Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Congratulations to Christophe and Ed ...

Author: José Carlos

Date: 07:47:41 11/05/01

Go up one level in this thread


On November 05, 2001 at 09:57:01, Frank Quisinsky wrote:

>On November 05, 2001 at 05:56:48, José Carlos wrote:
>
>>On November 05, 2001 at 03:40:36, Frank Quisinsky wrote:
>>
>>>Hi,
>>>
>>>the next big event and Fritz cann't win.
>>>In the most magazines where I can read information about Fritz-Kramnik I must
>>>read Fritz is the best program in the World. But our Fritz can not win the big
>>>tourneys.
>>>
>>>A pity:
>>>We have programs (I am to 70% sure) witch have a better chance against Kramnik
>>>with longer time controls.
>>
>>>Chess Tiger ...
>>>Shredder ...
>>
>>  I don't like Fritz, and I agree with the above, but don't forget Fritz is
>>number one in the SSDF list for a long time
>>
>>>Gandalf ...
>>
>>  Aha! joking again, you're so funny...
>>  You'd better write 'Rebel' in your list. Rebel gets always good results in
>>these events where programmers are involved. Also, Rebel has proved to be very
>>good against humans: playing against them.
>>  Gandalf has won nothing, hasn't played humans, is not top in SSDF. Gandalf is
>>just another strong program like Crafty, Phalanx, Yace... Very good, but not a
>>top program at all.
>>
>>>and now the World must see Fritz (not the 4x computer chess world champion
>>>Shredder), a program which cann't win the most of big computer chess events.
>>
>>  This is true. Also the way Fritz was selected was unfair and, I'd say, stupid.
>>
>>>GM Keene say Fritz is the best program in the World.
>>>Is GM Keene a grandmaster or have Keene tomatos on his eyes.
>>
>>  Everyone can make a mistake. It remainds me of your claim about Gandalf.
>>  But it's not important what program anyone believes is the strongets in the
>>world. There must be a competition, declare a winner, and that winner will be
>>the strongest in the world until the next competition. This is the fair way to
>>do it. Maybe not scientifically perfect, but fair in terms of competition.
>>
>>>What a pity ...
>>>
>>>But all this is not very important, the match Kramnik - Fritz is also boring if
>>>I look in my computer chess kalendar with the results of the last years.
>>>
>>>Now Chess Tiger won the Dutch-ch 2001 and I will say ... Christiophe we all know
>>>how strong is your program and I will say ...
>>>
>>>Congratulations
>>
>>  I agree.
>>
>>  José C.
>>
>>>Best
>>>Frank
>>>
>>>
>>>I wait of tourneys where Fritz can win.
>>>So it is better for me to understand a grandmaster with the name "Keene".
>
>Hi,
>
>make analyses with longer time controls with all top programs.
>You can see that the best analyses comes from Gandalf.

  1. Best analysis is a subjective concept.
  2. Is the match against Kramnik be a whose-analysis-are-better or a playing
chess match?

>In all test suites Gandalf have top results, Gandalf found the most of moves in
>5-10 minutes on stronger hardware. Other programs in different position in 1
>second or not in 30 minutes :-)) Gandalf found moves not in 1 second he must
>calculated and found the most critical moves in 5-10 minutes on very fast
>hardware.

  Is the match against Kramnik be a problem-solving match or a playing chess
match?

>Look in the SSDF
>Gandalf on 450MHz and Gandalf on 1.2GHz.

  If I had to trust SSDF, then I'd have to admit Fritz is #1.

>And now look in your analyses with 10 minutes and you can see what I say now a
>long time.
>
>For me is Gandalf with longer time controls one of the best chess programs. Also
>very strong is Shredder 5 and Junior 7, Chess Tiger 14.0 and Fritz.

  Yes, and GM Keene has a subjective opinion too.

>Or better look in my tourney:
>Amateurs vs. Professionals with 40 in 10 and you see that Fritz is clear
>stronger then all others. Now look in the results from my CCE tourney and you
>can see the different to the others is not to hight.

  Number of games please. Error bars? And more: I can't see how those games are
made. What I can see is public tounaments. I trust what I see, not what anyone
does at home. If I want, I can show you only games Averno has won in my tests.
That'd be a impressive performance.

>And if you now look in long time analyses I will see what you write about
>Gandalf, the program of the future. The problem is the endgame, here are other
>top programs better.

  About the 'program of the future': subjective.
  About endgame: do you believe the match against Kramnik will be a
mid-game-only tournament?

>I hope Steen made Gandalf stronger.
>
>Is Gandalf 5 50 ELO stronger as Gandalf 4, so Gandalf is at the moment on

  According to what? Can you give number of games and error bars of those
estimates?

Athlon
>1.4 and tournament time the strongest program and for sure is version 4 with
>Shredder the strongest program for long time analyses. I have here clear the
>best results with Shredder and Gandalf.
>
>Believe me, I can used x computers for an test and I used x computer.
>I know what I say ...

  I don't understand this last sentence. Please explain.

>Best
>Frank

  José C.



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.