Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: So why *does* Fritz beat Crafty?

Author: Sarah Bird

Date: 07:09:15 03/27/99

Go up one level in this thread


On March 27, 1999 at 03:48:15, blass uri wrote:

>
>On March 27, 1999 at 01:34:14, blass uri wrote:
>
>>
>>On March 27, 1999 at 00:09:50, Brian MacAskill wrote:
>>
>>>Excuse my ignorance (computer chess is a brand-new interest for me), but why
>>>*does* Fritz beat Crafty?
>>>
>>>I bought Fritz 5.32, and I run the Crafty 16.5 engine under it. Fritz usually
>>>beats Crafty (though not easily... Crafty puts up more of a fight than any other
>>>(free) engine I have).
>>
>>Crafty16.5 under Fritz5.32 is not the real crafty.
>>
>>It has not crafty's opening book and there are other problems.
>>
>>I find that even when there is mate in 1, crafty16.5 (as an engine for Junior)
>>"feels" a duty to use its time when the real crafty does not "feels" this duty.
>>It does not change the result of the game but it proves that crafty as an engine
>>for Junior is a modified crafty and not the real crafty.
>>
>>I do not have Fritz5.32 but I guess that Crafty as an engine for Fritz5.32 has
>>the same problem.
>>
>>The real crafty in ICC is a parallel machine and it is part of the program.
>>
>>
>>
>>>
>>>So what's the reason?
>>>
>>>Is Fritz the result of a team of programmers, while Crafty is the work of one
>>>(apparently very talented) man, Dr. Hyatt? Or is it vice-versa?
>>
>>No
>>The source code of crafty is free when the source code of Fritz is not free and
>>it gives an advantage to Fritz but I do not think that  Fritz is better than
>>Crafty.
>>
>>The real test is to give Crafty and Fritz the same rights to choose every
>>machine that money can buy (including parallel machine) and do a match between
>>them at tournament time control.
>>
>>If Fritz5.32 cannot earn from parallel machine then it is Fritz's problem
>>because the possibility to use parallel machine is part of the crafty's program.
>>
>>If Fritz can win Crafty in this test then Fritz is better than crafty.
>
>Mark young told me that his tests show that Fritz is better than parallel
>crafty.
>
>
>Results against Fritz5:
>
>Matches with Bob Hyatt.
>
>Fritz 5.00(PII300) Vs. Crafty(Quad Ppro200)    Fritz 5 wins +5 -0 =0   TC
>120min  180 sec incs.
>
>I read about these games here but I know that crafty has a bug in these games.
>
>
>
>
>
>Results against Hiarcs7
>
>Matches with Bob Hyatt.
>
>Hiarcs7(PII450)  Vs. Crafty(Quad Xeon PII 400)    Hiarcs7 wins +5 -2 =1  TC
>10min  20 sec incs.
>
>
>Server games with Hiarcs7 against Crafty.
>
>TC - Blitz from 5 min 5 sec incs. to 5 min 12 sec. incs.
>
>Hiarcs7(PII 450) wins
>
>+28 -5 =7
>
>Avg. speed of the crafty's per game was 668 MHz.  Most games were against
>multi-cpu Crafty's.
>
>
>Uri

I was going to mention these games of Mark's since you note them I also point
out the games between Rebel and Crafty at Rebel's homepage.
http://www.rebel.nl/match.htm
UltraB






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.