Testing Stockfish 11-03-13. 480 Games.

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
Ajedrecista
Posts: 2122
Joined: Wed Jul 13, 2011 9:04 pm
Location: Madrid, Spain.

Timestamps of Stockfish.

Post by Ajedrecista »

Hello:

I took a look on this post and following ones:

http://immortalchess.net/forum/showthre ... post380626

Unfortunately, some people do not know the meaning of timestamps. Let me clarify their meaning: these timestamps are 'seconds since 1970'. I did a quick search on Google and here is one of those timestamps converters/calculators:

UTC calculator.

So timestamp = 1368993649 means May 19th, 2013 at 20:00:49 GMT. It is a way to record the time of the compilation.

@George and Tom: thank you very much for your frequent tests. :)

Regards from Spain.

Ajedrecista.
Top
Tomcass
Posts: 786
Joined: Sun Apr 16, 2006 9:09 pm

Re: Timestamps of Stockfish.

Post by Tomcass »

Ajedrecista wrote:Hello:

I took a look on this post and following ones:

http://immortalchess.net/forum/showthre ... post380626

Unfortunately, some people do not know the meaning of timestamps. Let me clarify their meaning: these timestamps are 'seconds since 1970'. I did a quick search on Google and here is one of those timestamps converters/calculators:

UTC calculator.

So timestamp = 1368993649 means May 19th, 2013 at 20:00:49 GMT. It is a way to record the time of the compilation.

@George and Tom: thank you very much for your frequent tests. :)

Regards from Spain.

Ajedrecista.
Woooow, Jesús! This is a very surprising -for me- answer. Thank you very much for clarifying it for us. :-)

Kind regards from Barcelona.

Tom.
Top
ernest
Posts: 2053
Joined: Wed Mar 08, 2006 8:30 pm

Re: Timestamps of Stockfish.

Post by ernest »

Ajedrecista wrote:So timestamp = 1368993649 means May 19th, 2013 at 20:00:49 GMT.
...and indeed it says in http://abrok.eu/stockfish/ :

Date: Sun May 19 22:00:49 2013 +0200
Timestamp: 1368993649

Thanks, Jesus!
Top
Tomcass
Posts: 786
Joined: Sun Apr 16, 2006 9:09 pm

Re: Timestamps of Stockfish.

Post by Tomcass »

STOCKFISH 23-05-13 : 480 GAMES

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 4 min +0 sec/ game
No tablebases

SF230513 64x6 - Critter 1.6 64-bitx6_nob 9-24-7 21.0-19.0 52.50%
SF230513 64x6 - Deep Rybka 4.1 SSE42 x64 (x6) 13-24-3 25.0-15.0 62.50%
SF230513 64x6 - Houdini 3 Pro x64 8-19-13 17.5-22.5 43.75%
I7 980 3.33 Ghz.

6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 2 min +2 sec/ game
No tablebases

SF230513 64x6 - Critter 1.6 64-bitx6_nob 7-24-9 19.0-21.0 47.50%
SF230513 64x6 - Deep Rybka 4.1 SSE42 x64 (x6) 14-20-6 24.0-16.0 60.00%
SF230513 64x6 - Houdini 3 Pro x64_ 4-25-11 16.5-23.5 41.25%

240 games x6 = http://www.mediafire.com/?qz6353amz69y677

Overall average with 6 cores: (123.0-116.0) = 51.25%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 4 min + 0 sec/ game
Ponder: Off
No tablebases

SF230513 64 - Critter 1.6a 64-bitnob 9-26-5 22.0-18.0 55.00%
SF230513 64 - Deep Rybka 4 x64 (x4 13-23-4 24.5-15.5 61.25%
SF230513 64 - Houdini 3 Pro x64_ 6-17-17 14.5-25.5 36.25%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 2 min + 2 sec/ game
Ponder: Off
No tablebases

SF230513 64 - Critter 1.6a 64-bitnob_4 9-22-9 20.0-20.0 50.00%
SF230513 64 - Deep Rybka 4 x64 14-22-4 25.0-15.0 62.50%
SF230513 64 - Houdini 3 Pro x64_ 8-26-6 21.0-19.0 52.50%

240 games X4 = http://www.mediafire.com/?sjcskut0zz4w5jl

Overall average with 4 cores: (127.0-113.0) = 52.92%

Global Score: 250.0-230.0 (52.08%)

Against Critter 1.6: 51.25% Deep Rybka 4: 61.56% Houdini 3.0: 43.44%

Estimated ELO Performance: 3.109

Another excellent score for Stockfish.

