Author: Dieter Buerssner
Date: 13:28:24 04/22/05
Go up one level in this thread
On April 22, 2005 at 11:04:28, Uri Blass wrote:
>On April 22, 2005 at 09:29:13, Andrey Popov wrote:
>
>>This is the last position of the last game of Kasparov-Karpov match.
>>If Karpov (Black) reaches a draw, he wins the match.
>>Of course, White has a huge advantage.
>>However, I do not see any way to win.
>>5n1k/5Q2/4p1p1/2q1P2p/7P/6P1/5PK1/3B4 b - - 0 64
>>Why did Karpov resigned?
>>Can anybody see a series of White's moves leading to win after Qc5-b4?
>>Engines quickly go to score about +3.00 and stick on it forever.
>
>I could probe clearly higher score than +3 by yace after Qc5-b4
I assume you did this with backwards analysis. In this position, it is not even
needed - only moderate patience is needed. Yace, Centrino 1.7 GHz, 700 MB hash,
not TBs but 3/4 men bitbases + kppkp bitbases (no idea, if they make a
difference) shows +4 in around 3 minutes (in the analysis I show, it used about
90% of the CPU). In still reasonable time, it is over 5. The analysis also shows
some typical misbehaviour of Yace. Showing a fail high and the research just
settles at the border of the old window. In the past, I have tried to
investigate such cases (I suspected a bug), but I did find nothing. Perhaps it
is just because of the overloaded hash tables. It "smells" something real good,
but when researching with the more open window, the interesting HT entries are
gone already. Perhaps it is a real bug, which I am incapable to find. To my
liking, it happens too often. It might also be due to the fact, that HTs will
not handle repetitions correftly. Not storing draw scores will (of course) not
help. Typically, when I see this, I also see only short PVs (the engine tries to
reconstruct PVs from hash, always. At display time - not at search time)
132174421 1:48.7 2.71 15. 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qd8 5.Bxg6 Nxg6
6.Qxg6 Qxh4+ 7.Kg2 Qg4 8.Qxg4 hxg4 9.f4 Kh7
10.Kf2 Kh8 11.Ke3 Kh7 12.Ke4 Kg7 13.f5 Kf8
14.fxe6 {160}
189961947 2:37.4 3.11 16++ 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qd8 5.Bxg6 Nxg6
6.Qxg6 Qxh4+ 7.Kg2 Qg4 8.Qxg4 hxg4 9.f4 Kh7
10.Kf2 Kh8 11.Ke3 Kh7 12.Ke4 Kg7 13.f5 Kf8
14.f6 Kf7 {80}
237721325 3:17.2 4.06 16t 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qxe4 5.Qxf8+ Kh7
6.Qf7+ Kh6 7.g4H hxg4H 8.Qf8+H Kh7H 9.f3H gxf3H
10.h5H {HT} {-80}
267759954 3:42.3 4.06 16. 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qxe4 5.Qxf8+ Kh7
6.Qf7+ Kh6 7.g4 hxg4 8.Qf8+ Kh7 9.f3 gxf3 10.h5
{HT} {-80}
375396407 5:11.0 4.34 17t 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qa3 5.Bxg6 Nxg6
6.Qxg6 Qb3 7.Qxh5+ Kg7 8.Qg5+ Kf7 9.h5 Ke8H
10.Qg6+H Kd7H 11.Qf7+H Kc6H 12.h6H {HT} {240}
459512616 6:19.0 4.34 17. 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qa3 5.Bxg6 Nxg6
6.Qxg6 Qb3 7.Qxh5+ Kg7 8.Qg5+ Kf7 9.h5 Ke8
10.Qg6+ Kd7 11.Qf7+ Kc6 12.h6 Kd5H {240}
875037372 12:01.7 4.73 18t 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qa3 5.Bxg6 Qc5 6.Qf6+
Kg8 7.Bf7+H Kh7H 8.Bxh5H Qa7H 9.Qxf8H Qb6H {HT}
1126541386 15:26.3 4.73 18. 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qa3 5.Bxg6 Qc5 6.Qf6+
Kg8 7.Bf7+ Kh7 8.Bxh5 Qa7 9.Qxf8 Qb6
1476610424 20:09.4 5.13 19++ 2.Bc2 Qa3 3.Kh2 Qa8 4.Be4 Qa3 5.Bxg6 Qc5 6.Qf6+
Kg8 7.Bf7+ Kh7 8.Bxh5 Qa7 9.Qxf8 Qb6
1739679916 23:47.8 5.13 19t 2.Bc2 Qa3 3.Kh2 Qc5H 4.Bb1H Qc2H 5.Bxc2H Nd7H
{HT}
2401815394 32:55.2 5.13 19. 2.Bc2 Qa3 3.Kh2 Qc5 4.Bb1 Qc2 5.Bxc2 Nd7
2988481197 41:13.9 5.53 20++ 2.Bc2 Qa3 3.Kh2 Qc5 4.Bb1 Qc2 5.Bxc2 Nd7
3742374732 51:59.2 5.53 20t 2.Bc2 Qa3 3.Kh2 Qc5H 4.Be4H Qc1H 5.Qxf8+H Kh7H
{HT} {430}
1083244392 1:13:09 5.54 20t+ 2.Kh3 Qa3 3.Bc2 Qb4 4.Kh2 Qc5 5.Be4 Qb4 6.Bxg6
Nxg6 7.Qxg6H Qc4H {HT}
1411512008 1:17:27 5.54 20t 2.Kh3 Qa3 3.Bc2 Qb4H 4.Kh2H Qc5H 5.Be4H {HT}
3165768594 1:39:48 5.54 20. 2.Kh3 Qa3 3.Bc2 Qb4 4.Kh2 Qc5 5.Be4
Regards,
Dieter
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.