Search found 2456 matches

by Milos
Tue Aug 04, 2020 11:32 pm
Forum: Computer Chess Club: General Topics
Topic: Is this SF NN almost like 20 MB book?
Replies: 30
Views: 2048

Re: Is this SF NN almost like 20 MB book?

How bad is NNUE eval really one sees when testing SF depth 5 with minimum eval (material+mobility+PST) vs SF-NNUE depth 1. It's a slaughter house. I’m shocked! Shocked! Thing A at depth 5 slaughtered thing B at depth 1? Hard to believe. Well thing B having approximate eval of depth 12 or 16 or 18 o...
by Milos
Tue Aug 04, 2020 10:01 pm
Forum: Computer Chess Club: General Topics
Topic: Is this SF NN almost like 20 MB book?
Replies: 30
Views: 2048

Re: Is this SF NN almost like 20 MB book?

How bad is NNUE eval really one sees when testing SF depth 5 with minimum eval (material+mobility+PST) vs SF-NNUE depth 1. It's a slaughter house. I’m shocked! Shocked! Thing A at depth 5 slaughtered thing B at depth 1? Hard to believe. Well thing B having approximate eval of depth 12 or 16 or 18 o...
by Milos
Tue Aug 04, 2020 9:40 pm
Forum: Computer Chess Club: General Topics
Topic: Is this SF NN almost like 20 MB book?
Replies: 30
Views: 2048

Re: Is this SF NN almost like 20 MB book?

https://github.com/dkappe/leela-chess-weights/wiki/Leela-is-an-opening-book-nonsense Is this meant to be a fair analysis of the idea presented? I’m sorry I hurt your feelings. :D (BTW, I found your engine to be an excellent sparring partner during the development of a0lite.) It was entirely predict...
by Milos
Tue Aug 04, 2020 9:28 pm
Forum: Computer Chess Club: General Topics
Topic: Is this SF NN almost like 20 MB book?
Replies: 30
Views: 2048

Re: Is this SF NN almost like 20 MB book?

Oh boy. Most nnue are trained at the beginning without regard to game outcome, often at depth 8. Most of the positions they see are maybe 18 to 19 ply into the game and later. They are in essence an approximation of an engine eval at d8. And how is that different to Cerebellum book of Brainfish? It...
by Milos
Tue Aug 04, 2020 9:19 pm
Forum: Computer Chess Club: General Topics
Topic: Is this SF NN almost like 20 MB book?
Replies: 30
Views: 2048

Re: Is this SF NN almost like 20 MB book?

... Some tournaments disallow "books" but allow "neural networks", even if this distinction does not exist in reality, because you can train a neural network to remember openings. But the most of chess engine can not "read" neural net, so they need common opening book. So I encode the book for a cl...
by Milos
Tue Aug 04, 2020 5:30 am
Forum: Computer Chess Club: General Topics
Topic: Sure no one here can beat this simple free smartphone engine!
Replies: 41
Views: 5256

Re: Sure no one here can beat this simple free smartphone engine!

Mr. Milos the reason why I am using Shredder Gui and Larry uses Fritz Gui is because the Arena Gui causes Komodo to play either too fast or ignore that it first has to reach certain Depth before it let the engine select the candidate move. Take a look at this game where Jorge Sammour FIDE rated 245...
by Milos
Mon Aug 03, 2020 7:11 am
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 19716

Re: Sergio Vieri second net is out

and lastly , you will be able to switch back and forth easily from the command line https://i.ibb.co/YTFY10V/honeyx-I-r4.jpg right click and select open in new window to a more readable version hit the return key to get current status" "s nn true" to enable NN evaluation run your search "s nn false...
by Milos
Mon Aug 03, 2020 6:42 am
Forum: Computer Chess Club: General Topics
Topic: Repository for Stockfish+NNUE Android Builds
Replies: 3
Views: 609

Re: Repository for Stockfish+NNUE Android Builds

I think it is only a matter of time before SF NNUE becomes the actual SF as it seems it is much stronger. When that happens maybe there will be a switch of engines in the SF application for mobile. I am still not sure how the engines compare at the reduced hardware though, but theoretically it shou...
by Milos
Mon Aug 03, 2020 6:21 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: NNUE accessible explanation
Replies: 12
Views: 3959

Re: NNUE accessible explanation

As for inference, different projects rely on support on different levels. I don't know any projects which rely on high level libraries directly for neural network support. I think that is a fine route if you don't need the absolute maximum of performance and customization options. The main downside...
by Milos
Mon Aug 03, 2020 5:49 am
Forum: Computer Chess Club: General Topics
Topic: New chess rig
Replies: 64
Views: 4651

Re: TRX40 mobos

If you think you know something about my system that I don't please enlighten us. I think this will expose us to more substance than just your verbal diarrhea. I know that you either don't know how to reliably measure nps of SF or that you are simply lying, because gaining 15+% of average SF perfor...