Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hmm . . . is Junior5 not good enough??

Author: Enrique Irazoqui

Date: 05:28:12 10/18/98

Go up one level in this thread


On October 18, 1998 at 04:40:54, Harald Faber wrote:

>On October 17, 1998 at 23:09:16, Mark Young wrote:
>
>>I like computer test. I just wish we could stop testing the top of the line
>>programs Vs CST. Unless the point of this is to gauge which program is best by
>>how badly the beat CST.
>>>Enrique
>
>Maybe it is to demonstrate that Thorsten is wrong. He ALWAYS claims that CST is
>AS STRONG as the top commercial programs. So nothing to say against games
>between CST and the top.

It's not so much showing that Thorsten is wrong as being irritated by so many
ringing bells every time CST wins a game and so much silence when it loses,
presenting it as if it were an eater of giants. I still remember the first
results posted here, where CST supposedly defeated Nimzo 98 2-0 and Fritz 5 also
by 2-0. Then I did the same as now with these results:

CST-Nimzo98    3.5-15.5 (+2 -14  =3)
CST-Fritz 5    1.5-8.5  (+0  -7  =3)
CST-Junior 5   1.5-7.5  (+0  -6  =3)

CST scored a total of 17.1%, or -263 Elo, all games at 40/40.

Interesting as CST may be, it is far from being capable of competing on equal
basis with some much stronger programs. One thing is to see CST as a fun program
and an interesting project, and a very different one to present it as the
achievement it is not.

Enrique




This page took 0.01 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.