Author: Bernhard Bauer
Date: 04:28:33 09/15/98
Go up one level in this thread
On September 15, 1998 at 06:18:38, Djordje Vidanovic wrote:
>What Erkki claimed seemed sufficiently intriguing for me to immediately set up
>the position for Crafty 15_18 (AMD K6-2/300)and what I got does not confirm the
>bug. Here's the log:
>
>[Event "Computer chess game"]
>[Site "HAL"]
>[Date "1998.09.15"]
>[Round "-"]
>[White "wcrafty-15_18"]
>[Black "Djordje"]
>[Result "1-0"]
>[TimeControl "120+1"]
>[FEN "8/4k3/6KP/4N3/8/8/8/8 w - - 0 1"]
>[SetUp "1"]
>
>{--------------
>. . . . . . . .
>. . . . k . . .
>. . . . . . K P
>. . . . N . . .
>. . . . . . . .
>. . . . . . . .
>. . . . . . . .
>. . . . . . . .
>white to play
>--------------}
>1. Nc4 Ke8 2. Nd6+ Ke7 3. Nf5+ Ke8 4. h7 Kd7 5. h8=Q Kc6 6. Qc3+ Kd5 7. Kf6
>Ke4 8. Ke6 Kf4 9. Qg3+ Ke4 10. Qe3#
>{White mates} 1-0
>
>Crafty does go about blindly a bit (move ordering), but mates easily in the end.
>
>Regards,
>Djordje
>
>
>
>
>On September 15, 1998 at 05:56:02, Erkki Malkamaki wrote:
>
>>Wcrafty 15.18 without endgame tables can´t win as white from the following
>>position:
>>
>>White: Kg6,Ne5,ph6
>>Black: Ke7
>>
>>Crafty moves Knight or King aimlessly and "forgets" the pawn at h6.
>>Versions < = 15.17 move the pawn almost immediately.
>>
>>I have tested this with P90,P200MMX and AMD K6-2 300 with same results.
Here is what I get on a dual P6 233MHz without table bases:
9-> 12.65 16.37 1. Ng4 Kd6 2. Ne3 Kc5 3. Nf5 Kc4 4.
Ne7 Kc3 5. h7 Kd3
10 14.77 16.35 1. Ng4 Kd6 2. Ne3 Kc5 3. Nf5 Kc4 4.
Nd6+ Kc5 5. Ne8 Kd4 6. Ng7 Kd3 7. h7
Kc2 <HT>
10 40.88 ++ 1. h7!!
10-> 53.24 16.64 1. h77
11 1:02 ++ 1. h7!!
11 1:55 Mat08 1. h7 Kd6 2. h8=Q Ke6 3. Qe8+ Kd5 4.
Qb5+ Kd4 5. Qb2+ Ke3 6. Kf6 Ke4 7.
Qc3 Kf4 8. Qf3#
11-> 2:00 Mat08 1. h7 Kd6 2. h8=Q Ke6 3. Qe8+ Kd5 4.
Qb5+ Kd4 5. Qb2+ Ke3 6. Kf6 Ke4 7.
Qc3 Kf4 8. Qf3#
time: 2:01 cpu:202% mat:4 n:38180500 nps:315176
ext-> checks:6356378 recaps:0 pawns:142348 1rep:213275
predicted:0 nodes:38180500 evals:99116
endgame tablebase-> probes done: 0 successful: 0
hashing-> trans/ref:89% pawn:99% used:w61% b89%
mate in 8 moves
However, wcrafty-15.19-smp.exe crashes on WinNT4.0. The last 2 lines are
White(1): end-game phase
clearing hash tables
wcrafty-15.19-smp.exe
1. doesn't play under WinNT4.0 (crashes after some moves).
2. doesn't play after setup a position with setboard.
3. test and epdpfga work.
4. test may say "solution correct"
even if solution is not found.
Here an example:
======================================================================
! LCTPOS12 (...Rb6!) !
======================================================================
+---+---+---+---+---+---+---+---+
8 | | | | *R| | | *K| |
+---+---+---+---+---+---+---+---+
7 | *P| *P| | | *N| *P| *P| |
+---+---+---+---+---+---+---+---+
6 | | | *R| *Q| *P| | | *P|
+---+---+---+---+---+---+---+---+
5 | | | | | | | | |
+---+---+---+---+---+---+---+---+
4 | | | | P | Q | | | |
+---+---+---+---+---+---+---+---+
3 | | B | R | | | | P | |
+---+---+---+---+---+---+---+---+
2 | P | P | | | | P | | P |
+---+---+---+---+---+---+---+---+
1 | | | | R | | | K | |
+---+---+---+---+---+---+---+---+
a b c d e f g h
solution 1. Rb6
time surplus 0.00 time limit 30:00 (30:00)
depth time score variation (1)
11-> 3:07 0.14 1. ... Rxc3 2. bxc3 Nd5 3. Qf3 b5 4.
Bxd5 Qxd5 5. Qxd5 Rxd5 6. Kg2 Kf8 7.
Kf3 Rf5+ 8. Ke3 Ke7
12 4:08 0.14 1. ... Rxc3 2. bxc3 Nd5 3. Bxd5 exd5
4. Qe2 Rc8 5. Qb5 b6 6. Rd3 Qc6 7.
Qxc6 Rxc6 8. Re3 Kh7
12-> 5:52 0.14 1. ... Rxc3 2. bxc3 Nd5 3. Bxd5 exd5
4. Qe2 Rc8 5. Qb5 b6 6. Rd3 Qc6 7.
Qxc6 Rxc6 8. Re3 Kh7
13 8:00 0.13 1. ... Rxc3 2. bxc3 Nd5 3. Bxd5 Qxd5
4. Qxd5 Rxd5 5. Kg2 Kf8 6. Re1 Ke7
7. Kf3 b5 8. Ke4 Kd6 9. f4
time=30:00 cpu=199% mat=0 n=380004204 fh=92% nps=211005
ext-> checks=9013328 recaps=1471979 pawns=68235 1rep=3313840
predicted=0 nodes=380004204 evals=114261435
endgame tablebase-> probes done=0 successful=0
hashing-> trans/ref=22% pawn=9% used=w99% b99%
----------------------> solution correct (8/12).
Kind regards
Bernhard
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.