Author: Will Singleton
Date: 09:23:50 11/11/98
Go up one level in this thread
On November 10, 1998 at 22:22:29, Dann Corbit wrote:
>I propose an EPD study of a large number of EPD positions.
>It would require a large number of machines and operators.
>I would like to run the EPD benchmark on my ftp site through Rebel 10 {both
>settings}, through Crafty, through Hirarcs, and through any other tool with
>sufficient interest that can process EPD.
>
>There are 6751 distinct positions. The goals to be realized by this study are:
>1. Which is the best tool for general positional study?
>2. Which positions are unsolvable by current computer systems?
>3. Which standard EPD positions have the wrong answers in them? ;-) {I am sure
>we will find some}
>4. How does change in time per position affect solve rate?
>
>It will take *at least* 30 machines for each tool in the experiment. Is anyone
>interested? I plan to run this in parallel to my Chess Analysis Project. It
>will also give me calibration data for the accuracy of my answers in the C.A.P.
>
>I would also be very interested in a comprehensive list of tools that can
>analyze EPD input positions and produce an analyzed EPD row as output.
I'm sure people are interested in such a test, but it sounds like a tremendous
amount of cpu time you're asking for. What do you estimate are the time
requirements?
Will
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.