Author: Dann Corbit
Date: 12:30:32 06/17/05
Go up one level in this thread
On June 17, 2005 at 14:58:00, Christopher Conkie wrote: >Hello, >Thank you for all your answers, whoever you may be. I thought long and hard >about what I wanted to say. > >I should not like you all to think that my post was antagonisic as it is >something that I have a long held and very firm belief in. This post draws ire every year. People will spout off about it next year, and the year after that ad infinitum. It's a 'been there, done-that' thing. >There fore I will try to answer since I started this, all at once in one post. > >Lance > >>Its the best combination of many other things, including the best book > >Human games have no place in computer chess IMHO. If I want to see a human game >I would go and watch humans play. Let's remove all human knowledge from chess and play random moves then. After all, that is what it would mean. >Dann > >>If the goal of the contest is to try to find the combination that plays the >strongest chess, then it is well achieved by the current format. > >I think that the world championship should be about who made the best engine >only that performed in the competition under equal conditions. That cannot be achieved, and it cannot possibly be demonstrated in a contest of this kind. It is not the goal of the contest. You are welcome to create such a contest and I think it would draw some interest. >>At the end of the contest we literally have no idea which of the programs is >strongest (from the data supplied during the event). > >What a pity! I totally agree with this statement. It is true, we don't find out >and that is why I fear for the future of computer chess. Do you understand that with one weekend of chess your goal of finding out which PC engine is strongest is completely impossible? After thousands of SSDF chess games, we get an inkling of the answer. The SSDF answers exactly the question that you seek. Why try to make this contest do the same thing when it is clearly impossible to do it? >Charles > >>As far as results being "meaningless", this is not true. The point of it is >not to decide what people should go out and buy for their PC. The point is to >provide a competition that will all researchers in the field to test there >efforts. DarkThought ran on Alpha. We have had Cray supercomputers and dedicated hardware and Mac computers and mainfraimes and minis and unix boxes and PCs and all sorts of things. A PC program has won the contest during the last decade, despite the 100X hardware disadvantage in some of them. >Ask ChessBase if that is true. Not me. Why do you think they sponsor the event? >For fun? For money, of course. >>Also, these algorithms are quite applicable to other fields of science. I do >not see why you want to hold back progress in these fields inorder to see >which program is best on the PC at this single point in time. > >The ICCA ground to a screeching halt in 2003 and has done nothing much since for >computer chess. It pains me to say it as a Scotsman is in charge (or at least >has a major say in it) The ICCA has always been a pain in the posterior. It is inconsistent and also relatively uninvolved in the computer chess scene except for the magazine and this contest. They do not listen even to their membership -- including the most influential members. >>My guess is that you didn't understand the point of the tournament. > >Believe me, I do. I have my own firmly held beliefs and I have nothing to gain >from those beliefs, they are only honest observations. The contest you want does not show what you desire. What you are looking for is the SSDF results. >Mike > >To all of what you say, I say, fair enough but what a pity. No development takes >place where people have no morality left and money rules only. There are 10x more computer chess programmers now than there were ten years ago, at least. The innovations in the last 5 years top all those in the previous 15 years. >"Madhavan" >Most of what you say is to me is completely illogical and mostly rude. > >>Special appearance fees will also be given to the programmer by organisers >ICGA. > >How do you know this? Show me how you know this. I think it is false. But it does not matter one way or the other to me. >Matthew > >>That's OK. You have a right to learn nothing and be wrong. You have to admit, it was a great one-liner. If I had had coffee in my mouth it would have been all over the monitor. >True, it's my choice......man :-) > >Guy > >>'Equal hardware' events are no longer wanted by the competitors on the whole > >I don't how you know this and I disagree anyway. Did you ask everyone? It is pretty clearly a consensus. We talk about it every year until we are blue in the face. Most of the 'old-timers' are sick of it. >Mattheiu > >>What happen to the program which is program runs faster on a AMD? > >If the hardware were the same this would not be an issue. If you spent 1000 hours writing assembly and profiling for a particular platform and then you are told to use another one, it becomes an issue. >Peter > >>Not every strong program can take advantage of a parallel system > >Which is why I believe that there should be different classes for different >systems. We do after all have such a paralell in human chess tournaments after >all. Different classes has been done before. It's silly. >------------------------------------------------------------------------------ > >To sum up, there seems to be a need for a truly fair World Championship. Humans cannot achieve this. It is a laudable goal, of course. What we have now is quite fair. >One in >which the conditions are equal and therefore fair. Equal conditions does not mean fair conditions. Robert Hyatt has spent hundreds of hours writing a parallel program that is nicely debugged. Shall he throw that effort away so that toy programs can compete with it? What about all the other programmers that have toiled writing a parallel program and bought very expensive hardware to use in the contests. All for nothing? >One that is devoid of the >commercial considerations of a few, lining their pockets on supposed results of >their "manufactured" conditions. There is literally nothing wrong with the hardware and software conditions for either the WCCC or the WMCCC (do you know the difference?) >This is what I personally would set up. Somewhere where the worth of something >is appreciated and not somewhere where results are broadcasted as being "real" >when they are anything but real. Go for it. There is nothing to stop you from setting up a contest, gathering sponsors, and running it. If you are trying to discover which program is strongest, it will be a miserable failure, because the goal is literally impossible in one week's time. >Next you will all be telling me that the "World" Series take place between >baseball teams from all across the world. That would be fun to watch. BTW, the winner of the world series is not necesarily the best baseball team. >Only in North America could such a thing like this happen. If you tell >yourselves something for long enough I believe that eventually you will believe >that you are right. The WCCC is going to be in North America? That's great news. It sounds like the ICCA is finally going to listen to their own charter. >I want you to know that we love you even if you are misguided in your logic and >approach to the reward of the many hours that programmers put into their >engines. I do not think you understand the situation fully. I think that your approach is far worse for the programmers and most of them won't like it. >I will agree to differ. Let me know when you change your minds. > >I know that this is a nonsense and that the true World Champion (whoever that >is) will remain a mystery to those that do not think about fairness. 'World Champion' is a tag that gets assigned by a contest outcome, just like in Basketball or Soccer. It is not proof that the winner is the best team. It only means that a particular experiment had a given outcome and someone who is detached from the result stamps his golden seal of approval on it. >It is just as well that we know who has contributed most (secretly you know it >too) but that for now there is no proper environment for them to prove it. Is this some kind of secret message that needs a decoder ring? >Regards to you all > >Have a lovely (if ignorant) weekend Nanu-nanu. I'll have you know, I resemble that remark. >Christopher
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.