Rd7+ Rc7 57. R7d5 Ra1 58. Ke3 a4
59. Rb5+ Kc6 60. Rc2+ Kd7 61. Rd5+ Kc8 62. Rcd2 Rf1 63. Rd8+ Kb7 64. R8d5 Kb6
65. Rd6+ Kc5 66. R2d5+ Kc4 67. Ra5 Kb4 68. Rda6 Ra1 69. Ra8 Rb7 70. R5a7 Rxa7
71. Rxa7 Kb3 72. Rb7+ Kc2 73. Rb4 Ra2 74. Ke4 Kb1 75. Kf5 Rxb2 76. Rxa4 Rh2 77.
Ra7 Rxh3 78. Ke4 Rh7 79. Kd3 Kb2 80. Rc7 Kb3 81. Rb7+ Ka3 82. Rb1 Rh2 83. Rb7
Rh3 84. Ke4 Rh7 85. Rb1 Rh2 86. Rb7 Re2+ 87. Kd3 Re1 88. Rxg7 Rd1+ 89. Ke2 Rd6
90. Rb7 Rd4 91. Rb6 Rf4 92. Ke3 Ka4 93. Rb1 Ka5 94. Rb2 Rd4 95. Rb7 Ka6 96. Rb8
Rf4 97. Rb1 Rc4 98. Rb8 Rd4 99. Rb2 Ka7 100. Rb5 Rc4 101. Kd3 Ra4 102. Ke3 Rf4
103. Rb2 Rc4 104. Kd3 Rd4+ 105. Ke3 Rf4 1/2-1/2


CSTAL knew in the evaluations that a king, separated from the other pieces, cannot make the win. it knew the king should be CUT from the other pieces, and so it did. Dark Thought on DEC Alpha500 searched
much deeper, as Ernst A. Heinz did tell us almost each move in a very proud tone: "it is in ply 18 !" but it did not understand that you need the
King to bring your advantage home.



[d] 8/R5pr/5p2/4p1p1/6P1/3K1P2/1k6/8 w - - 42 80

here CSTAL decides that the King should be cut from the other pieces.
not using the search but using the evals that this is important strategy in the endgame.


[d] 8/R5pr/5p2/4p1p1/6P1/3K1P2/1k6/8 w - - 42 80

few moves later again:

[d] 8/1R4pr/5p2/4p1p1/6P1/k2K1P2/8/8 w - - 46 82



Ernst continued the game because in HIS main lines white allowed the black King to move to the center , to help the rook and the pawns.

His main lines showed this WIN because he had no knowledge that the King cannot win if he is cut.

in move 105 he - with a very depressed face - agreed on draw.

what do you want with all the plies AHEAD and your good , better and superior hardware if the search computes into an area of the tree that will never happen because the opponent does not allow you to make those moves ?