Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Dann Corbit: Question re:CAP

Author: Michel Langeveld

Date: 09:01:36 01/18/00

Go up one level in this thread


>>Dann,
>>
>>I put questionmarkes to the 2 million. Dann are the reaLLY having 2.0 million
>>positions??? On your ftp site in the private section (CAP-members only section)
>>there are only 950.000 unique positions. You said in another messaage on this
>>board that you expect a CD from overseas. Does this mean you don't have 2.0
>>million yet and expect to have 2.0 million positions if all data is merged??
>Yes

This is really great!

[snip]

>>I also saw in the CAP files a lot of positions where the nodecount is < 0. Is it
>>a good idea to analyze these positions 1 ply deeper and save them in the
>>database so eventually we don't have negative values in the primary base
>>anymore....?? (Project "Negakiller")
>Negative nodecount is more than 2^31 nodes.  Some have even wrapped again to
>positive (e.g. you will see 18 plies but only 100K nodes or something strange
>like that).  It is a function of the programs that count the nodes.  Most use a
>4 byte long to do it.
>
>>Michel Langeveld

Ahh... ok I remembered the problem in the newsgroups, craftylist, etc.
Thanks for the clear explaination of the questions!

Regards,

Michel Langeveld



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.