Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Diep 17 ply +0.49

Author: Vincent Diepeveen

Date: 20:50:04 08/17/00

Go up one level in this thread


On August 17, 2000 at 21:37:24, Paulo Soares wrote:

>On August 17, 2000 at 21:03:58, Vincent Diepeveen wrote:
>
>>On August 17, 2000 at 10:00:37, Vivek Rao wrote:
>>
>>>[D]2nq2rb/3pr2k/1pp1p2p/p3Pp1B/3P1P2/PPN5/5Q1P/R5RK w
>>>
>>>The Solution is Ne4!>>The Winning Line for White is;
>>>1. Ne4            fxe4
>>>2. f5             Qf8
>>>3. f6             Reg7
>>>4. Rxg7+          Rxg7
>>>5. Qh4            Rg5
>>>6. Qxe4+          Kg8
>>>7. h4             Bxf6
>>>8. hxg5           Bxg5
>>>9. Qg6+           Kh8
>>>10. Qe8           Qxe8

at the screen i see the line till here now.
so a 20 ply line. at ply==19. exactly as the above 10 moves are.

Bxe8 is probably overwritten in hashtable. I assume Ne7 and Bxd7 also.

score at 19 ply is dissappointing though +0.26

850M nodes needed in a bit more as 3 hours.
>>>11. Bxe8          Ne7
>>>12. Bxd7
>>
>>At 17 ply DIEP finally sees white is winning here. Incredible!
>>Pawn less in a nearly closed position and winning by a "here goes my
>>knight" move.
>>
>>339M nodes +0.49  PV 17 ply is now the first 16 plies of the above
>>variation. The plies after that probably fell away because of
>>hash overwrites.
>>
>>This is my first test at home with 200MB hash and i like it!
>>
>>At WMCC i will play with a bit less hash as i lose 20mb memory for tables
>>to 5 men then. 180mb or so hash then, still not bad!
>
>And what is the main line(pv)?
>Thanks,
>Paulo Soares



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.