lichess.org
Donate

Stockfish 16.1 is here!

#41
Official rating ...
tests.stockfishchess.org/tests/view/65d666051d8e83c78bfddbd8
computerchess.org.uk/ccrl/4040/rating_list_all.html
computerchess.org.uk/ccrl/4040/cgi/engine_details.cgi?print=Details&each_game=0&eng=Stockfish%2016.1%2064-bit%204CPU#Stockfish_16_1_64-bit_4CPU
2.4 Creating and Testing Engines
2.3 Endgame Tablebases
2.3.3 Reference Fallible Endgame Play
2.3.4 Can Endgame Tablebases Weaken Play?
digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=1276&context=cpesp
www.reddit.com/r/chess/comments/o57eqt/stockfish_has_removed_the_contempt_option/?rdt=33925
Are engines competing with contempt enabled or not?
computerchess.org.uk/ccrl/4040/about.html
What was used to create the rating ? Hardware, software, engine settings and time controls.
What caused the rating change?

youtu.be/As0b_zhKo88
#42
I pressed the Quote link and this is what it shows. @wonderk

CCRL Blitz Rating List - Custom engine selection
1141591 games played by 2031 programs, run by 23 testers
Ponder off, General books (up to 12 moves), 3-4-5 piece EGTB
Time control: Equivalent to 2'+1" on an Intel i7-4770K. See the "about" page.
Computed on March 3, 2024 with Bayeselo based on 1'141'591 games
Tested by CCRL team, 2005-2024, http://ccrl.chessdom.com/ccrl/404/

Rank Engine Elo + - Score AvOp Games
1 Stockfish 16 64-bit 8CPU 3804 +16 -16 65.7% -97.2 2786
Stockfish 16.1 64-bit 8CPU 3794 +18 -18 61.0% -65.7 1498

computerchess.org.uk/ccrl/404/cgi/compare_engines.cgi?print=Rating+list+%28text%29&eng=Stockfish%2016%2064-bit%208CPU&eng=Stockfish%2016.1%2064-bit%208CPU&only_public_releases=0&only_stable_versions=0&only_default_settings=0&only_best_in_class=0&num_best_in_class=1

computerchess.org.uk/ccrl/404/cgi/compare_engines.cgi?family=Stockfish&print=Rating+list&print=Results+table&print=LOS+table&print=Ponder+hit+table&print=Eval+difference+table&print=Comopp+gamenum+table&print=Overlap+table&print=Score+with+common+opponents
#43
As you wrote:
Stockfish 16 rating 3804.
Stockfish 16.1 rating 3794.
I understand that the difference is minimal and nothing changes for us humans, but the question is another: the developers work hard to build an engine that is stronger than the previous one,
I hope the developers find a way to make a more stable version, maybe let's wait for stockfish 16.2 soon
#44
rating, shrating!

stronger. stable. accuracy while at it?

edit: did it occur to anyone that there might be other better engine criteria than the gladiator rating, one being knowing what the engine is actually doing, and maybe not enthralling all its cousins developpers to keep doing the same unknowable thing to chess, and maybe want to get something under control and optimization for out of arena among counsins benefits, like us humans using them where they are not optimized for, as per position analysis tools.
#45
Wonderful! I'll use this for my game analysis. Thanks for telling us! thanks for developing Stockfish 16.1, Lichess developers!
#46
To @kimia_ashirley’s #45:

Lichess developers are separate from Stockfish’s developers, aren’t they? They didn’t create or develop Stockfish—at least that is true for most of them, I believe.
#47
7 March ccrl blitz
Stockfish 16 , 64 bit rating 3761
Stockfish 16.1 64 bit rating 3746
I Hope into new release stockfish 16.2 soon
#49
Can I ask you something? Does normal stockfish have access to the table base? Or does it find in the start of the 8 peice and use only it's calculation?