Stockfish

on Sat Sep 08, 2012 8:17 pm by

stockfish 12-09-02 x 64, se encuentra en la pagina anterior ,la 24.

on Sun Sep 09, 2012 4:01 pm by

Stockfish source - Sep 09, 2012

- Remove redundancy in move generation
- Simplify generate_castle()
- Rename *last to *end
- Enable link time optimization only when optimizing

https://github.com/mcostalba/Stockfish/commits/master

on Mon Sep 10, 2012 9:05 am by

Stockfish 120909 x64 SSE4.2
Download

on Mon Sep 10, 2012 9:32 am by

Stockfish 120909 x64
Download

on Tue Sep 11, 2012 1:16 am by

@Quocvuong82 wrote:Stockfish 120909 x64
Download

which is best Stockfish120902 or Stockfish120909? confused

on Tue Sep 11, 2012 2:48 am by

from Sep 02 to Sep 09
Sep 04, 2012

Code:: Get rid of struct Time We just need the milliseconds of current system time for our needs. This allows to simplify the API. No functional change.

Sep 06, 2012

Code:: Simplify Option c'tor No functional change.

Sep 07, 2012

Code:: Revert "Simplify Option c'tor" std::to_string() is C++11 material, not c++03. So revert the patch.

Sep 09, 2012

Code:: Enable link time optimization only when optimizing Because it is quite slow, skip it when 'optimize' flag is 'no' No functional change.

Code:: Rename *last to *end It is a more correct name because it points past the last move of the list. No functional change.

Code:: Simplify generate_castle() Skipping the calls to std::min(), std::man() we get even a nice speed-up on perft. No functional change.

Code:: Remove redundancy in move generation Introduce generate_all_moves() and remove a good bunch of redundant code. No functional change.

bench 128 1 16 default depth
Stockfish 120902 x64 SSE4.2

Code:: Total time (ms) : 37611 Nodes searched : 41791181 Nodes/second : 1111142

Stockfish 120909 x64 SSE4.2

Code:: Total time (ms) : 37697 Nodes searched : 41791181 Nodes/second : 1108607

Nodes searched : 41791181

on Thu Sep 13, 2012 6:38 am by

Stockfish source - Sep 11, 2012

- Speed-up generate

The trick here is to check for legality only in the
(rare) cases we have pinned pieces or a king move
or an en-passant.

This trick is able to increase the speed of perft
of more then 20%!

https://github.com/mcostalba/Stockfish/commit/e0bd0f250b861bfd4ee2999fc518eaf9b0a58941

on Sat Sep 15, 2012 5:08 am by

Stockfish 120915P SSE4.2 x64 by Peterpan
http://www.spaadyshare.com/AUM4HWIYZND6/StockFish_120915Px64.rar

on Sat Sep 15, 2012 6:01 am by

easl wrote:Stockfish 120915P SSE4.2 x64 by Peterpan
http://www.spaadyshare.com/AUM4HWIYZND6/StockFish_120915Px64.rar

Bad compile. I get random crashes (blue screen of death followed by a restart). cry

on Sat Sep 15, 2012 7:52 am by

Try this: mediafire.com ?5f3338ex17b3t9z

Stockfish 120909 - compiled by Quocvuong82, very strong and stable.

on Sat Sep 15, 2012 8:22 am by

Stockfish 2.3 update available

Stockfish 2.3 JA by the Stockfish Team

Changes:

ELO increase is very limited, it is mainly a minor release to flush
the accumulated work. From the user point of view the biggest thing is
that SF should not crash anymore even with many threads because a
nasty SMP bug causing a rare but repetitive crash has been fixed.

Marco.

Window/Linux/Android 32/64 bit

https://dl.dropbox.com/u/5047625/stockfish-23-ja.zip
Mirror:
http://cl.ly/JTpn/stockfish-23-ja.zip

on Sat Sep 15, 2012 8:54 am by

Stockfish 120915 x64 SSE4.2
Download

on Sat Sep 15, 2012 9:36 am by

JA GCC version is very fast! eek

on Sat Sep 15, 2012 10:04 am by

@Quocvuong82 wrote:Stockfish 120915 x64 SSE4.2
Download

Thanks! is there/will be a non SSE version too?

on Sat Sep 15, 2012 10:05 am by

Code:: Games Completed = 1000 of 1000 (Avg game length = 15.186 sec) Settings = RR/32MB/2000ms+100ms/M 700000cp for 7 moves, D 150 moves/EPD:E:\LittleBlitzer\openings.epd(4000) Time = 4009 sec elapsed, 0 sec remaining 1. Stockfish 2.3 JA 64bit 515.0/1000 267-237-496 (L: m=237 t=0 i=0 a=0) (D: r=380 i=67 f=26 s=6 a=17) (tpm=113.3 d=15.57 nps=1348166) 2. Stockfish 2.2.2 JA 64bit 485.0/1000 237-267-496 (L: m=267 t=0 i=0 a=0) (D: r=380 i=67 f=26 s=6 a=17) (tpm=113.6 d=15.29 nps=1373859)

on Sat Sep 15, 2012 12:14 pm by

Bob wrote:

@Quocvuong82 wrote:Stockfish 120915 x64 SSE4.2
Download

Thanks! is there/will be a non SSE version too?

