Author: Robert Hyatt
Date: 15:02:45 11/21/03
Go up one level in this thread
On November 21, 2003 at 12:52:37, Peter Kappler wrote: >On November 21, 2003 at 10:30:21, Robert Hyatt wrote: > >>On November 21, 2003 at 05:59:05, Amir Ban wrote: >> >>>On November 20, 2003 at 23:23:16, Robert Hyatt wrote: >>> >>>>On November 20, 2003 at 14:23:10, Amir Ban wrote: >>>> >>>>>On November 20, 2003 at 08:59:25, Matthew Hull wrote: >>>>> >>>>>>On November 20, 2003 at 06:57:30, Amir Ban wrote: >>>>>> >>>>>>>On November 19, 2003 at 18:12:12, Matthew Hull wrote: >>>>>>> >>>>>>>>On November 19, 2003 at 17:30:36, Amir Ban wrote: >>>>>>>> >>>>>>>>>On November 19, 2003 at 12:02:56, Matthew Hull wrote: >>>>>>>>> >>>>>>>>>>On November 19, 2003 at 11:51:59, martin fierz wrote: >>>>>>>>>> >>>>>>>>>>>On November 19, 2003 at 11:34:17, Matthew Hull wrote: >>>>>>>>>>> >>>>>>>>>>>>On November 19, 2003 at 11:30:37, martin fierz wrote: >>>>>>>>>>>> >>>>>>>>>>>>>On November 19, 2003 at 11:06:21, Matthew Hull wrote: >>>>>>>>>>>>> >>>>>>>>>>>>>>On November 19, 2003 at 10:55:26, martin fierz wrote: >>>>>>>>>>>>>> >>>>>>>>>>>>>>>On November 19, 2003 at 10:31:54, Robert Hyatt wrote: >>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>here. Makes a _lot_ of sense. And it shows just how "world" aware the >>>>>>>>>>>>>>>>ICCA actually is. >>>>>>>>>>>>>>> >>>>>>>>>>>>>>>i don't really want to be involved in this thread, but i can't resist this >>>>>>>>>>>>>>>one... >>>>>>>>>>>>>>> >>>>>>>>>>>>>>>disclaimer: of course it would be much more sensible to have the championship in >>>>>>>>>>>>>>>the US from time to time. >>>>>>>>>>>>>>> >>>>>>>>>>>>>>>cheapo: so the ICCA does something which is not good for *one* country >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>>That's one cheapo that doesn't work. It would be like 2000 years ago holding >>>>>>>>>>>>>>gladiator events that discommode only one country, Rome. >>>>>>>>>>>>>> >>>>>>>>>>>>>>MH >>>>>>>>>>>>> >>>>>>>>>>>>>of course it works, and you just invite the next follow up cheapo ;-) >>>>>>>>>>>>> >>>>>>>>>>>>>2000 years ago the romans were perhaps not aware that there was much more to the >>>>>>>>>>>>>world than rome. sometimes one gets the feeling that the US citizens are no >>>>>>>>>>>>>different in this respect... >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>>Ok, how about holding a world chess championship that only inconviences >>>>>>>>>>>>Russians. >>>>>>>>>>>> >>>>>>>>>>>>I think you get the idea. :) >>>>>>>>>>>> >>>>>>>>>>>>MH >>>>>>>>>>> >>>>>>>>>>>of course i get the idea! i put a disclaimer on my first post stating clearly >>>>>>>>>>>that IMO the championship should be held in the US from time to time, and i >>>>>>>>>>>labelled my posts as cheapos :-) >>>>>>>>>>>i thought that made it clear enough... >>>>>>>>>>> >>>>>>>>>>>going back to your comparison with the russians: exactly how many american >>>>>>>>>>>programs are in the top 10 of the SSDF list? >>>>>>>>>> >>>>>>>>>> >>>>>>>>>>The SSDF list only uses consumer-grade technology to test programs. Programs >>>>>>>>>>tuned to that limited technology will always top that list. That is why the >>>>>>>>>>list is of limited importance. A real WCCC is going to attract high performance >>>>>>>>>>projects, not just consumer oriented projects. This is what the New World has >>>>>>>>>>always offered. But, Old Worlders have a problem with that I guess. >>>>>>>>>> >>>>>>>>> >>>>>>>>>Do any such New World high performance projects exist ? >>>>>>>> >>>>>>>> >>>>>>>>Crafty can be such a project on practically a moment's notice (I believe). >>>>>>>>Other programs are similiarly suited. If the WCCC comes to North America, the >>>>>>>>projects will materialize. This was the benefit of limiting the event to every >>>>>>>>three years and making it a practical event, length-wise. It provided time for >>>>>>>>the husbanding of resources, planning, development and sponsorship along with a >>>>>>>>relative rarity that made the event that much more important and compelling (and >>>>>>>>thus an easier sell to the people with the expensive resources). >>>>>>>> >>>>>>>>The current cycle with it's awkward timing and extended length, along with it's >>>>>>>>persistent location in Europe (not to mention its archaic modus operendi) seems >>>>>>>>calculated to favor European commercial interests while excluding projects from >>>>>>>>North America. >>>>>>>> >>>>>>>>Perhaps it is the punishment Europeans are determined to mete out to us for the >>>>>>>>DB2 triumph, which seems to be universally reviled overseas. EU types are maybe >>>>>>>>fed up with the dominance of North American, high-end computer chess projects. >>>>>>>> >>>>>>> >>>>>>>There's nothing to be fed up with, since the dominance is long gone. >>>>>> >>>>>>Yes, the