LS-Testrun of Stockfish 4: disappointing intermediate result

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
pohl4711
Posts: 2906
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

LS-Testrun of Stockfish 4: disappointing intermediate result

Post by pohl4711 »

Hello,

3300 of 10000 games played and Stockfish 4 scores 5 Elo weaker than the final result of Stockfish 130809 (latest and strongest Stockfish development-version in the LS-ratinglist).
Inside Errorbar, of course, but it looks like Stockfish 4 is a regression no progress...

Final result on Saturday on my website.
Stay tuned!

Stefan

http://ls-ratinglist.beepworld.de/


(Of course I used the right compile (windows x64 for modern computers + SSE42) and the nodes per second are as high as in other Stockfish-versions, which I tested in the LS-ratinglist)
carldaman
Posts: 2287
Joined: Sat Jun 02, 2012 2:13 am

Re: LS-Testrun of Stockfish 4: disappointing intermediate re

Post by carldaman »

pohl4711 wrote:Hello,

3300 of 10000 games played and Stockfish 4 scores 5 Elo weaker than the final result of Stockfish 130809 (latest and strongest Stockfish development-version in the LS-ratinglist).
Inside Errorbar, of course, but it looks like Stockfish 4 is a regression no progress...

Final result on Saturday on my website.
Stay tuned!

Stefan

http://ls-ratinglist.beepworld.de/


(Of course I used the right compile (windows x64 for modern computers + SSE42) and the nodes per second are as high as in other Stockfish-versions, which I tested in the LS-ratinglist)
Regardless, it's no regression when compared to Stockfish 3, even if the latest patches since Stockfish 130809 were regressive, which is not yet a certainty.