STS Test Results

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

melajara
Posts: 213
Joined: Thu Dec 16, 2010 4:39 pm

Re: STS Test Results

Post by melajara »

Wondering how the latest Stockfish and Leela would score.
Per ardua ad astra
Dann Corbit
Posts: 12540
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: STS Test Results

Post by Dann Corbit »

xr_a_y wrote: {snip}
A lot of work to do ... Is this a good idea to work on evaluation based on STS ?
Using STS to work on your evaluation will ensure that your evaluation is complete and correct.

However, tuning to a test set will not maximize your engine's strength as well as tuning to games.

So I would conclude that it is a good first step, but not a final step.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
User avatar
xr_a_y
Posts: 1871
Joined: Sat Nov 25, 2017 2:28 pm
Location: France

Re: STS Test Results

Post by xr_a_y »

Ok thanks, I'll try to implement Texel tuning method soon to start tuning the eval.
User avatar
xr_a_y
Posts: 1871
Joined: Sat Nov 25, 2017 2:28 pm
Location: France

Re: STS Test Results

Post by xr_a_y »

Here's Minic results (10sec per position): 908/1500

STS1: 73
STS2: 58
STS3: 66
STS4: 68
STS5: 77
STS6: 56
STS7: 65
STS8: 44
STS9: 57
STS10: 57
STS11: 54
STS12: 62
STS13: 72
STS14: 70
STS15: 31