Search found 760 matches

by xr_a_y
Mon Oct 28, 2019 6:05 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Pawn hash table, a little disappointment
Replies: 15
Views: 748

Pawn hash table, a little disappointment

I added a pawn hash table to Minic in order to store everything pawn related (passed, isolated, doubled, backward, candidate, holes, attack, openfile, semiopenfile, ...) So the entry of the hash table is quite big (sizeof(PawnEntry) = 140) ... I'm using a total size for this table that is 1/16 of th...
by xr_a_y
Mon Oct 28, 2019 5:58 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: deep history pruning
Replies: 8
Views: 701

Re: deep history pruning

If you try the history scores indexed by ply, let me know how this worked out for you Vivien, this has been on my "Try THIS" list for along time now and I haven't come around to do it... But hearing what bob has said maybe it would be more sensible to have history scores indexed by subtree, but thi...
by xr_a_y
Mon Oct 28, 2019 5:55 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: deep history pruning
Replies: 8
Views: 701

Re: deep history pruning

Well it was easy to implement, so I tried it yesterday. This was not a success. I try to accumlate history by ply with a bandwidth of +/-2 6 10 and 20 ply. I also try different sum strategies to aggregate history from various ply. Nothing seems to be good, and some were really really bad. Of course ...
by xr_a_y
Sun Oct 27, 2019 4:24 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: deep history pruning
Replies: 8
Views: 701

Re: deep history pruning

A while back (can't say about today since I have not followed this) history counters were found to be not very useful. With the very deep searches, the same move can be good in parts of the tree and bad in others. When Fruit came out, I played with the history pruning setting on our cluster and dis...
by xr_a_y
Wed Oct 23, 2019 2:25 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: deep history pruning
Replies: 8
Views: 701

deep history pruning

history leaf pruning is a thing. But how about history prune at higher depth with a history score limit that depends on depth ? It seems to be a little win inside Minic, but not really tuned yet.
by xr_a_y
Tue Oct 22, 2019 6:08 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: FGRL 60 sec + 0.6 sec Rating list - Komodo 13.1, Topple 0.7.2, RuyDos 1.1.10, Cheese 2.1
Replies: 10
Views: 1728

Re: FGRL 60 sec + 0.6 sec Rating list - Komodo 13.1, Topple 0.7.2, RuyDos 1.1.10, Cheese 2.1

xr_a_y wrote:
Tue Sep 17, 2019 6:45 am
Ok so this GUI thinks "Hash" option is available in all engines. I think this is not stated in the protocol spec. But anyway, I will make that available in next release (or at least not an error).

Thanks for testing.
Should be working now ;)
by xr_a_y
Sun Oct 20, 2019 6:12 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Looking for TT policy advice
Replies: 28
Views: 4115

Re: Looking for TT policy advice

150 elo diff sounds like an engine playing without any TT at all. I remember measuring the impact of tt in Igel since I completely re-written it since the fork from Greko. Everybody had a 1Mb TT during this test. So there are two possibilities : - either Minic TT policy is stupid for small TT - or ...
by xr_a_y
Thu Oct 17, 2019 5:55 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Looking for TT policy advice
Replies: 28
Views: 4115

Re: Looking for TT policy advice

xr_a_y , I cannot really answer yes to that question ... It may indeed be the case. Are you thinking of a particular parameter ? I thought TT was a good candidate because with the used replacement scheme. I think instead of spending significant time on some idea it may be worth rechecking it: if yo...
by xr_a_y
Wed Oct 16, 2019 5:19 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Looking for TT policy advice
Replies: 28
Views: 4115

Re: Looking for TT policy advice

xr_a_y , I cannot really answer yes to that question ... It may indeed be the case. Are you thinking of a particular parameter ? I thought TT was a good candidate because with the used replacement scheme. I think instead of spending significant time on some idea it may be worth rechecking it: if yo...