Search found 829 matches
- Thu Jan 14, 2021 10:18 am
- Forum: Computer Chess Club: General Topics
- Topic: NNUE nets with depth of 20?
- Replies: 8
- Views: 412
Re: NNUE nets with depth of 20?
It's a quantity vs quality tradeoff. Required nodes to reach a given depth behaves exponentially. A billion positions analyzed to depth 8 takes a 24 hours with a modern 12 core. Depth 12 takes 10 times that. Depth 20 takes (guesstimate) 750-1000 times as long as depth 8. Without distributed generat...
- Thu Jan 14, 2021 3:08 am
- Forum: Computer Chess Club: General Topics
- Topic: NNUE nets with depth of 20?
- Replies: 8
- Views: 412
- Tue Jan 12, 2021 10:55 am
- Forum: Computer Chess Club: Programming and Technical Discussions
- Topic: HalfKP Structure in NNUE
- Replies: 7
- Views: 328
Re: HalfKP Structure in NNUE
Apologies for cross-posting on different sites. I've already posted this question on chess.stackexchange, but it seems it does not reach the proper audience there (few views, no replies despite bounty). I took a stab at answering each of your questions: https://chess.stackexchange.com/questions/336...
- Tue Jan 12, 2021 10:03 am
- Forum: Computer Chess Club: General Topics
- Topic: Komodo Dragon is Kicking Ass in TCEC
- Replies: 46
- Views: 4709
Re: Komodo Dragon is Kicking Ass in TCEC
Good thing I am not in charge, as your ass would be banned purely on stupidity! Are you always this chippy? Actually, no! However, the statement fits the one it addressed, even if a bit to harsh. Hey Robert. I have a mailing list for people who like to trash talk Dkappe. Reach out if you want to be...
- Tue Jan 12, 2021 8:58 am
- Forum: Computer Chess Club: General Topics
- Topic: Interesting Puzzle for Engines
- Replies: 10
- Views: 617
Re: Interesting Puzzle for Engines
As you see the below, both Eman and Ethereal didn't see a4! and a win coming. I deem your claim Fake News. Ethereal, some version, running some Network. info depth 37 seldepth 80 multipv 1 score cp 839 time 189702 nodes 13713555160 nps 72289000 tbhits 0 hashfull 460 pv a5a4 b3a4 e6f6 h3h4 c5c4 h4g4...
- Tue Jan 12, 2021 8:27 am
- Forum: Computer Chess Club: General Topics
- Topic: Interesting Puzzle for Engines
- Replies: 10
- Views: 617
Re: Interesting Puzzle for Engines
As you see the below, both Eman and Ethereal didn't see a4! and a win coming. I deem your claim Fake News. Ethereal, some version, running some Network. info depth 37 seldepth 80 multipv 1 score cp 839 time 189702 nodes 13713555160 nps 72289000 tbhits 0 hashfull 460 pv a5a4 b3a4 e6f6 h3h4 c5c4 h4g4...
- Fri Jan 08, 2021 9:46 am
- Forum: Computer Chess Club: Programming and Technical Discussions
- Topic: Help with Texel's tuning
- Replies: 33
- Views: 1542
Re: Help with Texel's tuning
Depending on your mathematical education, you can read https://github.com/AndyGrant/Ethereal/blob/master/Tuning.pdf. I recommend atleast some basic knowledge about multivariate calculus. Then it is easy. For more questions on the contents of this, you can also ask Andrew directly I suppose or me he...
- Thu Jan 07, 2021 10:47 pm
- Forum: Computer Chess Club: General Topics
- Topic: The Next very strong Engine will be rofChade NNUE
- Replies: 17
- Views: 1937
- Thu Jan 07, 2021 8:48 pm
- Forum: Computer Chess Club: Programming and Technical Discussions
- Topic: How to calc the derivative for gradient descent?
- Replies: 14
- Views: 843
Re: How to calc the derivative for gradient descent?
Thank you and thanks also to AlvaroBegue. Your paper seems very interesting and I'll read it asap. Hopefully you won't mind if I ask if something is not clear to me :) Of course. The code snippets are not a great source of utility, but the words behind them layout a framework for anyone to follow. ...
- Thu Jan 07, 2021 8:47 pm
- Forum: Computer Chess Club: General Topics
- Topic: The Next very strong Engine will be rofChade NNUE
- Replies: 17
- Views: 1937
Re: The Next very strong Engine will be rofChade NNUE
you really must work on the way you use the forums. Quoting yourself means I just skip the post, since I assume I've already read it. Plus you can delete part of the quote history that is not revelent. Otherwise in a long enough forum, the quote takes up an entire page per posting.