Search found 2183 matches

by cdani
Tue Jul 28, 2020 4:27 pm
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 28204

Re: Sergio Vieri second net is out

Now in the download site there is a graphic of rating (fixed d12 games).

https://www.comp.nus.edu.sg/~sergio-v/nnue/
by cdani
Tue Jul 28, 2020 4:22 pm
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 28204

Re: Sergio Vieri second net is out

Now the latest one 20200728-0633.bin seems the strongest, almost outside error margins again in 1000 games. Now there are substantial Elo gains versus old 2141, 10-12 Elo points or so. Fist I tested a bit the two new nets: 20+0.1 # PLAYER : RATING ERROR POINTS PLAYED (%) 1 stnnue20200728-1442 : 285...
by cdani
Tue Jul 28, 2020 7:48 am
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 28204

Re: Sergio Vieri second net is out

The new one 20200727-0928.bin seems on a par with the best in 1000 games. Yes. It seems equal to 2051. 20+0.1: # PLAYER : RATING ERROR POINTS PLAYED (%) 1 stnnuesergio20200727-0928 : 2856.4 1.8 4629.5 9239 50.1% 2 stnnuesergio20200725-2051 : 2855.6 1.8 4609.5 9239 49.9% In 1000 games matches, today...
by cdani
Mon Jul 27, 2020 10:20 pm
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 28204

Re: Sergio Vieri second net is out

The new one 20200727-0928.bin seems on a par with the best in 1000 games. Yes. It seems equal to 2051. 20+0.1: # PLAYER : RATING ERROR POINTS PLAYED (%) 1 stnnuesergio20200727-0928 : 2856.4 1.8 4629.5 9239 50.1% 2 stnnuesergio20200725-2051 : 2855.6 1.8 4609.5 9239 49.9%
by cdani
Mon Jul 27, 2020 2:25 pm
Forum: Computer Chess Club: General Topics
Topic: SF-NNUE going forward...
Replies: 90
Views: 20169

Re: SF-NNUE going forward...

Tuning search is always fine, tuning position- learning code based on hash- learning wouldn't make code much more complicated, I think, even if I'm not good enough in proramming to prove so. Andscacs had a nice hash- storage before SugaR had so, no interest in position- learning based on selected h...
by cdani
Mon Jul 27, 2020 9:55 am
Forum: Computer Chess Club: General Topics
Topic: SF-NNUE going forward...
Replies: 90
Views: 20169

Re: SF-NNUE going forward...

Tuning search is always fine, tuning position- learning code based on hash- learning wouldn't make code much more complicated, I think, even if I'm not good enough in proramming to prove so. Andscacs had a nice hash- storage before SugaR had so, no interest in position- learning based on selected h...
by cdani
Mon Jul 27, 2020 7:58 am
Forum: Computer Chess Club: General Topics
Topic: SF-NNUE going forward...
Replies: 90
Views: 20169

Re: SF-NNUE going forward...

I also think that tuning search for the new NNUE eval will net nice redits. This will complicate the code of Stockfish, though.
by cdani
Mon Jul 27, 2020 6:53 am
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 28204

Re: Sergio Vieri second net is out

2051 seems the best: Someone found at Discord that experimental-20200722-1621.bin is even better. Testing... 2051 prevails. 20+0.1 # PLAYER : RATING ERROR POINTS PLAYED (%) 1 stnnuesergio20200725-2051 : 2857.3 1.8 4659.5 9249 50.4% 2 stnnuesx20200722-1621 : 2854.7 1.8 4589.5 9249 49.6%
by cdani
Mon Jul 27, 2020 6:01 am
Forum: Computer Chess Club: General Topics
Topic: SF-NNUE going forward...
Replies: 90
Views: 20169

Re: SF-NNUE going forward...

My first idea was specialized nets. For example a net for rook endgames. But the fact that the previous net that arrives to a rook endgame does not know about those endgames, will weaken it. So once you start using a net, I think the best is use it till the end. So maybe various nets specialized in ...
by cdani
Sun Jul 26, 2020 8:08 pm
Forum: Computer Chess Club: General Topics
Topic: Sergio Vieri second net is out
Replies: 167
Views: 28204

Re: Sergio Vieri second net is out

cdani wrote:
Sun Jul 26, 2020 6:32 pm
2051 seems the best:
Someone found at Discord that experimental-20200722-1621.bin is even better. Testing...