Stockfish 2.3

Discussion about chess-playing software (engines, hosts, opening books, platforms, etc...)
Post Reply
User avatar
noctiferus
Posts: 122
Joined: Thu Jun 10, 2010 7:57 am
Location: Ivrea (To), Italy

Stockfish 2.3

Post by noctiferus » Sat Sep 15, 2012 9:07 pm


User avatar
Carlos Ylich
Posts: 56
Joined: Thu Jun 10, 2010 3:04 am
Location: São Paulo

Re: Stockfish 2.3

Post by Carlos Ylich » Sun Sep 16, 2012 1:10 am

thanks for the news!
Congratulations to the Stockfish team :!:

User avatar
noctiferus
Posts: 122
Joined: Thu Jun 10, 2010 7:57 am
Location: Ivrea (To), Italy

Re: Stockfish 2.3

Post by noctiferus » Sun Sep 16, 2012 1:26 pm

A bug found ( limited to chess960). Wait a few days for a new compile. I hope to be able to advise you as soon as published.

ThinkingALot
Posts: 144
Joined: Sun Jun 13, 2010 7:32 am
Contact:

Re: Stockfish 2.3

Post by ThinkingALot » Sun Sep 16, 2012 2:01 pm

SSE42 compile appears to be slower than the regular x64 on 2600K@4.5Ghz HT on:
32-bit: 1563kns
32-bit legacy: 1509kns
64-bit: 2091kns
64-bit SSE42: 1718kns (!!!)
64-bit GCC: 2142kns.
Node count is 5416292 in all cases.
I ran the benchmark several times for each build. The results do not change significantly.

User avatar
noctiferus
Posts: 122
Joined: Thu Jun 10, 2010 7:57 am
Location: Ivrea (To), Italy

Re: Stockfish 2.3

Post by noctiferus » Sun Sep 16, 2012 3:23 pm

Marco Costalba, Jim Ablett and others are discussing this issue here:
http://talkchess.com/forum/viewtopic.php?t=45163

Post Reply