CEGT - rating lists March 01st 2009

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

User avatar
Werner
Posts: 2871
Joined: Wed Mar 08, 2006 10:09 pm
Location: Germany
Full name: Werner Schüle

CEGT - rating lists March 01st 2009

Post by Werner »

Hi all :-),
our updated rating lists are now online and can be found under the attached links.

40 /120:
New results as usual in our forum: http://cegt.foren-city.de/topic,175,-40 ... -4cpu.html

CEGT 40/400 repeated

Code: Select all

1   Rybka 3 x64 4CPU  ½1½½½1½½½½½½½½½1½½½½0½1½½½½½½½1½1½1½½½ 22.0/38 
2   Naum 4 x64 4CPU   ½0½½½0½½½½½½½½½0½½½½1½0½½½½½½½0½0½0½½½ 16.0/38

Code: Select all

1   Rybka 3 Human x64 4CPU  ½½½1½10½½½½½1½½1½½½½½½½10½½½½½01½½½½½ 20.0/37 
2   Naum 4 x64 4CPU         ½½½0½01½½½½½0½½0½½½½½½½01½½½½½10½½½½½ 17.0/37



40 / 20:
This week we added 965 games made with 32 engines to our list. See more in our list "Games of the week". In total our 40/20 list is based now on 311.088 games.

New engines:
298 Tornado 2.2 x64 2CPU 2650 +33 -33 268 games:
The start-rating is about 50 points ahead of 1CPU-version!

Updated engines:

Code: Select all

35 Stockfish 1.2def x64 4CPU 2959 +23 -23 484 games (-20 cause of bad results against Bright and Rybka 2.2n2)
72 Stockfish 1.2def x64 2CPU 2915 +20 -20 707 games (-7 cause of a bad result against Thinker 5.4A inert)
78 Glaurung 2.1 x64 4CPU 2910 +18 -18 860 games (-1 and now only 30 points behind Stockfish)
30 Deep Shredder 11 x64 4CPU 2968 +11 -11 2281 games ( +-0)
32 Thinker 5.4Ai x64 2CPU 2963 +20 -20 660 games (+12)
42 Thinker 5.4Ai x64 4CPU 2950 +25 -25 370 games (now again behind 2CPU)
48 Bright 0.4a 4CPU 2942 +21 -21 583 games (+7)
232 Crafty 22.9 x64 2CPU 2726 +39 -39 209 games (again -17 - more games are necessary)
286 Crafty 22.9 x64 1CPU 2664 +23 -23 616 games (-3)

40 / 4:
New games 3664; total games = 439.660 in the list. See more in the list of the played games: http://cegt.foren-city.de/topic,67,-upd ... nline.html

New engines

Onno 0.12.0 x64 1CPU has a rating of 2888 after 700 games and is +19 to previous version. The engine will probably be commercial later.

Tornado 2.2: We have started this version for our Blitz Rating List too:

Code: Select all

426 Tornado 2.2 w32 2CPU 2610 +38 -38 240 games
437 Tornado 2.2 x64 2CPU 2602 +38 -38 240 games
484 Tornado 2.2 w32 1CPU 2563 +31 -31 350 games
all have good startratings but too less games!

Updates

Code: Select all

252 Booot 4.15.0 2718 +17 -17 1090 games (-4)
110 Deep Junior 10.1 4CPU 2890 +15 -15 1600 games (+1)
436 Hermann 2.4 x64 2CPU 2603 +24 -24 600 games (-23 now behind 2.3)
399 Junior 5.0 2629 +18 -18 1073 (+11, Gerhard likes such oldies... )
A big „Thank you“ to all testers as usual! :)

links:
40/20: http://www.husvankempen.de/nunn/rating.htm
Blitz: http://www.husvankempen.de/nunn/blitz.htm
40/120: http://www.husvankempen.de/nunn/rating120.htm
Tester: http://www.husvankempen.de/nunn/testers/testers.htm
Games of the week: http://www.husvankempen.de/nunn/40_40%2 ... on/gow.JPG
Elo-comparison: http://www.husvankempen.de/nunn/Replay/ ... arison.htm

