15,0000 game Komodo 11.01 test results

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
PaulieD
Posts: 205
Joined: Tue Jun 25, 2013 6:19 pm

Re: 15,0000 game Komodo 11.01 test results

Post by PaulieD » Tue May 30, 2017 9:01 pm

FastGMs Rating Lists (FGRL)
http://fastgm.de/60-0.60.html

Time control: 60 sec./game + 0.6 sec./move

Playing conditions

CPU: Dual Intel Xeon E5-2670 @ 2.6 GHz
OS: Windows 10 64-Bit
Tool: LittleBlitzer GUI, no draw and resign rules
Engines: 1 Core, 64 Bit, default settings
Hash-Table: 128 MB
Time control: 60 seconds + 0.6 seconds
Tablebases: None
Openings: 8moves_v3.epd, random
Ponder: Off
Learning: Off

Ordo v1.2.6: 3000

Best version list: 26.05.2017

New: 4. Komodo 11.01 64-bit

Code: Select all

     Engine                    :    Elo  Error   Games    Points    (%)      W       D      L     D(%)   CFS
 ------------------------------------------------------------------------------------------------------------
   1 asmFish 161212 x64        :   3519      9   13000   11968.5   92.07   11110   1717    173   13.21   100
   2 Stockfish 8 64            :   3501      9   13000   11875.5   91.35   10926   1899    175   14.61   100
   3 Houdini 5 x64             :   3483      9   13000   11781.5   90.63   10818   1927    255   14.82    97
   4 Komodo 11.01 64-bit       :   3472      8   13000   11718.0   90.14   10765   1906    329   14.66   100

yanquis1972
Posts: 1766
Joined: Tue Jun 02, 2009 10:14 pm

Re: 15,0000 game Komodo 11.01 test results

Post by yanquis1972 » Wed May 31, 2017 5:35 pm

i dont understand why +contempt would be stronger than 0 contempt, which is recommended for analysis. personally id like to see testing where contempt & time management are eliminated, but i'm not sure the best way to do that. contempt can be disabled but if you allocate a hard 10s/move (for example) i assume theres a risk of the engine being caught out pondering a better move, which doesnt correspond to real-life usage.

Post Reply