Author: Robert Hyatt
Date: 13:36:50 06/19/03
Go up one level in this thread
On June 19, 2003 at 01:59:13, Tom Kerrigan wrote: >On June 19, 2003 at 00:22:38, Uri Blass wrote: > >>On June 18, 2003 at 22:59:49, Tom Kerrigan wrote: >> >>>On June 18, 2003 at 20:25:24, Robert Hyatt wrote: >>> >>>>On June 18, 2003 at 13:35:08, Tom Kerrigan wrote: >>>> >>>>>On June 17, 2003 at 20:44:22, Robert Hyatt wrote: >>>>> >>>>>>>I'm not sure how Stobor would do vs. Crafty right now, as I haven't really >>>>>>>worked hard on my program for several years now, but Stobor has been stronger >>>>>>>than Crafty in the past so don't be so sure that Bob is more "clever" than me in >>>>>>>that regard. >>>>>>> >>>>>>>-Tom >>>>>> >>>>>> >>>>>>We only count reality, not dreams. >>>>>> >>>>>>:) >>>>> >>>>>Ha ha, that's a lame dream, to have a program stronger than Crafty. My dream is >>>>>to have a program stronger than Shredder. >>>>> >>>>>-Tom >>>> >>>> >>>>You have to pass crafty first, along the way. I'm knowledgable enough to know >>>>my program isn't stronger than Fritz, etc. I also know enough to know Stobor >>>>has _never_ been stronger than crafty, nor even as strong. Yes you might >>>>have won a game here and there. But pick any 10 consecutive games you want. >>>> >>>>ICC has only 4 such games in its database. Crafty won two and drew two. No >>>>losses. >>>> >>>>I would not write checks _my_ program can't cash. You shouldn't either. >>> >>>10 consecutive games? Okay. How about games 11 to 20 when you do a "search >>>crafty kerrigan" on ICC? I scored 55%. >>> >>>Funny thing is, my hardware has always been much slower than yours, too. >>> >>>Should I also say something inane about check cashing? >>> >>>-Tom >> >>Games on ICC cannot prove that program X is better than program Y. > >No need to have another endless debate about what it takes to prove that program >X is better than Y. Bob's terms were 10 games and I gave him 10 games. If you think that proves superiority, you are sadly mistaken. See my other post that shows that of the 30 10=game samples over that 39 game population, you won 2, three were equal, and you lost 25. I wouldn't count my superiority if I were you. > >>I think that it may be more interesting if there was a date when latest public >>Stobor could do better than latest public Crafty(not only in matches between it >>and crafty but also against other opponents). > >That'll be hard because there has never been a public version of Stobor. I've >given copies to people I trusted under special circumstances and it seems that >one of those got leaked somehow. I'm pretty disappointed about that. Who knows >what's going on with that version, but I wouldn't read much into its results. > >-Tom
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.