Search found 634 matches

by Desperado
Wed Aug 19, 2015 12:02 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

I made this up, but believe me, the tuner results are close to this example and it doesn't matter how my tuner configuration looks like. The effect will be seen as the tuner is close to finish its doing. If i would stop the tuner in the middle of the runtime, reasonable values appear occasionally, b...
by Desperado
Wed Aug 19, 2015 11:37 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

I would prefer a general solution, that does not depend on restrictions on the values, but on a adaption for phase computing or more general weighted evaluations. That would be good indeed for positional factors but for piece values I have some doubts. The crux is: 1: simple parameter types score_t...
by Desperado
Wed Aug 19, 2015 11:27 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

Mentioned this... well, the fitness function returns winning probabilities and it is pretty obvious that mg advantages/values are more exhalable. It is the mg values that matters and not the winning probabilities. Winning probabilities is the more exhalable. I am not sure what you mean. double Data...
by Desperado
Wed Aug 19, 2015 8:15 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

The result values of the regression algorithm described at the beginning of the thread are the problem . Think of a single value for the queen which may be 800 , but if i measure values depending on phase i do not get values like [mg,eg] / [750,850] but it looks like [200,1400] for example. So, it ...
by Desperado
Wed Aug 19, 2015 7:42 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

Hi, i just wanted to point out, that i use a standard form of phase computation. Is this what you call standard where you set Q, R, B and N to 1? int Evaluation::getMaterialPhase(int tid,Position* pos) { const int PHASE_P = 0; const int PHASE_N = 1; const int PHASE_B = 1; const int PHA...
by Desperado
Wed Aug 19, 2015 6:27 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

Another note: As described above, phase values like [200,1400] which have the property (200 + 1400) / 2 == the single point value of 800. This is an arbitrary example, but finally the regression algorithm will prefer strongly one phase. My guess was that the reason is an unequal distribution of the ...
by Desperado
Wed Aug 19, 2015 6:11 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

Hi, i just wanted to point out, that i use a standard form of phase computation. You can also choose Fruit computation which is described on chessprogramming.wikispaces.com. The result values of the regression algorithm described at the beginning of the thread are the problem . Think of a single val...
by Desperado
Tue Aug 18, 2015 8:43 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

Hello again. Here is a task i want/need to solve but i guess some help is needed by the forum members. Although the introduced algorithm works pretty fine for my purposes, "Tapered Evaluation" is a real handicap. I am not sure if a new thread would be more useful, but i would like to talk about Tape...
by Desperado
Thu Aug 13, 2015 4:35 pm
Forum: Computer Chess Club: General Topics
Topic: A question to Mr.Gerd Isenberg
Replies: 5
Views: 1561

Re: A question to Mr.Gerd Isenberg

Hi.

I already noticed that yesterday. There seems to be a point in the alphabet where several sites are mixed up. :(
by Desperado
Thu Aug 13, 2015 5:15 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Evaluation Tuning
Replies: 47
Views: 10554

Re: Evaluation Tuning

Hi, @Jon & Matthew I need to reformulate that i find the best possible solution with the given restrictions by RESOLUTION and STEPSIZE. The highest resolution i use is one centipawn and there is no limitation for the stepsize. If i use as shown in my example for the material computation a resolution...