Author: James Robertson
Date: 17:53:11 02/10/99
Go up one level in this thread
On February 10, 1999 at 19:37:01, Bruce Moreland wrote: > >On February 10, 1999 at 15:19:40, James Robertson wrote: > >>Hi Mr. Moreland, >> >>Speaking of null moves: >>My problems with the null move seem to be diminishing.... I added futility >>pruning and several other improvements, and with every speed increase the null >>move helps significantly more. I am hoping that by the time I add hash tables >>and more pruning that I will get gains something like what you described. :) >> >>Thanks for you help. >> >>James > >If you put enough ketchup on it, anything is edible. > >Null move should work well right ouf of the box. I'm not convinced yet that you >are bug-free. > >I suggest you add some #ifdef NULL_MOVE statements at the appropriate locations, >and try your new code with null move turned on and off. It's very possible that >all you are doing is hiding the problem. You should see a big difference with >null move on and off. > >You sent me some code in the mail. I looked at it a bit, but I haven't replied >to your email yet. Since I presume you are reading this, I'll tell you one >thing here: > >"short" is the data-type of Satan. It is way better to use "int" at every >possible opportunity, if you are talking about a Pentium-type machine running >32-bit code. > >bruce Hmmmmm. Ok, that sounds cool. And I'll keep hunting for bugs. Just the other day I found a huge one in the pawn hash that was effectively rendering it useless.... I keep thinking I have cleared all the bugs because it plays a reasonable game, but then I find a new one..... :( James
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.