Author: Dann Corbit
Date: 18:38:03 03/09/01
Go up one level in this thread
On March 09, 2001 at 21:28:22, Dieter Buerssner wrote:
>> 8/1p6/5Pp1/7p/7k/3p4/7K/8 w - - 0 1
>>
>> Moravec, 1961
>
>I allready posted a followup to this one. For some reason, I could not believe
>the analysis of Yace. I recompiled, started with larger hash (20 MB) - and the
>unbelievalbe happened:
>
> nodes time score depth
>[many lines with 0.0 score]
> 9667999 1:13.6 0.00 20. 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qf6+ Qg5
> 5.Qd4+ Qg4 6.Qf6+ {-240}
> 10703164 1:21.1 0.40 21++ 1.f7 d2 2.f8=Q g5
> 19801708 2:22.7 7.00 21t 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 b4 8.Qe7+ Qg5 9.Qxb4+ Qg4
> 10.Qe7+ Qg5 11.Qe4+ Qg4 12.Qe3 Qe2+ 13.Qxe2 g5
> 14.Qf3 g4H 15.Qb7H g3+H 16.Kg2H {HT} {841}
> 20042337 2:24.5 7.00 21. 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 b4 8.Qe7+ Qg5 9.Qxb4+ Qg4
> 10.Qe7+ Qg5 11.Qe4+ Qg4 12.Qe3 Qe2+ 13.Qxe2 g5
> 14.Qf3 g4 15.Qb7 g3+ 16.Kg2 {HT} {841}
> 32871596 4:01.5 7.25 22t 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 b4 8.Qe7+ Qg5 9.Qxb4+ Qg4
> 10.Qe7+ Qg5 11.Qe4+ Qg4 12.Qe3 Qh3+ 13.Qxh3+
> Kg5 14.Qf3 h4 15.Kh3H {HT} {841}
> 33314343 4:03.9 7.25 22. 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 b4 8.Qe7+ Qg5 9.Qxb4+ Qg4
> 10.Qe7+ Qg5 11.Qe4+ Qg4 12.Qe3 Qh3+ 13.Qxh3+
> Kg5 14.Qf3 h4 15.Kh3 {HT} {841}
> 52448548 5:41.5 7.39 23t 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 b4 8.Qe7+ Qg5 9.Qxb4+ Qg4
> 10.Qe7+ Qg5 11.Qe4+ Qg4 12.Qe3 Qh3+ 13.Qxh3+
> Kg5 14.Kg3 Kf6 15.Kh4 {HT} {841}
> 53138865 5:45.3 7.39 23. 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 b4 8.Qe7+ Qg5 9.Qxb4+ Qg4
> 10.Qe7+ Qg5 11.Qe4+ Qg4 12.Qe3 Qh3+ 13.Qxh3+
> Kg5 14.Kg3 Kf6 15.Kh4 {HT} {841}
> 72398003 7:19.0 7.79 24++ 1.f7 d2 2.f8=Q d1=Q 3.Qf4+ Qg4 4.Qe3 b5 5.Qe7+
> Qg5 6.Qe4+ Qg4 7.Qe3 Qg2+ 8.Kxg2 Kg4
>
>I guess this will give me some work.
You must have made some improvements to search or eval recently. The 0.99
version won't find anything in 24 plies, even with 256 megs of hash.
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.