ICGA have seen to that by keeping the WCCC out of North America and >>>>>>making inconvenient for North Americans to participate. Nicely done, IMO. >>>>>> >>>>>>>Hong Kong >>>>>>>1995 was the swansong. There were 4 of them there, but losing to Fritz, and even >>>>>>>before that, in 1992, to Schroeder, underscored that they have lost their >>>>>>>advantage and so their reason in life. >>>>>> >>>>>> >>>>>>That is a not entirely unreasonable opinion, though still incorrect, IMO. Bob >>>>>>addressed the competitive issue in another thread here. There are American >>>>>>programs suited to high performance hardware which would have a definite >>>>>>advantage, even over your project. Yes? >>>>>> >>>>> >>>>>Sure. There are tens if not hundreds of Americans who would make me look silly >>>>>with multi-million $ projects and $10 million hardware. The only thing holding >>>>>them back is that they can't afford to go to Europe. >>>>> >>>>>It has been tested once in a Rebel vs. Crafty match where Crafty was given a 100 >>>>>to 1 time advantage. The match was aborted after Rebel won the first game. >>>>> >>>> >>>>How about doing a couple of things: >>>> >>>>(1) tell the entire story. (a) one game doesn't mean _anything_. (b) Ed >>>>played multiple games with crafty and rebel having a _very_ long time for >>>>each move. Crafty won. Does that prove anything? Nope, other than the >>>>one handicap game was meaningless. >>>> >>>>(2) I'll be _happy_ to take you on at 100:1 time odds, anything you think >>>>you are ready. I'll even put up a wager to make it interesting. I am >>>>talking about a match of at least 10 games. Interested? You'd be >>>>stupid if you were. Because I wouldn't play _any_ program at that time >>>>handicap, including the original Sargon... >>>> >>> >>>You're on. Please suggest format and let's discuss after WCCC. >>> >>>Amir >>> >> >>Any format is fine by me. 100 minutes to 1. 6000 minutes to 60. Totally >>irrelevant to me. I don't have to guess on the outcome here. I don't >>believe your program is _that_ much more knowledgable than mine. It is >>certainly better tactically. But 100:1 is going to eliminate that tactical >>advantage totally and swing it the other way.. You had better have a _bunch_ >>of chess knowledge I don't have to beat me. I don't believe _any_ program >>has that big an advantage today... >> > > >I assume this will be a ponder=off match? > >-Peter The normal set-up for such a match would be the following: 1. ponder=off, obviously. 2. If a program leaves book with a bad score (< -1.00) then that game is aborted, to eliminate bad book lines from influencing a match that is not about books at all. 3. There probably should be two players for each side. IE I could run Crafty, but then someone Amir chooses _also_ runs the same version to verify that the program is playing as I say it plays. Ditto for him, he can run his own or choose a proxy, I choose a second person to run his program to show that the computer is playing as expected. 4. We run Crafty on whatever hardware he chooses for Junior, then we run it on whatever hardware (non-SMP) we choose for running Crafty. Then we give Crafty enough time on the hardware we use, so that it searches 100X as long as it would on the machine Junior uses.. IE my hardware might start out being 2x faster, so the time handicap is 50x, to make the difference between the two programs exactly 100x. Other than that, it is just "open 'em up and let 'em dance." > > > >> >> >> >>> >>>> >>>>> >>>>>>But that's not good for business, ist it? It looks to me that the status quo >>>>>>favors your interests. >>>>>> >>>>> >>>>>So it's the money motive working here ? This would be an object lesson on how to >>>>>bring industry giants and ivy-league colleges to their knees: make them travel, >>>>>or make them get a $50,000 sponsor. >>>>> >>>>>Amir >>>>> >>>> >>>>It is _several_ things. The biggest is that we have an organization >>>>that was formed with the sole purpose of fostering interest in computer >>>>chess "around the world". It is no longer living up to that charter. >>>>It is now fostering computer chess interest in Europe, mainly. Which >>>>is fine. I've already re-named it to the ECCI or ECGA, which is much >>>>more descriptive... >>>> >>>>It is easily possible to get a company to provide hardware, and some >>>>publicity money, and even some prize money. But not a big chunk of >>>>change that goes into a black hole called the JICGA, which won't >>>>benefit the donor whatsoever... >>>> >>>>If you tax someone too much, they move away. >>>> >>>> >>>>> >>>>>>Matt >>>>>> >>>>>> >>>>>>> >>>>>>>To remind you, the current world champion is not European. >>>>>>> >>>>>>>Amir
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.