CEGT - rating lists September 27th 2009

[Werner]

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

40 / 20:

This time we added 2301 games made with 46 different engines to our list. See more in our list "Games of the week". In total our 40/20 list is based now on 351.652 games (cracked 350.000 mark).

New engines:
With Twisted Logic 20090922 x64 we managed to make 558 games and the engine got a rating of 2808 - which is 41 points over previous version. Very good result for an engine with a "serious bug in the search"
Hope we soon can test the next version!
Another new version here is Amyan 1.62. The engines has 2542 elos after 540 games and is 34 points in front of version 1.595!!

Updated engines:

Code: Select all

5 Naum 4 x64 4CPU 3096 +15 -15 1124 games (the engine did hold its rating and is still best engine behind Rybka 3).
20 Stockfish 1.4 x64 4CPU 3006 +12 -12 2059 games (holds its rating too - we are waiting for the next version)
178 Onno 1.1.1 w32 2830 +24 -24 434 games (-14 and with -59 still a huge gap to x64 version)
443 Arasan 11.5 x64 2CPU 2562 +22 -22 700 games (-20)
467 Arasan 11.5 x64 1CPU 2541 +20 -20 794 games (-7)
It is the strongest Arasan version in our list. Arasan 11.5 x64 2CPU is 15 points in front of version 10.2 and the 1CPU version is 3 points in front of Arasan 11.3x64 1CPU. As there is not so much difference we need thounsands of games to see real gap. But I think, Jon is on the right way.

In our forum you can see some more matches for 40/120 with Stockfish 1.4 x64 4CPU and 2CPU and Naum 4 x64 2CPU. For the next update we hope to show some results with Shredder 12 and Fritz 12.

40 / 4:
Update follows tomorrow!

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
CEGT-Team

[Werner]

And here our update for the Blitz-List:

New Games: 7.876
All Games: 526.522

New Engines

Amyan 1.61: 2586 - 1000 (much better than previous version, +92 to 1.597)
NanoSzachy 3.6: 2474 - 1096 (little improvements, +28)
Pirat B.0.4.9: 2525 - 700 (no prev. version in the list)
Sloppy 0.22 x64: 2664 - 1000 (+44 to w32)
Twisted Logic 20090922 x64: 2802 - 1000 (+44 to 20090105 - same result as in our 40/20 list)
Twisted Logic 20090922 w32: 2752 - 900 (+38 to 20090105)

We started to test Fritz 12. Here the link: http://cegt.foren-city.de/topic,301,-te ... tz-12.html

Some updates:

Bright 0.4a 2CPU (+300 games +2)
Counter 1.1 (+500 games -1)
DeepSjeng 3.0 w32 2CPU (+500 games +2)
Hiarcs 12.1 2CPU (+300 games -4)
Rotor 0.4 (+400 games +-0)
Stockfish 1.4 JA x64 2CPU (+700 games +2)

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

Werner
CEGT-Team

[adieguez]

Now this is weird, I didn't make such a significant change in order to make it +94 elo better. Maybe it was mainly tested now with its favorite hardware or something like that. Anyway it looks cooler now in that list.

Still, in the other rating list (40/20) it got only +34 ELO better (amyan 1.62 tough but which is definitely not noticeably worse), and that indeed is what I more or less expected.

Have a nice day.

And here our update for the Blitz-List:

New Games: 7.876
All Games: 526.522

New Engines

Amyan 1.61: 2586 - 1000 (much better than previous version, +92 to 1.597)

CEGT-Team

[Edsel Apostol]

Thanks Werner and the CEGT team for the tests. It's nice to see TL making some improvements. I will try harder to make the next version stronger.

[Leto]

Hi Werner, on the CEGT 40/20 coordination forum you ask what happened to the first Fritz 12 vs Naum 4 match. Well if you add up all three results Fritz 12 averaged 34% which is only a difference of 7% from what it got on the first match (26.85%). It may be that the results from the first match are within normal expectancy.

That 45% result it got from the 20 game match might have involved a bit of luck, a lot of things can happen in 20 games.

From the match results I've seen thus far from CEGT Blitz and 40/20 it seems clear that Fritz 12 is weaker than Naum 4.