Author: Robert Hyatt
Date: 14:12:19 01/24/01
Go up one level in this thread
On January 24, 2001 at 13:07:44, Robert Hyatt wrote:
I should add that the jury is still out on this extension, IMHO. IE you might
try running a test suite with the extension set to each of the values {0, .25,
.5, .75, 1.0} and see how you do overall. I have run a few tests that suggest
that .25 and .5 are actually _faster_ overall as opposed to the .75 I use by
default. I usually test by running a test suite, stopping a search when two
consecutive iterations get the right score, and then taking all the times to
first find and keep the right move and squaring them and summing the result.
this gives a pretty good idea of how the things stack up. I am running the
suite WAC right now with all 5 extensions being tested one at a time with
the values above. I will try to summarize and post the results tonite for
fun...
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.