Enjoy your Sunday!

Tom.
Top
Tomcass
Posts: 786
Joined: Sun Apr 16, 2006 9:09 pm

Re: Timestamps of Stockfish.

Post by Tomcass »

STOCKFISH 27-05-13 : 480 GAMES

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 4 min +0 sec/ game
No tablebases

201305_SF27052013_4.0

Stockfish 270513 64 SSE4.2x6 - Critter 1.6 64-bitx6_nob 19.0 - 21.0 +9/=20/-11 47.50%
Stockfish 270513 64 SSE4.2x6 - Deep Rybka 4.1 SSE42 x64 (x6) 24.5 - 15.5 +15/=19/-6 61.25%
Stockfish 270513 64 SSE4.2x6 - Houdini 3 Pro x64_6 11.5 - 28.5 +1/=21/-18 28.75%

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 2 min +2 sec/ game
No tablebases

201305_SF27052013_2.2

Stockfish 270513 64 SSE4.2x6 - Critter 1.6 64-bitx6_nob 21.5 - 18.5 +10/=23/-7 53.75%
Stockfish 270513 64 SSE4.2x6 - Deep Rybka 4.1 SSE42 x64 (x6) 23.0 - 17.0 +14/=18/-8 57.50%
Stockfish 270513 64 SSE4.2x6 - Houdini 3 Pro x64_6 16.0 - 24.0 +7/=18/-15 40.00%

Global 6 cores: (115.5-124.5) = 48.13%

240 GAMES x6 CORES: http://www.mediafire.com/?roopchehis51dqm

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 4 min + 0 sec/ game
Ponder: Off
No tablebases

20130527_SF270513_4+0-1 2013

Stockfish 270513 64 SSE4.2 - Critter 1.6a 64-bitnob_4 21.0 - 19.0 +9/=24/-7 52.50%
Stockfish 270513 64 SSE4.2 - Deep Rybka 4 x64 (x4) 26.0 - 14.0 +13/=26/-1 65.00%
Stockfish 270513 64 SSE4.2 - Houdini 3 Pro x64_4 15.0 - 25.0 +6/=18/-16 37.50%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 2 min + 2 sec/ game
Ponder: Off
No tablebases

20130527_SF270513_2+2 2013

Stockfish 270513 64 SSE4.2 - Critter 1.6a 64-bitnob_4 21.0 - 19.0 +7/=28/-5 52.50%
Stockfish 270513 64 SSE4.2 - Deep Rybka 4 x64 (x4) 24.0 - 16.0 +12/=24/-4 60.00%
Stockfish 270513 64 SSE4.2 - Houdini 3 Pro x64_4 16.0 - 24.0 +4/=24/-12 40.00%

240 gamesX4 cores : http://www.mediafire.com/?2qbocj4w1vfjxix

Overall average with 4 cores: (123.0-117.0) = 51.25%

Global Score: 238.5-241.5 (49.69%)

Against Critter 1.6: 51.56% Deep Rybka 4: 60.94% Houdini 3.0: 36.56%

Estimated ELO Performance: 3.092

Not so good score this time.

Regards,

Tom.
Top
Uri Blass
Posts: 10882
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Timestamps of Stockfish.

Post by Uri Blass »

I see no functional change from 23.5 to 27.5 so I could expect no improvement.

I also guess that 27.5 is slightly slower than 23.5

did not test it but I am based on the following post of marco:

https://groups.google.com/forum/#!topic ... 3OoPjmqBkY

"I have merged pawn structure in main evaluation (with the small speed penalty of an additional apply_weight call every time we call evaluate, instead before was done only at pawns evaluation)."

In other words marco admits that the change from 25.5 made stockfish slightly slower.

http://abrok.eu/stockfish/
Top
Tomcass
Posts: 786
Joined: Sun Apr 16, 2006 9:09 pm

Re: Timestamps of Stockfish.

