These are versions 1.0 till 1.7.
Quite interesting if you ask me
chess22k progress overview
Moderator: Ras
-
sandermvdb
- Posts: 160
- Joined: Sat Jan 28, 2017 1:29 pm
- Location: The Netherlands
chess22k progress overview
These are versions 1.0 till 1.7.
Quite interesting if you ask me
-
Guenther
- Posts: 4718
- Joined: Wed Oct 01, 2008 6:33 am
- Location: Regensburg, Germany
- Full name: Guenther Simon
Re: chess22k progress overview
How did you calculate the eval error (or against what)?
Guenther
Guenther
-
sandermvdb
- Posts: 160
- Joined: Sat Jan 28, 2017 1:29 pm
- Location: The Netherlands
Re: chess22k progress overview
Using a quiet testset with 725.000 positions with known outcome of the match.Guenther wrote:How did you calculate the eval error (or against what)?
-
sandermvdb
- Posts: 160
- Joined: Sat Jan 28, 2017 1:29 pm
- Location: The Netherlands
Re: chess22k progress overview
Btw, the eval error is calculated using the formula that is described on the following page:
https://chessprogramming.wikispaces.com ... ing+Method
The depth and nps are measured when analyzing a middlegame position for 1 minute:
2b5/1p3k2/7R/4p1rP/1qpnR3/8/P4PP1/3Q2K1 w - - 0 47
https://chessprogramming.wikispaces.com ... ing+Method
The depth and nps are measured when analyzing a middlegame position for 1 minute:
2b5/1p3k2/7R/4p1rP/1qpnR3/8/P4PP1/3Q2K1 w - - 0 47