Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: TCSP - Rating?

Author: James Swafford

Date: 19:10:07 05/12/99

Go up one level in this thread


On May 12, 1999 at 20:52:22, James Robertson wrote:

>On May 12, 1999 at 20:14:56, James Swafford wrote:
>
>>On May 12, 1999 at 14:58:58, William H Rogers wrote:
>>
>>>Can anyone tell me the approximate rating of Tom's TCSP.
>>>Although I do not program in 'C', I have studied his program and find it
>>>very much like mine, with the exceptions of the language and hash tables for
>>>mover generation.
>>>Any help would be greately appreciated.
>>>Thanks in advance
>>>Bill
>>
>>
>>I did a small scale experiment several months ago.
>>I don't have the exact figures handy, but off the top
>>of my head...
>>
>>A 1 ply searcher achieved an ICC rating of just under
>>700.  2 ply was 950 or so.  A 3 ply was 13 something.
>>If I remember correctly, a 4 ply searcher with quiescence
>>rated around 1700 or 1750.
>>
>>The experimental program didn't use iterative deepening,
>>and therefore just computed a move and made it without
>>regard to time.  It always had _lots_ of time left
>>at the end of the game....
>>
>>This program wasn't TSCP, but I suspect the rating would
>>be pretty close.  At 4 ply, lots of tactical things
>>get missed, even with a quiescence function.
>>
>>--
>>James
>
>A tremendous amount also depends on the extensions... IE, a recapture and check
>extension alone would probably raise the program's score a hefty amount. I do
>not know much about TSCPs extensions, but I think it does a check extension,
>doesn't it?
>
>James

Yeah, sure does.  The program I was testing did the same thing;
nothing more.  I later added other extensions.



--
James




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.