ChessUSA.com TalkChess.com
Hosted by Your Move Chess & Games
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Tuning again
Post new topic    TalkChess.com Forum Index -> Computer Chess Club: Programming and Technical Discussions Flat
View previous topic :: View next topic  
Author Message
Robert Hyatt



Joined: 27 Feb 2006
Posts: 15869
Location: Birmingham, AL
PostPost subject: Re: Tuning again    Posted: Thu Nov 03, 2011 7:11 pm Reply to topic Reply with quote
rbarreira wrote:
michiguel wrote:
Rebel wrote:
Rebel wrote:
Eval tuning I strictly do at fixed depth. I don't want external factors like time control or permanent brain to interfere. Enough volume will flatten all the horizon effects eventually, both sides.

mcostalba wrote:
IMHO the main drawbacks are: impossible to test depth sensible stuff like king safety and artificial same depth for midgame and endgame. But I agree for some evaluation parameters could be good, actually I will give it a try.

For self-play-ply-depth testing I am planning the following:

1. introduce a special parameter for internal testing;
2. when the flag is on increase the depth with 1 when queens are exchanged;
3. depth+2 entering the endgame
4. depth+5 entering the simple endgame

Or something like that.


I limit my search by nodes and I have been happy thereafter. I enthusiastically recommend it.

Miguel


I suppose that's good for parameter tuning but not for bigger eval changes (due to it not accounting for eval speed).


The other issue is one I have pointed out repeatedly. If your program speeds up (or slows down) in nps in a certain phase of the game, you would normally search deeper (if it speeds up, for example). But if you slow down, say in a complicated attacking position, a fixed node count makes it appear you do not. And as a result, you can tune your program to try to reach those kinds of positions, where you do better because you are not getting penalized by slowing down when doing a fixed node test. That NPS variation adds a new variable that is not obvious, and tuning against that is not always a good idea. I've tried both fixed depth, and fixed nodes. Each has places where they work reasonably. But NOTHING replaces using time, overall, because that is how you actually have to play the game, and tuning like you play is much safer overall...
Back to top
View user's profile Send private message
Display posts from previous:   
Subject Author Date/Time
Tuning again Ed Schroder Tue Nov 01, 2011 10:20 am
      Re: Tuning again Edmund Moshammer Tue Nov 01, 2011 11:42 am
            Re: Tuning again Ed Schroder Tue Nov 01, 2011 4:13 pm
                  Re: Tuning again Robert Hyatt Tue Nov 01, 2011 8:04 pm
      Re: Tuning again Joona Kiiski Tue Nov 01, 2011 4:20 pm
            Re: Tuning again Ed Schroder Tue Nov 01, 2011 6:11 pm
                  Re: Tuning again Joona Kiiski Tue Nov 01, 2011 7:25 pm
                  Re: Tuning again Marco Costalba Tue Nov 01, 2011 9:30 pm
                        Re: Tuning again Ed Schroder Tue Nov 01, 2011 11:55 pm
                              Re: Tuning again Marco Costalba Wed Nov 02, 2011 6:04 am
                                    Re: Tuning again Ed Schroder Wed Nov 02, 2011 12:29 pm
                                          Re: Tuning again Michael Hoffmann Wed Nov 02, 2011 1:00 pm
                                          Re: Tuning again Larry Kaufman Wed Nov 02, 2011 2:04 pm
                                          Re: Tuning again Miguel A. Ballicora Wed Nov 02, 2011 2:52 pm
                                                Re: Tuning again Larry Kaufman Wed Nov 02, 2011 3:59 pm
                                                      Re: Tuning again Sam Hamilton Wed Nov 02, 2011 5:18 pm
                                                      Re: Tuning again H.G.Muller Wed Nov 02, 2011 5:29 pm
                                                Re: Tuning again Ed Schroder Wed Nov 02, 2011 7:36 pm
                                                      Re: Tuning again H.G.Muller Wed Nov 02, 2011 8:56 pm
                                                      Re: Tuning again Sam Hamilton Wed Nov 02, 2011 10:45 pm
                                                Re: Tuning again Ricardo Barreira Thu Nov 03, 2011 1:23 pm
                                                      Re: Tuning again Robert Hyatt Thu Nov 03, 2011 7:11 pm
                                                            Re: Tuning again Ricardo Barreira Thu Nov 03, 2011 8:39 pm
                                                            Re: Tuning again H.G.Muller Thu Nov 03, 2011 9:43 pm
                                                                  Re: Tuning again Robert Hyatt Mon Nov 07, 2011 2:08 am
                                          Re: Tuning again H.G.Muller Wed Nov 02, 2011 5:24 pm
                                                Re: Tuning again Ed Schroder Thu Nov 03, 2011 9:16 am
                                                      Re: Tuning again Steve B Thu Nov 03, 2011 11:20 am
                                    Re: Tuning again Rein Halbersma Wed Nov 02, 2011 6:32 pm
                                          Re: Tuning again david nash Thu Nov 10, 2011 8:57 am
                              Re: Tuning again Vincent Diepeveen Sun Nov 06, 2011 3:50 pm
                                    Re: Tuning again Ed Schroder Mon Nov 07, 2011 11:27 am
                                          Re: Tuning again Vincent Diepeveen Thu Nov 10, 2011 4:28 pm
      Re: Tuning again Sergei S. Markoff Thu Nov 03, 2011 10:11 pm
Post new topic    TalkChess.com Forum Index -> Computer Chess Club: Programming and Technical Discussions

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum




Powered by phpBB © 2001, 2005 phpBB Group
Enhanced with Moby Threads