Post by Tomcass »

Thanks for your comment, Uri Blass. :-)

STOCKFISH PA_GTB_ 29-05-13 : 480 GAMES

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 4 min +0 sec/ game
No tablebases

SF290513 64x6 - Critter 1.6 64-bitx6_nob 4-30-6 19.0-21.0 47.50%
SF290513 64x6 - Deep Rybka 4.1 SSE42 x64 (x6) 12-22-6 23.0-17.0 57.50%
SF290513 64x6 - Houdini 3 Pro x64 5-20-15 15.0-25.0 37.50%

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 2 min +2 sec/ game
No tablebases

SF290513 64x6 - Critter 1.6 64-bitx6_nob 9-27-4 22.5-17.5 56.25%
SF290513 64x6 - Deep Rybka 4.1 SSE42 x64 (x6) 17-17-6 25.5-14.5 63.75%
SF290513 64x6 - Houdini 3 Pro x64_ 5-20-15 15.0-25.0 37.50%

240 games x6 = http://www.mediafire.com/?7jf7bjcd62xmxx2

Overall average with 6 cores: (120.0-120.0) = 50.00%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 4 min + 0 sec/ game
Ponder: Off
No tablebases

SF290513 64 - Critter 1.6a 64-bitnob 11-22-7 22.5-17.5 56.25%
SF290513 64 - Deep Rybka 4 x64 (x4 16-20-4 26.0-14.0 65.00%
SF290513 64 - Houdini 3 Pro x64_ 4-21-15 14.5-25.5 36.25%


i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 2 min + 2 sec/ game
Ponder: Off
No tablebases

SF290513 64 - Critter 1.6a 64-bitnob_4 10-26-4 23.0-17.0 57.50%
SF290513 64 - Deep Rybka 4 x64 10-21-9 20.5-19.5 51.25%
SF290513 64 - Houdini 3 Pro x64_ 5-21-14 15.5-24.5 38.75%

240 games X4 = http://www.mediafire.com/?lce7fhuscy54kxd

Overall average with 4 cores: (122.0-118.0) = 50.83%

Global Score: 242.0-238.0 (50.41%)

Against Critter 1.6: 54.37% Deep Rybka 4: 59.37% Houdini 3.0: 37.50%

Estimated ELO Performance: 3.097

Regards,

Tom.
Top
ernest
Posts: 2053
Joined: Wed Mar 08, 2006 8:30 pm

Re: Timestamps of Stockfish.

Post by ernest »

Tomcass wrote:Not so good score this time.
Don't forget that the 95% error bar is of the order of ±3%
Top
Tomcass
Posts: 786
Joined: Sun Apr 16, 2006 9:09 pm

Re: Timestamps of Stockfish.

Post by Tomcass »

Thanks for reminding this point, Ernest. I don't forget it. :)

STOCKFISH 01-06-13 : 480 GAMES

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 4 min +0 sec/ game
No tablebases

Stockfish 010613 64 SSE4.2_6 - Critter 1.6 64-bitx6_nob 17.0 - 23.0 +6/=22/-12 42.50%
Stockfish 010613 64 SSE4.2_6 - Deep Rybka 4.1 SSE42 x64 24.5 - 15.5 +18/=13/-9 61.25%
Stockfish 010613 64 SSE4.2_6 - Houdini 3 Pro x64_6 14.0 - 26.0 +6/=16/-18 35.00%
I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 2 min +2 sec/ game
No tablebases

Stockfish 010613 64 SSE4.2_6 - Critter 1.6 64-bitx6_nob 19.0 - 21.0 +9/=20/-11 47.50%
Stockfish 010613 64 SSE4.2_6 - Deep Rybka 4.1 SSE42 x64 24.5 - 15.5 +14/=21/-5 61.25%
Stockfish 010613 64 SSE4.2_6 - Houdini 3 Pro x64_6 15.0 - 25.0 +6/=18/-16 37.50%

240 games x6 = http://www.mediafire.com/?gxwplc5af77im6a

Overall average with 6 cores: (114.0-126.0) = 47.50%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 4 min + 0 sec/ game
Ponder: Off
No tablebases

Stockfish 010613 64 SSE4.2 - Critter 1.6a 64-bitnob_4 22.0 - 18.0 +8/=28/-4 55.00%
Stockfish 010613 64 SSE4.2 - Deep Rybka 4 x64 (x4) 20.5 - 19.5 +5/=31/-4 51.25%
Stockfish 010613 64 SSE4.2 - Houdini 3 Pro x64_4 16.5 - 23.5 +5/=23/-12 41.25%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 2 min + 2 sec/ game
Ponder: Off
No tablebases

