Author: blass uri
Date: 02:47:26 10/29/99
Go up one level in this thread
On October 29, 1999 at 01:39:46, Harald Faber wrote: >On October 28, 1999 at 12:22:06, Christophe Theron wrote: > >>On October 28, 1999 at 09:44:11, Harald Faber wrote: >> >>>I have also seen many outstanding results for Tiger, but I can only remember >>>G/60. Give me (outstanding) results of tournament games on (almost) equal >>>hardware and I will take back all I said and claim the opposite. :-) >> >>You are right. Some of the outstanding results published here were not on equal >>hardware. In fact Tiger had the SLOWEST hardware (remember Enrique's results?) >>but managed to win. That's unfair! > > >I am sure in 60/60 Nimzo7.32 e.g. would also be able to win against MCP8 on >30-50% slower hardware. > > >>I know you are going to find a way out, but anyway: >> >>latest result is from Thorsten. It's Tiger 12.0 against Fritz6, on 2 K6-400 >>computer, games are 40/120 (is it ok for you?) >> >>Tiger 12.0 - Fritz 6: +4 =3 -1 >> >>What? Games are not long enough? Hardware is outdated? What excuse are you going >>to find? > > >I never find excuses like that, it is T. who does, e.g. as he mentioned MY >hardware (K6-200) was outdated. > > >>Thorsten has received Fritz6 recently, so it is physically impossible that he >>biases the result: I have received the games "live" so he would have no time to >>play extra games and select only Tiger's wins, and it is possible to check the >>moves played by both side. >> >>You have explicitely implied that Thorsten was not honest in his testings, which >>is definitely a personal attack to somebody that has not even taken part to the >>discussion. >> >>For your information, Thorsten has not always given superior results to Tiger. >>For example he was not that enthousiastic with previous versions (11.7-11.9). >> >>If you imply that Thorsten's results are not honest, be prepared to face the >>reaction of several other testers that have EXACTLY the same results. > > >I am still waiting. Where are they? IIRC Enrique played 60/60. > > >>It is very easy for you to just sit, relax, and critize when other people spend >>days and night doing real tests. >> >>Have you seen Enrique's results, playing Tiger on INFERIOR hardware? Are you >>trying to make us believe that the result would be completely different if the >>time controls were longer? > > >I don't know if completely different, but different. > > >>That's a classical strategy: if your favorite program does not do well, just >>argue that it would do better at longer time controls. > > >My FAVOURITE program is not Fritz5.32/6 so you are assuming the wrong intention. > > >>I pretend that I accept results of Tiger AT ANY TIME CONTROLS ON ANY COMPUTER, >>given that both opponents have the same hardware and time controls. You can even >>play on one computer with PB OFF. Which programmer is ready to accept this >>challenge? Bob has given up for example. > > >You are ready to, but who takes this? > > >>What is the purpose of your allegations exactly? Defend the last minutes of the >>ChessBase autoplayer supremacy? >> >>Good luck. > > >Sometimes you should read my other posts not concerning Tiger. I already stated >what I think of the CB-autoplayer. I remember that you said that this is a stupid autoplayer and I think that all the autoplayers are stupid. I remember that Ed did not find a special problem with the chessbase autoplayer but found that the autoplayer problem is a general problem. I understood that there is an autoplayer error in the sddf games. It may be 20 elo for some programs and 50 elo for some other programs. I do not know. Uri
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.