## Search found 1994 matches

- Sun Jan 13, 2019 8:57 am
- Forum: Computer Chess Club: Programming and Technical Discussions
- Topic: Writing bugs
- Replies:
**66** - Views:
**7973**

### 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...

- 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:
**2856**

### 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 ...

- 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:
**2856**

### 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 ...

- 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:
**2856**

### 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 ...

- 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:
**2856**

### 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...

- 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:
**2856**

### 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...

- 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:
**2856**

### 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 ...

- 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:
**2856**

### 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...

- 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:
**2856**

### 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

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

- Mon Jan 07, 2019 8:41 am
- Forum: Computer Chess Club: Programming and Technical Discussions
- Topic: Fixed nodes games and the pentanomial model.
- Replies:
**25** - Views:
**2856**

### Re: Fixed nodes games and the pentanomial model.

I don't want to make a big deal out of this, but nonetheless let me repeat. What you call correlation in the case of unbalanced opening is not what is technically called correlation. Correlation between two random variables means that the covariance is not zero. https://en.wikipedia.org/wiki/Covari...