Please wait!

on Sat Sep 15, 2012 12:15 pm by

Code:: Games Completed = 200 of 2000 (Avg game length = 13.473 sec) Settings = Gauntlet/64MB/2000ms+100ms/M 500cp for 5 moves, D 150 moves/EPD:openings.epd(4000) Time = 2852 sec elapsed, 25664 sec remaining 1. Stockfish 120915Q x64 SSE4.2 106.0/200 51-39-110 (L: m=0 t=0 i=0 a=39) (D: r=91 i=13 f=2 s=1 a=3) (tpm=117.0 d=14.69 nps=1780924) 2. Stockfish 23GCCJA x64 SSE4.2 47.5/100 18-23-59 (L: m=0 t=0 i=0 a=23) (D: r=48 i=5 f=2 s=1 a=3) (tpm=116.5 d=14.98 nps=1836041) 3. Stockfish 120902Q x64 SSE4.2 46.5/100 21-28-51 (L: m=0 t=0 i=0 a=28) (D: r=43 i=8 f=0 s=0 a=0) (tpm=118.8 d=14.52 nps=1840633)

on Sat Sep 15, 2012 3:48 pm by

Marco Costalba wrote:Just a couple of quick notes.

1) For people compiling themself the Stockfish bench signature is: 5416292 It means that if you run ./stockfish bench then it should show 5416292 nodes searched. This means that the compile is correct.

2) For people running / using under Android / iPhone with more than one thread, I'd suggest to set "Use Sleeping Threads" option. This allows to save power and battery and strength is practically the same.

on Sun Sep 16, 2012 2:23 am by

Just saw that Stockfish 2.3 is out. Mainly maintenance realease, but I cant help thinking its a bit faster also smile

http://stockfishchess.org/download/

on Sun Sep 16, 2012 10:12 am by

@Quocvuong82 wrote:

Bob wrote:

@Quocvuong82 wrote:Stockfish 120915 x64 SSE4.2
Download

Thanks! is there/will be a non SSE version too?

Please wait!

on Sun Sep 16, 2012 12:02 pm by

Bob wrote:

@Quocvuong82 wrote:

Bob wrote:

@Quocvuong82 wrote:Stockfish 120915 x64 SSE4.2
Download

Thanks! is there/will be a non SSE version too?

Please wait!

Sorry Bob!
Stockfish 2.3Q x64

on Sun Sep 16, 2012 1:57 pm by

Code:: Games Completed = 1000 of 1000 (Avg game length = 14.933 sec) Settings = RR/32MB/2000ms+100ms/M 700000cp for 7 moves, D 150 moves/EPD:E:\LittleBlitzer\openings.epd(4000) Time = 3977 sec elapsed, 0 sec remaining 1. Stockfish 2.3 JA 64bit 510.0/1000 246-226-528 (L: m=226 t=0 i=0 a=0) (D: r=378 i=104 f=25 s=3 a=18) (tpm=113.4 d=15.45 nps=1299907) 2. Stockfish 2.3Q 64bit 490.0/1000 226-246-528 (L: m=246 t=0 i=0 a=0) (D: r=378 i=104 f=25 s=3 a=18) (tpm=113.4 d=15.37 nps=1272451)

Stockfish 2.3 JA : nps=1299907
Stockfish 2.3 Q : nps=1272451

on Mon Sep 17, 2012 11:49 am by

Stockfish 'Barracuda' 2.3 32-bit
http://www.computerchess.info/tdbb/phpBB3/viewtopic.php?f=9&t=619

Download

on Mon Sep 17, 2012 1:15 pm by

(B=Barracuda)
Stockfish 2.3B x64 SSE4.2

bench 128 1 16 default depth:

Code:: Total time (ms) : 25714 Nodes searched : 29223250 Nodes/second : 1136472

on Mon Sep 17, 2012 1:25 pm by

Code:: Games Completed = 1000 of 1000 (Avg game length = 14.979 sec) Settings = RR/32MB/2000ms+100ms/M 700000cp for 7 moves, D 150 moves/EPD:E:\LittleBlitzer\openings.epd(4000) Time = 3911 sec elapsed, 0 sec remaining 1. Stockfish 2.3 JA 736.0/1000 574-102-324 (L: m=102 t=0 i=0 a=0) (D: r=237 i=59 f=18 s=0 a=10) (tpm=113.7 d=14.96 nps=969211) 2. Stockfish 'Barracuda'2.3 264.0/1000 102-574-324 (L: m=567 t=7 i=0 a=0) (D: r=237 i=59 f=18 s=0 a=10) (tpm=114.5 d=14.89 nps=917210)

Stockfish 2.3 JA : nps=969211

Stockfish 'Barracuda'2.3 : nps=917210 (very weak)