Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: On ICC: Which engines are best against GMs at tc >= 30 0?

Author: Matthew Hull

Date: 11:34:00 08/07/02

Go up one level in this thread


On August 07, 2002 at 00:41:32, Brian Katz wrote:

>On August 06, 2002 at 16:54:37, Matthew Hull wrote:
>
>>On ICC, which engines perform best against strong humans at time controls
>>greater than 30 0?  Is there a way to find out?  If not, maybe the programmers
>>have observations.
>>
>>Thanks!
>
>On ICC I have two sign ons.
>My own sign on is "Katzparov", where no computer chess program use is involved.
>My computer account sign on is "Deep-Katzparov"
>With this sign on I have been using Fritz 7. I was using Deep Fritz with a
>single processor Pentium II 350 MHz 384 SD Ram.
>I use the Time control of 5/12 against mostly Human opposition. I usually set
>the Hash Tables to 32 MB. I also have the Nalimov 4 Disc set of Tablebases plus
>some essential downloaded tablebase files, and some not so essential tablebase
>files as well. This is a Chessbase GUI with Chessbase engines, so in all of the
>games the moves must be manually entered. (Which by the way, I enjoy.)
>Check out my record
>Statistics for Deep-Katzparov(C)         On for:    0     Idle:    0
>
>          rating [need] win  loss  draw total   best
>Bullet      2335  [3]    12     2     1    15
>Blitz       2833        388    30    34   452   2958 (31-Jan-2002)
>Standard    2497  [6]    10     0     3    13
>
>I do have Winboard with the Automatic play feature which I have only used for
>Bullet with the Crafty 18.13 engine. The first.......Approx 197 wins and 10
>losses were by Deep Fritz              (The Number of draws escapes me)
>Most of these games were against Humans. Since then, I have been using Fritz 7
>which has been achieving some astonishing results. I have gotten more losses
>with Fritz 7 only because I have been playing against more Computer programs.
>Either way, 5/12 with 32 MB Hash Tables has been giving some tremendous results.
>Sometimes I have to force Fritz to move when too much time has been used and ICC
>time is running low.) I also set Fritz' clock to add 60 minutes to the Human
>player's time on the Fritz Program. This way the opponent won't lose on time on
>the Fritz GUI. It takes time to manually enter moves and the opponents Fritz
>clock will run out of time. This will screw up the game results for your data
>base. So rather than have to edit the game result later, I avoid that this way.
>
>Best
>Brian Katz


Brian, Thanks!

Also, do you keep track of titled human opponents and their record against your
engine?

Regards,
Matt



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.