Author: Michael Henderson
Date: 10:39:01 10/09/04
Go up one level in this thread
On October 09, 2004 at 10:00:05, Stuart Cracraft wrote:
>On October 08, 2004 at 20:37:29, Michael Henderson wrote:
>
>>I'm not sure if I saw code for adjusting mate scores for the hash table. That's
>>really what I'm worried about. If you do not adjust them, your mate threat
>>detection will not work! Btw, are your recaptures _any_ two captures in a
>>row...I don't know if you still have that. If you do, that will definitely get
>>you in the <10 sec range even w/out mate threat extensions.
>>
>>I also saw this code:
>> if (MTHR_EXT != 0.0 && value == MATE+ply+2) { threat=1; }
>>isn't it supposed to be if (value == -MATE...) ?
>>
>
>My MATE is -MAXNUM+100 so in the above, it is like value == -MAXNUM+ply+2.
>
>My recaptures is nearly any two captures in a row but I don't want to solve
>141 with that. I'd like to have mate threat working. Not sure about what is
>meant by adjusting mate scores for the hash table. What's to adjust?
>I find a mate, a modify it by +ply, then I store that position with that
>score in the hash table. Don't I want to do that?!?
>
>Stuart
There is something I forgot to mention. A temporary solution to this is to not
store any score/bound in the vicinity of a mate score. Then your mate handling
in the search will be flawless and your mate threats should then work. I don't
recommend it though because you miss out on TT hits. But you should see
yourself solve WAC 141 fast (without recapture ex.) I tried it and it works
well for me. Once you get adjusted mate scores working (see other post),
compare it with the anti-mate handling TT method.
Michael
Michael
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.