Author: Christophe Theron
Date: 12:54:03 05/02/02
Go up one level in this thread
On May 02, 2002 at 11:32:16, Peter Berger wrote:
>On May 02, 2002 at 00:57:02, Christophe Theron wrote:
>
>>On May 01, 2002 at 19:01:07, Peter Berger wrote:
>>
>>>The games were played with 30 minutes sudden death (blitz) time control; match
>>>was scheduled for 12 games.
>>>
>>>Yace used 2.5 MB for hashtables, no tablebases and a little opening book. The
>>>Palm Vx has been overclocked with AfterBurner - Tiger displays Speed index of
>>>1.10 for this device.
>>>
>>>The hardware advantage for Yace in this match should have been somewhere around
>>>factor 4-6.
>>
>>
>>
>>Thanks for playing the games!
>>
>>Any idea of the rating of Yace on this computer?
>>
>>
>>
>> Christophe
>
>A good starting point for calculation would be the list of Andreas Herrmann IMHO
>( great work):
>
>http://www.wbholmes.de/elo/eloceleron.txt
>
>For adjusting to SSDF niveau engines Crafty, and Nimzo2000b ( similar in
>strength to Nimzo 7.32 and Nimzo 8 ) can be used.
>
>After some more crude maths from my side one should expect at least 2230-2250
>SSDF on 486dx50 for Yace I think - so a very good match result for the little
>Tiger.
>
>The match results still suffer from the little number of games played of course.
>But my personal impression is that Yace doesn't really play at this level on the
>486 - battle against Tiger was open and there weren't any lucky games; it felt
>like equal forces battling it out all the time( except of the first game). I
>suspect Yace is weaker than the expected value on the slow hardware. It missed
>quite a lot tactically and suffered from the low depths it reached very much,
>especially in the endgame.
I have not tried Yace myself, but I hear that it is a tactical monster with an
incredible NPS... So I thought that CT for Palm would have lots of trouble
against it...
>I think I should add a calibration partner to the "contest" soon - this one
>looks quite useful:
>
>82 Genius 4.0 DOS 486/50-66 MHz 2266 23 -23 900 49% 2270
CT for Palm will simply get slaughtered by this one. Try it and you will find
yourself stopping the match after just a few games. :-(
>In fact I was much impressed by Comet B42's play in the 20 games against the
>Tiger so far - it might well be simply much stronger than expected on this slow
>hardware. Yace felt weaker in comparison in its match.
>
>Lots of guesswork - don't slaugter me, lady and guys :).
No problem, your tests are interesting!
Christophe
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.