Search found 1985 matches

by Michel
Sun Jan 13, 2019 10:09 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Writing bugs
Replies: 63
Views: 2023

Re: Writing bugs

Of course it is obvious how to emulate the correct object oriented design in C in this simple case. That was left as an exercise to the reader. You were claiming "state is evil" (as usual a statement devoid of any nuance). I am claiming instead: "there is no problem with state as long as it is prope...
by Michel
Sun Jan 13, 2019 8:57 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Writing bugs
Replies: 63
Views: 2023

Re: Writing bugs

State is fine but it should be encapsulated in an object. Like so: rg=random_generator(seed) r=rg.rand() The state maintained by the rg object could be much more complicated than a single number (this would likely be the case for a cryptographically secure random generator). The rg object completely...
by Michel
Sat Jan 12, 2019 6:25 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Yes everything seems to be consistent now. It is interesting to compare the real game measurements with SF's eval (the Gauss curve you posted). It seems that 1cp of SFeval is worth about 1.6 elo. This is quite a bit higher than what I had expected. A small issue is that the ratio between SFeval and ...
by Michel
Fri Jan 11, 2019 4:38 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Ok here are the results (not having much time I did it very quickly so I hope I did not make mistakes). I took the pgns of about 150 tests on fishtest and grouped the results by opening (i.e. by starting fen). The raw data is here http://hardy.uhasselt.be/Toga/global.digest. There were some strange ...
by Michel
Thu Jan 10, 2019 11:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Thx! Great data as usual. It is nice that the normal distribution shows up. Very comforting. Some care must be taken in interpreting this data. Nowadays SF uses of course contempt so that may inflate the bias when measured with real games (unknown in what way). The transformation bias->SF eval has ...
by Michel
Thu Jan 10, 2019 7:21 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Thx! Great data as usual. It is nice that the normal distribution shows up. Very comforting. Some care must be taken in interpreting this data. Nowadays SF uses of course contempt so that may inflate the bias when measured with real games (unknown in what way). The transformation bias->SF eval has i...
by Michel
Wed Jan 09, 2019 2:04 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Thanks! I think this is all very interesting! I still need to do the calculation but except for the exotic case of SF playing against itself with fixed nodes, maybe the elo model can explain things after all... The pentanomial model has 4 degrees of freedom. If we use the BayesElo model then we now...
by Michel
Wed Jan 09, 2019 11:04 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Thanks! I think this is all very interesting! I still need to do the calculation but except for the exotic case of SF playing against itself with fixed nodes, maybe the elo model can explain things after all... The pentanomial model has 4 degrees of freedom. If we use the BayesElo model then we now ...
by Michel
Wed Jan 09, 2019 6:37 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Ok I decided to try to reproduce Kai's results for myself... It turns out that correlation between games in a game pair is real and quite strong... and not only for fixed node games. https://github.com/glinscott/fishtest/issues/348#issue-397088218 It is not important for the application indicated i...
by Michel
Tue Jan 08, 2019 10:00 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Fixed nodes games and the pentanomial model.
Replies: 25
Views: 1133

Re: Fixed nodes games and the pentanomial model.

Ok I decided to try to reproduce Kai's results for myself...

It turns out that correlation between games in a game pair is real and quite strong... and not only for fixed node games.

https://github.com/glinscott/fishtest/i ... -397088218