Stockfish 010613 64 SSE4.2 - Critter 1.6a 64-bitnob_4 21.0 - 19.0 +8/=26/-6 52.50%
Stockfish 010613 64 SSE4.2 - Deep Rybka 4 x64 (x4) 22.5 - 17.5 +8/=29/-3 56.25%
Stockfish 010613 64 SSE4.2 - Houdini 3 Pro x64_4 20.5 - 19.5 +10/=21/-9 51.25%

240 games X4 = http://www.mediafire.com/?g5u8438z8v6w7eo

Overall average with 4 cores: (123.0-117.0) = 51.25%

Global Score: 237.0-243.0 (49.37%)

Against Critter 1.6: 49.37% Deep Rybka 4: 57.50% Houdini 3.0: 41.25%

Estimated ELO Performance: 3.090


Marco, thanks for making the name shorter. Now it is easier for me to test Stockfish. :wink:

Regards from Barcelona,

Tom.
Top
Tomcass
Posts: 786
Joined: Sun Apr 16, 2006 9:09 pm

Re: Timestamps of Stockfish.

Post by Tomcass »

Following Chess Dragon suggestion I retested Stockfish 01/06/2013, now with Sleeping Threads = Off and Min Split Depth= 6

The result was better than using standard settings. (See post above). Let's keep testing.


TESTING STOCKFISH 01/06/2013 WITH SLEEPING THREADS OFF AND MIN SPLIT DEPTH 6: 480 GAMES.

STOCKFISH 01/06/2013 6_6_6 Sleeping Threads: Off. 240 games

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 4 min +0 sec/ game
No tablebases

Stockfish 010613 64 SSE4.266 - Critter 1.6 64-bitx6_nob 22.5 - 17.5 +13/=19/-8 56.25% Stockfish 010613 64 SSE4.266 - Deep Rybka 4.1 SSE42 x64 23.0 - 17.0 +12/=22/-6 57.50% Stockfish 010613 64 SSE4.266 - Houdini 3 Pro x64_6 16.5 - 23.5 +7/=19/-14 41.25%

I7 980 3.33 Ghz.
6 real cores
Ponder: Off.
GUI: Fritz 12
Book: Perfect 2012
Time control: 2min +2 sec/ game
No tablebases

Stockfish 010613 64 SSE4.266 - Critter 1.6 64-bitx6_nob 22.0 - 18.0 +11/=22/-7 55.00% Stockfish 010613 64 SSE4.266 - Deep Rybka 4.1 SSE42 x64 26.0 - 14.0 +15/=22/-3 65.00%
Stockfish 010613 64 SSE4.266 - Houdini 3 Pro x64_6 17.0 - 23.0 +5/=24/-11 42.50%

240 games : http://www.mediafire.com/?ut8mb1ct77qk2r8

Overall average with 6 cores: (127.0-113.0) = 52.92%

STOCKFISH 01/06/2013 6_4_4 Sleeping Threads: Off. 240 games

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 4 min + 0 sec/ game
Ponder: Off
No tablebases

Stockfish 010613 64 SSE4.255 - Critter 1.6a 64-bitnob_4 20.5 - 19.5 +7/=27/-6 51.25%
Stockfish 010613 64 SSE4.255 - Deep Rybka 4 x64 (x4) 25.0 - 15.0 +15/=20/-5 62.50%
Stockfish 010613 64 SSE4.255 - Houdini 3 Pro x64_4 18.0 - 22.0 +7/=22/-11 45.00%

i7 975 3.33 Ghz.
4 real cores
GUI: Fritz 12
Book: Fritz 12
Time control: 2 min + 2 sec/ game
Ponder: Off
No tablebases
20130601_SF01061355_2+2 2013

Stockfish 010613 64 SSE4.255 - Critter 1.6a 64-bitnob_4 19.5 - 20.5 +4/=31/-5 48.75%
Stockfish 010613 64 SSE4.255 - Deep Rybka 4 x64 (x4) 20.0 - 20.0 +8/=24/-8 50.00%
Stockfish 010613 64 SSE4.255 - Houdini 3 Pro x64_4 16.5 - 23.5 +4/=25/-11 41.25%

240 games: http://www.mediafire.com/?a362x87c1mj57yj

Overall average with 4 cores: (119.5-120.5) = 49.79%

Global Score: 246.5-233.5 (51.35%)

Against Critter 1.6: 52.81% Deep Rybka 4: 58.75% Houdini 3.0: 42.50%

Estimated ELO Performance: 3.101

Regards,

Tom.
Top

Return to “Computer Chess Club: Tournaments and Matches”