Author: Enrique Irazoqui
Date: 08:28:16 01/19/99
Go up one level in this thread
On January 19, 1999 at 10:57:20, Harald Faber wrote: >On January 19, 1999 at 10:36:39, Enrique Irazoqui wrote: > >>>>>>>>Next match will be Genius3-MCP8 to compare MCP8 a bit with MCP7. >>>>>> >>>>>>Please note, that Genius is "Angstgegner" to Mchess! In SSDF play M8 - G5 >>>>>>result was 15,5 - 24,5, even if M8 is rated higher. >>>>> >>>>>Yes, I have seen it too with a smile in my face. :-) >>>>>However, what can we say if MCP8 wins? :-) >>>>>MCP8 is stronger than MCP7 or Genius3 is weaker than Genius5? ;-) >>>> >>>>I don't think you can say anything regarding strength after only one match. >>> >>>For it contains 40 games one can say a bit. >>>Of course for statistical relevance we need >500 games but none of us will do >>>that. >> >>I was not only referring to the number of games, but to the fact that some >>programs are "customers" of others. Program A can intensly dislike program B, >>without this meaning that B is stronger than A. I think it’s important to make >>them play as many different opponents as feasible. > >Yu are right but my time is limited. If I had 10 PCs I'd do such tests but I >haven't. And from time to time I have other things to do then to let my both >computers play each other. ;-) > >>>And I also know of Genius being "Angstgegner" for MCP. But I repeat, I only >>>wanted to see if old predecessors are able to compete with actual programs. In >>>case of Genius3 I would certainly say YES very clearly. >> >>I am very interested in this too. Genius 3( or 4/5/6, quite the same)is, I >>think, quite a bit weaker than Junior 5, Hiarcs 7 and Tiger, for example. >IF weaker then I'd say not more than 20pts. Based on what? According to the SSDF, Junior 5 is 60 points stronger than Genius 5, both on P200MMX, and G5 is 16 points stronger than G3, both on P90s. In my own tournament, J5 is getting 101 more Elo points than G6. Aside from all these games, watching them play is quite revealing: Junior 5 is obviously stronger, and if it were by only 20 points we wouldn't even notice. >As I already wrote, the top10 programs are equal strong, maybe Junior5 and Hiarcs7 another bit above. Again: why do you think so? >Theyonly differ in style. Among other things. What you are saying is that chess programming did not advance a bit in the last 5 years. I strongly disagree. Look at comp-comp results, at Aegon results, at matches against world class players, at the opinions expressed by Kramnik, Kasparov and Anand... Or else, watch them play and you'll see. Alright, if this is what you truly believe, I'm ready to make a bet: I'll enter Genius 3 in my tournament. If it scores at only 20 Elo under the top, you win. If it scores worse than -50 under the top, I win. Now: what do we bet? You must make it really worthwhile.... One thousand Euros? :) Enrique >>>>In my opinion, there is no difference of strength between the engines in >>>>G3/4/5/6. Instead, I think M8 is stronger than M6/7 and Genius. >>>>Enrique >>> >>>Me too, but that is only a feeling, nothing based on facts so far. But I hope to >>>see it soon although it is hard for me to recognize strength improvement at that >>>level... >> >>It is getting very difficult... > >At least for us patzes who can almost only run testsuites... > >I am looing forward to that Genius3-MCP8 match, I am really curious. :-)
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.