Booot 6.2 64-bit 4CPU Gauntlet for CCRL 40/40

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
Graham Banks
Posts: 46068
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Booot 6.2 64-bit 4CPU Gauntlet for CCRL 40/40

Post by Graham Banks »

gbanksnz at gmail.com
User avatar
Graham Banks
Posts: 46068
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Booot 6.2 64-bit 4CPU Gauntlet for CCRL 40/40

Post by Graham Banks »

Code: Select all

CCRL 40/40 Rating List - Custom engine selection
723446 games played by 1896 programs, run by 19 testers
Ponder off, General books (up to 12 moves), 3-4-5 piece EGTB
Time control: Equivalent to 40 moves in 40 minutes on Athlon 64 X2 4600+ (2.4 GHz), about 15 minutes on a modern Intel CPU.
Computed on July 1, 2017 with Bayeselo based on 723'446 games
Tested by CCRL team, 2005-2017, http://computerchess.org.uk/ccrl/4040/

Rank              Engine                Elo   +    -   Score  AvOp  Games
1 Booot 6.2 64-bit 4CPU             3221  +25  -24  44.7%  +33.8   456
  Booot 6.1 64-bit 4CPU             3178  +21  -21  39.5%  +63.9   632
gbanksnz at gmail.com
lucasart
Posts: 3243
Joined: Mon May 31, 2010 1:29 pm
Full name: lucasart

Re: Booot 6.2 64-bit 4CPU Gauntlet for CCRL 40/40

Post by lucasart »

If you want 4 CPU and 1 CPU ratings to be comparable (which is important to measure SMP scaling), you should really do more 4 CPU vs 1 CPU. Things like Engine A 4 CPU vs. Engine B 2 CPU, etc.

Without those missing links, BayesElo (or Ordo) have very little information to make the 4 and 1 cpu ratings coherent.
Theory and practice sometimes clash. And when that happens, theory loses. Every single time.