Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Move order %

Author: Andreas Guettinger

Date: 01:13:03 03/04/04

Go up one level in this thread


On March 03, 2004 at 21:29:57, Andrew Wagner wrote:

>On the plus side, I think I've *finally* managed to clear up all my
>move-generation bugs. Unfortunately, the ratio of first-move fail-highs to total
>fail highs is still VERY low...many times in the 30-50% range. I have it playing
>on ICC now, and it plays reasonably (1400ish blitz), but only at 4-5 ply because
>the low move ordering makes the nps low.

Move ordering has not much to do with NPS.
Better move ordering will give yo a lower branching factor. But if you're NPS
are low, I suspect there is something wrong in your movegen, makemove or
qsearch.

regards
Andy



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.