Stockfish 220917 vs SugaR XPrO 1.3

[gogamoga]

1 core per engine
Hash: 256 Mb
TC: 180s+1
Opening book: HERT openings
No Syzygy

Code: Select all

   # PLAYER              :    Elo  Error  POINTS  PLAYED   (%)  CFS(next)    W    D    L  D(%)
   1 SugaR XprO 1.3      :   3202      4   505.0    1000  50.5         78   87  836   77  83.6
   2 Stockfish 220917    :   3198      4   495.0    1000  49.5        ---   77  836   87  83.6

Games - http://www44.zippyshare.com/v/W2Qgh9cl/file.html

[gogamoga]

Code: Select all

Games Completed = 1000 of 1000 (Avg game length = 393.535 sec)
Settings = Gauntlet/256MB/Book/180000ms+1000ms/M 650cp for 8 moves, D 150 moves/PGN:C:\LB_2.75\Hert500.pgn(500)

 1.  Stockfish 220917         	495.0/1000	77-87-836  	(L: m=0 t=19 i=0 a=68)	(D: r=663 i=103 f=30 s=6 a=34)	(tpm=3174.1 d=30.34 nps=2268564)
 2.  SugaR XprO 1.3           	505.0/1000	87-77-836  	(L: m=1 t=9 i=0 a=67)	(D: r=663 i=103 f=30 s=6 a=34)	(tpm=3181.4 d=30.14 nps=2149477)

p.s for SF I was using zipproth.de compile

[kranium]

Code: Select all

Games Completed = 1000 of 1000 (Avg game length = 393.535 sec)
Settings = Gauntlet/256MB/Book/180000ms+1000ms/M 650cp for 8 moves, D 150 moves/PGN:C:\LB_2.75\Hert500.pgn(500)

 1.  Stockfish 220917         	495.0/1000	77-87-836  	(L: m=0 t=19 i=0 a=68)	(D: r=663 i=103 f=30 s=6 a=34)	(tpm=3174.1 d=30.34 nps=2268564)
 2.  SugaR XprO 1.3           	505.0/1000	87-77-836  	(L: m=1 t=9 i=0 a=67)	(D: r=663 i=103 f=30 s=6 a=34)	(tpm=3181.4 d=30.14 nps=2149477)

p.s for SF I was using zipproth.de compile

Thanks Goga!

Ordo:

Code: Select all

# PLAYER              : RATING    ERROR  POINTS   PLAYED    (%)
1 SugaR XprO 1.3      : 3051.8    8.7    505.0    1000   50.5%
2 Stockfish 220917    : 3048.2    8.7    495.0    1000   49.5%

I calculated LOS at 63%
Ordo gives CFS at 66.9%

conclusion: might as well flip a coin to see which is better


But it's best of course to remove the 19 time losses for Stockfish and 9 for Sugar
and it becomes:

Code: Select all

1.  Stockfish 220917         	486.0/972      68-68-836 
2.  SugaR XprO 1.3           	486.0/972	   68-68-836

and no need to run Ordo

[gogamoga]

and no need to run Ordo

Yep they are pretty the same regarding the strength, though I find it odd that new time management code has passed framework so easily.

[Lyudmil Tsvetkov]

I wonder how people don't get tired of testing engines that are +-0.5 elo from each other.

[Milos]

I wonder how people don't get tired of testing engines that are +-0.5 elo from each other.

They not testing engines that are +/-0.5 Elo from each other, they are testing identical engines. I guess they like to use CPU power for home heating.

[JJJ]

Big surprise, Stockfish has the exact same strenght than Stockfish !

[Nay Lin Tun]

Wonder who will test Nayfish? I got some tips how to change/create new the engine and author names!