Search found 206 matches

by Tony P.
Thu Oct 01, 2020 7:55 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Re-activation Programmer Code of Honor
Replies: 57
Views: 9633

Re: Re-activation Programmer Code of Honor

Re clause 2. I feel it does need to be stricter, like, 'Not to look at reverse-engineered work of other programmers without their express permission.' So, not just to refrain from RE on your own, but also not to borrow ideas from the results of RE done by a whistleblower who's not an engine author a...
by Tony P.
Thu Oct 01, 2020 7:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Re-activation Programmer Code of Honor
Replies: 57
Views: 9633

Re: Re-activation Programmer Code of Honor

The only code I'm probably not going to write myself is endgame TB access. That's "just" database access, and I have no interest in writing code like that. I do that for work half of the time already. I'll happily wrap either Fanthom or Phyrric in my own class, and thank the developers of said code...
by Tony P.
Thu Oct 01, 2020 6:29 pm
Forum: Computer Chess Club: General Topics
Topic: NNUE variation
Replies: 22
Views: 3489

Re: NNUE variation

Minic uses the nn-97f742aaefcd net from the SF NNUE page dated August 3 long before the SF12 release thus most likely a Sergio net which they renamed from bin to nnue. Which would indicate the Minic results are just an accident. Accident or not, I will do 2 extra tests, 1) run Minic with another ne...
by Tony P.
Wed Sep 30, 2020 9:21 pm
Forum: Computer Chess Club: General Topics
Topic: NNUE variation
Replies: 22
Views: 3489

Re: NNUE variation

^ At 1000 ms, Minic's correlations are <=48.28%, whereas between any pair of other engines, they're >=54.77%. If I didn't know which engine had a net trained on a different dataset, I'd easily guess it from the table. (Sorry for the edit, I was looking at the 500 ms table initially; there, Minic <=4...
by Tony P.
Wed Sep 30, 2020 7:04 pm
Forum: Computer Chess Club: General Topics
Topic: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance
Replies: 152
Views: 16863

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

I released a tuning paper a few months ago. It was the culmination of a years effort on various implementations, as well as likely over a hundred different methodologies for building datasets. I shared that paper, and I shared pieces of the Ethereal data, about 10 million positions at a time, to al...
by Tony P.
Tue Sep 29, 2020 9:58 pm
Forum: Computer Chess Club: General Topics
Topic: Is Allie a 'Leela net player'?
Replies: 57
Views: 7282

Re: Is Allie a 'Leela net player'?

^^ SlowBlitz, RofChade, Schooner, and Pedone (and Komodo, though I'm not sure how long it will stick around) are closed-source, and I trust that their authors are honest enough not to copypaste GPLed code. Also, Winter is on a very original route, and I haven't stopped believing in it yet.
by Tony P.
Tue Sep 29, 2020 9:24 pm
Forum: Computer Chess Club: General Topics
Topic: Is Allie a 'Leela net player'?
Replies: 57
Views: 7282

Re: Is Allie a 'Leela net player'?

So Allie can play GM strength (weak) on TCEC hardware, and I can have my name as author on an engine I had no hand in writing? I assumed that Allie's search would be heavily rewritten. With the current search, it would indeed not benefit from running the Gyals at TCEC. I'd also assume that an AB en...
by Tony P.
Tue Sep 29, 2020 8:52 pm
Forum: Computer Chess Club: General Topics
Topic: Is Allie a 'Leela net player'?
Replies: 57
Views: 7282

Re: Is Allie a 'Leela net player'?

^^ Before the NNUE hype started, many thought that small nets wouldn't work, but now that can be reconsidered. That being said, I think the new TCEC rules require the data used to be from the engine itself, so if TCEC applied its rules fairly to all engines, Allie would have to use a net trained usi...
by Tony P.
Tue Sep 29, 2020 8:21 pm
Forum: Computer Chess Club: General Topics
Topic: Is Allie a 'Leela net player'?
Replies: 57
Views: 7282

Re: Is Allie a 'Leela net player'?

All NNUE training data should be generated by the unique engine's own search and/or eval code. Fwiw, I don't think it's the final version of the rule. As is, it's unenforceable, as training code doesn't have to be released, and if the organizers required that code, they'd have a hard time reproduci...
by Tony P.
Tue Sep 29, 2020 7:06 pm
Forum: Computer Chess Club: General Topics
Topic: NNUE variation
Replies: 22
Views: 3489

Re: NNUE variation

Ed, you wrote yourself in the 2019 similarity report : All tested engines in this report are of the alpha-beta type, so our proposed baseline is an alpha-beta baseline. When we test as many neural net engines as possible for our next report, we may well discover a different baseline figure for move ...