Author: Peter Hegger
Date: 10:20:04 02/05/03
Go up one level in this thread
On February 05, 2003 at 11:45:09, Rolf Tueschen wrote: >On February 04, 2003 at 20:42:54, Peter Hegger wrote: > >>On February 04, 2003 at 13:33:17, Rolf Tueschen wrote: >> >>>On February 04, 2003 at 12:38:07, Peter Hegger wrote: >>> >>>>On February 04, 2003 at 11:35:55, Rolf Tueschen wrote: >>>> >>>>(much snippage) >>>> >>>>>4.) in a defined period of time, say half a year, the progs are forbidden to be >>>>>changed; new games into book are allowed, techno bugs are allowed to be >>>>>corrected; but the chess system of the engine version jan-june is constant; >>>>>books are allowed but without lines no computer ever could solve actually; >>>>>tables must be discussed by the nasters themselves and possibly forbidden and >>>>>reduced or such! >>>>> >>>>>Rolf Tueschen >>>> >>>>Rolf, >>>>Yes, this would definetely give the poor human a much better chance to compete. >>>>But don't you think this is stacking the deck in favour of the humans a little >>>>too much? >>>>I, personally, could never beat my car in a race if it were set up and running >>>>properly. However, if I pulled off a few spark plug wires, let the air out of >>>>the tires and drained out the oil, I would then stand an excellent chance of >>>>defeating it. >>>>This is basically what you want to do to the programs. You want to weaken it, >>>>like I would weaken my car and then when it loses you can say " see, I told you >>>>that the programs are not that strong". >>>>Would that really be fair? >>>>Regards, >>>>Peter >>> >>>No, I don't see the analogy here. You reason as if the prog should simply play >>>chess as it is. And I am saying that we need rules for human-comp competition. >>>You might be right that today we would get advantage for humans, but that is >>>also exactly what is the truth, if you believe experts, but I wanted to give a >>>general approach also for the next couple of years. It will become more and more >>>difficult. I don't get what you mean with advantages. You insinuate that it's >>>unfair if we would play like this? So fair is only fair if we let the computer >>>side do what they want? Peter, I know that a constant identity is a problem. >>>Computerchess programmers do want hourly tweakings. But I fear you all oversee >>>what tweaking means for human players. They must feel like secondary garbage >>>just being used for the glory of CC and the machine. >>> >>>Peter, just forget it, let's skip the tournament chess of humans and machines >>>and continue with the hoax of show events. Kill your own future. Do what you >>>want. I won't stop you. >>> >>>Rolf Tueschen >> >>Rolf, >>Let me approach the problem from a different angle. >>Would you accept the following proposition? I'll not tweak or fiddle with my >>program for 6 months before the GM-comp. match as you suggest above. We'll also >>get a committee of masters together to determine which opening book can be used. >>If the computer doesn't understand the resulting position of a certain opening >>then that position will be stricken from the book as you also suggest. >>In return, however, we must also level the playing field. The human master must >>also sit in front of the same masters committee and explain each opening he/she >>intends to use. If the committee feels that the GM isn't 100% certain on >>resulting positions then he/she will also be restricted from using that opening. >>Also, in all fairness, the human GM will not be allowed to play, study or open a >>chess book for the same six months that the computer can't. >>Do you now see just how absurd your suggestions for computer restrictions are? >>If its not good for the goose then its not good for the gander either. >>Regards, >>Peter > > >Peter, >you don't understand me! Read above! I do NOT talk about show events. I mean >tournament chess events! Several for these 6 months! And I allow that you twist >you book, you can define openings prog should play in tournament 1 and not in >tournament 2. Do you get this now? Rolf, I've understood your basic premise all along. But, alas, it appears that we will have to agree to disagree on alot of points. What you think is fair and what I think is fair and what you think is show and what I think is show are obviously two different things. I still stand by my basic assertion that programs are now playing 2700-2800 chess. I also fear that the only way to put this discussion to rest is by bribing the super GM with lots and lots money for winnning games and by penalizing him/her lots and lots of money for drawing or losing to programs. This could be done in a tournament or a match setting. Offer $10,000 in the tournament setting for anyone who can beat the program(s), $2,000 for a draw and $0 for a loss. Would that be incentive enough to produce "fierce" chess and not mere show? In a match seeting against, say Kasparov, what we do is set up an absolute World Championship match between him and ProgramX for all the marbles. Have a one million dollar prize fund divided in the following way. 1 million to the winner for a 10 game match win, $100,000 to each for a drawn match and $0 for losing. Now, if that's not enough incentive I don't know what is. >The ONLY think I want to exclude is that > >1) you fiddle between single games, because I think chess is variable enough I could accept this in the above match. >2) you fiddle with openings where NOT the final positions are BAD for comp or >couldn't be understood! No, I am talking about openings whose positions these >programs actually could NOT understand and where they would play GARBAGE if the >book would not be there! To be clear, I want to exclude lines who might well >lead to a equal position in the end but who contain positions where a comp would >commit suicide - IF without book! I don't agree with this though. Do we also tell Kasparov he can't play the Sicilian any more because he didn't understand some resulting positions and lost the other day? > >Peter, please ask all the questions you like. Until you have understood what I >meant with my main point. Otherwise the debate is moot. > >Rolf Tueschen Now its off to the restaurant. All this Hamburger talk lately is making me hungry. Here I can get a juicy flame broiled cheeseburger for $1.59 on Wednesdays. I'll just pretend its a $40 one while I'm eating it. Regards, Peter
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.