TalkChess.com

Why are you linking an article that bases its conclusions on 240/250 game samples? In his defense, thats 10x the sample size of the average talkchess thread :lol: well you're right..good point it's unfortunate, the paper is so well done otherwise, and extremely thorough tested so many variables tab...

All this gloating about how Ethereal NNUE trainer is "original" or whatever while Fire's isn't etc for what exactly ? Reinventing the wheel doesn't make one original or smarter. If one enjoys doing that, that is fine, but bashing others because they don't do it your way is bullshit. People disagree...

No need...the CalPoly study states the facts clearly enough for me. BTW- if you read the study, it comes to the conclusion that SSD or HDD doesn't make a difference, "Engine strength does not appear to have any relationship with tablebase access speed and relative performance gains/losses." so that...

Have you ever actually tried a match between Stockfish with EGTB vs. without EGTB? With 6 man EGTB, one can expect less than 10 elo. For comparison, let's try Stockfish with an evaluation function vs. without an evaluation function... oh wait, that would be stupid. They're not remotely comparable a...

Have you ever actually tried a match between Stockfish with EGTB vs. without EGTB? With 6 man EGTB, one can expect less than 10 elo. For comparison, let's try Stockfish with an evaluation function vs. without an evaluation function... oh wait, that would be stupid. They're not remotely comparable a...

MikeB wrote: ↑

Fri Jun 18, 2021 9:54 pm

I guess this is a fitting epilog to my random walk down NNUE street:

https://github.com/official-stockfish/S ... /pull/3568

the latest and greatest Stockfish net .. until tomorrow perhaps..

wow nice work Michael

Like Fire author said NNUE and Syzygy are pretty similar. Both probe (syzygy code snatched from SF...) evaluation binaries and return evaluation. I also disagree. Syzygy probing code is only really of much use in the rare tricky endgame. For top engines, it has a fairly negligible impact on strengt...

I agree that it's all quite subjective, though I find it somewhat amusing that you suggest unique moves as a criterion for originality after decrying Ed's similarity tests as inadequate for establishing originality (when it was pointed out that Fire had high similarity to Houdini). You know very we...

Thanks Connor I do appreciate all this work...and your attention to detail, unfortunately the exact details of differences between engines is just not important to me personally. I do have an understanding of NNUE, being one of the 1st to help Nodchip last summer, when NNUE 1st appeared. I went so f...

Houdini's lower similarity can be attributed to its evaluation function being somewhat original (at least among publicly availably evaluation function). Regardless, it is a clone due to its search function and many related component being copy pasted from Stockfish on inspection of the code. That's...