Werner
Engin
Posts: 918
Joined: Mon Jan 05, 2009 7:40 pm
Location: Germany
Full name: Engin Üstün

Re: CEGT - rating lists March 01st 2009

Post by Engin »

Werner wrote:Hi all :-),

New engines:
298 Tornado 2.2 x64 2CPU 2650 +33 -33 268 games:
The start-rating is about 50 points ahead of 1CPU-version!

Tornado 2.2: We have started this version for our Blitz Rating List too:

Code: Select all

426 Tornado 2.2 w32 2CPU 2610 +38 -38 240 games
437 Tornado 2.2 x64 2CPU 2602 +38 -38 240 games
484 Tornado 2.2 w32 1CPU 2563 +31 -31 350 games
all have good startratings but too less games!

Werner
why is the x64 version 8 points weaker then the w32 ??
User avatar
Matthias Gemuh
Posts: 3245
Joined: Thu Mar 09, 2006 9:10 am

Re: CEGT - rating lists March 01st 2009

Post by Matthias Gemuh »

Engin wrote:
Werner wrote:Hi all :-),

New engines:
298 Tornado 2.2 x64 2CPU 2650 +33 -33 268 games:
The start-rating is about 50 points ahead of 1CPU-version!

Tornado 2.2: We have started this version for our Blitz Rating List too:

Code: Select all

426 Tornado 2.2 w32 2CPU 2610 +38 -38 240 games
437 Tornado 2.2 x64 2CPU 2602 +38 -38 240 games
484 Tornado 2.2 w32 1CPU 2563 +31 -31 350 games
all have good startratings but too less games!

Werner
why is the x64 version 8 points weaker then the w32 ??

Lowest rating for Tornado 2.2 w32 2CPU is 2610-38 = 2572.
Highest rating for Tornado 2.2 x64 2CPU is 2602+38 = 2640.

What do you mean by "x64 version 8 points weaker then the w32" ?


Matthias.
My engine was quite strong till I added knowledge to it.
http://www.chess.hylogic.de
Engin
Posts: 918
Joined: Mon Jan 05, 2009 7:40 pm
Location: Germany
Full name: Engin Üstün

Re: CEGT - rating lists March 01st 2009

Post by Engin »

Matthias Gemuh wrote:
Engin wrote:
Werner wrote:Hi all :-),

New engines:
298 Tornado 2.2 x64 2CPU 2650 +33 -33 268 games:
The start-rating is about 50 points ahead of 1CPU-version!

Tornado 2.2: We have started this version for our Blitz Rating List too:

Code: Select all

426 Tornado 2.2 w32 2CPU 2610 +38 -38 240 games
437 Tornado 2.2 x64 2CPU 2602 +38 -38 240 games
484 Tornado 2.2 w32 1CPU 2563 +31 -31 350 games
all have good startratings but too less games!

Werner
why is the x64 version 8 points weaker then the w32 ??

Lowest rating for Tornado 2.2 w32 2CPU is 2610-38 = 2572.
Highest rating for Tornado 2.2 x64 2CPU is 2602+38 = 2640.

What do you mean by "x64 version 8 points weaker then the w32" ?


Matthias.
mean that x64 version have 2602 (without +- 38)
and that w32 version have 2610 (also without +- 38)

but 240 games maybe not enough to get exact rating differenz
ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Re: CEGT - rating lists March 01st 2009

Post by ernest »

With infinite analysis on the start_position (using 1 Thread, for reproducibility), I found that Tornado 2.2 x64 is 40% faster than w32.

This means that the x64 should be approximately 35 Elo stronger than w32.

With more games, today's -8 Elo will be certainly be replaced by a positive Elo (unless something is wrong 8-) )