TalkChess.com

So for small nets is it a good idea to increase thread count or is that insignificant (or completely irrelevant)? I’m only typically getting ~30kns, fwiw

1, Every type of contempt weakens the engines. No, if this was the case it'd have lower elo on the rating lists. This contempt is just a draw-avoider, for analysis, the user is expected to take care of the cases where avoiding draw is worse than the alternatives, for the rest of scenarios Contempt ...

thanks, looks like it could be interesting to play against. anyone have a link to a windows compile?

after 93 games at 5+5, test30 is stable at -70 elo vs SF10 & -21 elo vs SF10+book (perfect2017/2018). will attach pgn's later, going to try to get a decent sample of SF10 vs SF10+book results. Any more news, yanquis? Been running ltc (classical otb) vs SF10, but only partly to get an idea of streng...

thanks, curious if a code alteration can be intuited that would prevent such extreme depths being necessary.

i can't think atm & this is probably over my head anyway, but in a long time control game stockfish went from >+2.00 to ~0.00 in one move (& that move matched SF's pv) https://lichess.org/study/LHEmOGKM/LAtGrw7r#111 see move comments or pgn (~move 56) [Event "Lc0 test30 gtx1080 vs SF10 6core@4.7ghz ...

theres a link and summary here: https://sites.google.com/site/computers ... t2017books

It’s meant for engine-engine matches, seems to be a good compromise of quality and diversity.

But kais main point, I think, was that experimental runs should be as efficient as possible. Could eg tests20+30, and again test40 (if the aim is truly experimental as the name implies) be performed entirely on 128x10 (or smaller) nets, or does it not translate? test30 is an experimental run, and s...

But kais main point, I think, was that experimental runs should be as efficient as possible. Could eg tests20+30, and again test40 (if the aim is truly experimental as the name implies) be performed entirely on 128x10 (or smaller) nets, or does it not translate?

I’ve assumed the bignet has extended its schedule markedly after publication of several A0 parameters that have been implemented into the training. It’s also only had one LR drop and provided they are aiming for SF10+ strength with it I don’t see the issue with expense.