Author: Robert Hyatt
Date: 20:32:46 06/29/03
Go up one level in this thread
On June 29, 2003 at 05:15:46, Geoff wrote:
>Hello Tony
>
>Thanks for the help, I checked this again in a TSCP clone, and it is ply not
>depth.
>
>Also double checked with Crafty, it is ply not depth. Mine doesnt have the -1
>difference, I will try that small correction but I dont expect it to fix it
>either
>Crafty code is hard to follow, I think Bob's term "Draft" is the distance from
>the current node to the leaf node, but I easily could be wrong ;-)
That's correct...
>
>When I tried depth not ply, it again didnt work properly, I got to a winning
>position then reported the mate as getting farther away as it in reality got
>closer to mate. Another game locked up 2 moves from mate with a score reported
>of 10002 (2 past mate) Grrrrrrrrr !! Back to the drawing board.
You _definitely_ don't want depth. The mate score should _never_ depend
on what extensions are triggered, yet that is exactly what using depth
would introduce.
If you do fractional extensions it would be even further confusing... IE mate
in 4.5 moves. :)
>
> if (value > MATE-300)
> value=value+ply-1;
> else if (value < -MATE+300)
> value=value-ply+1;
>
>On June 29, 2003 at 04:26:04, Tony Werten wrote:
>>>Recapping , my hash insert adjust code looks like this
>>>
>>> /* Adjust mate scores so they are not relative to the position... */
>>> if (eval > (INFINITY - MAX_PLY)) /* > 9,900
>>> {
>>> eval = eval + ply;
>>
>>This should be depth, not ply.
>>
>>Tony
>
> Regards Geoff
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.