Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: EGTB Test Positions

Author: David Rasmussen

Date: 11:02:15 01/23/02

Go up one level in this thread


On January 23, 2002 at 10:43:42, David Rasmussen wrote:

>I've just implemented EGTB use in Chezzz. Are there any good test positions that
>will reveal mistakes in the implementation, just as fine70 often reveals
>mistakes in hashing?
>
>I often see people say "if your program can handle 6-man tablebases". Does that
>mean that the normal egtb.cpp code by Nalimov doesn't handle this, or can I just
>download some 6-man tablebases and then the code will automatically use that
>also?
>
>/David

Thanks for all the answers!

I haven't tried any of the tests yet, because I am not at home with my fast
connection these days, so I haven't yet downloaded the 5 man TB's. When I get
home I will try them.

What I am really hoping for, besides positions that can "only" be solved by
programs with TB's, is test positions that will catch typical errors, such as
having converted the mate score from the TB's wrongly or having dealed with en
passant the wrong way etc.

/David



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.