Author: Enrique Irazoqui
Date: 10:37:54 11/08/99
Go up one level in this thread
On November 08, 1999 at 11:37:36, Thorsten Czub wrote: >On November 08, 1999 at 08:33:29, Enrique Irazoqui wrote: > >>>I guess that chesstiger is going to be number 1 in the ssdf list. >>>I guess the difference will be less than 100 elo but more than 30 elo. >> >>After the few hundred games played and posted by different people, I think it >>will be just about 30 points better than the second. >> >>Enrique > >i don't think 30 points are true. >WHO has posted games on fast machines ? >Markus, you and i. I have played 40/120 games. >and my final result against fritz6 is 11.5-6.5 >(+8 =7 -3). >I guess results against CM6 reported by Markus Kaestner in this forum >are also very high for Ctiger. >Fritz6 and CM6 have very high elo-ratings. >Also hiarcs7.32 who loses high. >SO - i don't see why this only gives 30 points. Well... If I give credit only to my games, Tiger was about 100 points above the average of F6, F5.32, N732, H732 and J5, which would make it about 60 points above the second (F532). On the other hand, by other means, I believe that more accurate than my 80 games or so at 60/30, I get this +30. The surprise came today, when I realized that the Windows version is up to 15% (?!) faster than the dos T12 I played in the matches. So let's add 10 more points or so and we have +40 if tests are done with the commercial Windows version. A huge lead anyway... >what i see in the moment is that the ssdf guys do test >uninteresting different hardware-platforms. interesting is IMO >only 450Mhz vs. 450Mhz. The SSDF always did this and their results are accurate as far as comp-comp goes. This has been discussed already a zillion times. Enrique >but as i am testing almost on this level they should get similar kind >of results than i do have. >and my results do result in more than 30 elo. >but we will see. time will tell. right. > > >>>I do not count tournament like paderborn and the NL-ch because opening >>>preperation is more important in these tournament relative to ssdf games and I >>>count only the results of the ssdf. > >the tournament results always differ from ranking lists. >chalifman e.g. is world-chess-champion but kasparov, no longer champion, >has a higher rating, although these rating are made via incest. >so he is overrated. >whatever. >the tiger from paderborn was NOT the tiger from now. >The paderborn tiger was weaker than the tiger we test.
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.