Search found 51 matches

by Ross Boyd
Sun Sep 20, 2009 10:35 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Null move pruning
Replies: 32
Views: 18635

Re: Null move pruning

There are two well known solutions to safegard against bad zugzwang searches. There is verified null move search (by Omid David) where you go ahead and do a quick search to refute the move, and double null move technique (by Vincent Diepveen) which says that you can never do two null moves in a row...
by Ross Boyd
Fri Aug 07, 2009 10:46 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: CCC Practice Game
Replies: 1150
Views: 100314

Re: Move 28..

<lurking mode off>

A sacrificial move like ...Rf3 or even ...Bf3 (at the right moment and with the right preparation) is possibly a gift that the R30 could not easily refuse...

Something to think about....

<lurking mode on>
by Ross Boyd
Wed Jan 07, 2009 8:55 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Repetition Detection
Replies: 28
Views: 8864

Re: Repetition Detection

Thanks, I understand that. I don't have a 50 move counter at present, but may make one at another time. Searching backwards does seem a little inefficient, cause with your idea, you have to search 50 positions at each node in the tree. I think a improvement would be too have another value attached ...
by Ross Boyd
Fri Apr 18, 2008 3:28 am
Forum: Computer Chess Club: General Topics
Topic: Dead End for Computer Chess? Some ideas...
Replies: 26
Views: 6909

Re: Dead End for Computer Chess? Some ideas...

I must admit that whilst I'm not much of a chess player (maybe 1600 ELO, who knows?)... I certainly enjoy playing vicariously through my engine TRACE. So, chess programming is a way of playing chess by proxy - through a silicon agent. And I study chess, and observe thousands of games, and have becom...
by Ross Boyd
Fri Apr 18, 2008 12:57 am
Forum: Computer Chess Club: General Topics
Topic: To authors: Have you ever beaten your program?
Replies: 4
Views: 1441

Re: To authors: Have you ever beaten your program?

I managed a draw once by avoiding tactics and simplifying to a R and P ending - but - to actually win, its just not going to happen.
by Ross Boyd
Fri Apr 11, 2008 8:51 am
Forum: Computer Chess Club: General Topics
Topic: Test Position: Bd3!
Replies: 11
Views: 2880

Re: Test Position: Bd3!

Ah yes, TRACE 2 solves this, no problem... It just takes 19 minutes, that's all.. Old TRACE 1.37a doesn't see it even after 90 minutes.... FEN: 2r1nrk1/2qbbpp1/p3p2p/1p1pP1N1/3B1P2/6Q1/PPP1B1PP/3R1R1K w - - 0 18 TRACE 2.00: 1 00:00.156 462 0 -1.19 Nf3 2 00:00.156 660 0 -1.19 Nf3 Qxc2 3 00:00.156 1.1...
by Ross Boyd
Wed Apr 09, 2008 9:08 pm
Forum: Computer Chess Club: General Topics
Topic: Medium difficulty testmove 17.Qe3
Replies: 17
Views: 4858

Re: Medium difficulty testmove 17.Qe3

Build 70: tries to reign in extensions in QSearch but is not successful enough, extensions only allowed if depth > -1 instead of depth > 0, still QSearch explodes on ply 9, does find Ne3 at this depth what Build 69 does not. Changes in search_full.cpp: By using depth > -1 instead of depth > 0 you w...
by Ross Boyd
Wed Apr 09, 2008 12:22 pm
Forum: Computer Chess Club: General Topics
Topic: Test Position: Nxg2 followed by d4!
Replies: 5
Views: 1527

Re: Test Position: Nxg2 followed by d4!

I see that you got Trace 2 almost ready... I wish, I wish... She usually solves faster than 1.37a but she doesn't play well. And the search has some 'sticking' points which I haven't solved yet. Time is very limited here... doing some studies as well as full-time work.... so don't get excited... ju...
by Ross Boyd
Wed Apr 09, 2008 12:13 pm
Forum: Computer Chess Club: General Topics
Topic: Test Position: Bh6!!!
Replies: 17
Views: 5126

Re: Test Position: Bh6!!!

TRACE 2 needs 3 minutes 41 secs on a Q6600 (single threaded) FEN: 2r3k1/1pqnbppp/p2pb1P1/4p2P/2r1P3/1NN1BP2/PPPQ4/2K3RR w - - 0 18 TRACE 2.00: 1/2 00:00.156 1.281 0 -1.11 gxf7+ Bxf7 2/11 00:00.156 1.570 0 -0.80 gxf7+ Bxf7 Kb1 2/6 00:00.156 2.028 0 -0.71 Kb1 fxg6 3/12 00:00.156 2.980 0 -0.71 Kb1 fxg6...
by Ross Boyd
Wed Apr 09, 2008 12:04 pm
Forum: Computer Chess Club: General Topics
Topic: Test Position: Nxg2 followed by d4!
Replies: 5
Views: 1527

Re: Test Position: Nxg2 followed by d4!

Yes... TRACE 2 needs about 35 seconds on a Q6600 (single threaded) FEN: r2qr1k1/p4pbp/bp4p1/3p4/8/BPN1n1P1/P2QN1BP/R2R2K1 b - - 0 18 TRACE 2.00: 1/4 00:00.157 2.537 0 +0.70 Nxd1 2/11 00:00.157 2.716 0 +0.70 Nxd1 Rxd1 3/12 00:00.157 6.306 5.000 +0.28 Nxd1 Rxd1 Bxe2 3/13 00:00.157 8.363 5.000 +0.56 Bx...