Author: Don Dailey
Date: 09:48:53 05/22/98
Go up one level in this thread
To Thorsten Czub: On May 21, 1998 at 21:54:07, Thorsten Czub wrote: >On May 21, 1998 at 19:40:07, Mark Young wrote: >>Yes Junior, Fritz 5, and Nimzo 98 are fast searchers. > >Right. > > >> And one could >>argue from there results that these three programs are the top programs >>out right now. > >Maybe right. > > >> I think what Junior, Fritz 5, and Nimzo 98 have shown >>that its not amount of knowledege that important. Its understanding what >>positional knowledege that the program does not need. I agree 100% with this statement. >Why ? I don't see that these 3 programs are much stronger than >Hiarcs or Mchess. I do not believe that these 3 programs understand >chess better than Hiarcs or Mchess. Do you ? I do. If they are winning they understand chess better. It might very well be that we need to modify our view of what understanding chess is. You seem to strongly believe that tactics are not a part of chess. I have seen the phenomenon in people, where certain "clumbsy" players seemed to get better results than they deserved. They didn't know thematic plans, didn't understand opening theory etc. but they got good results compared to much "better players." But they did more thing right than the apparently more knowledgable players. They fought hard, didn't make errors and were good tacticians. So I ask you, who really understood chess better? But I even dispute your belief that Fritz and Nimzo are dumb programs. As I've observed, Fritz seems to come up with consistantly good moves, otherwise how could it possibly win unless you admit these other programs are not "smart" enough to deal with this. >>In my mind it >>takes more skill to write a program in this way. I even disagree here (since I'm being disagreeable today.) The magic of programs like Fritz is the great engineering involved in making such a fast program so smart. Do you think this is easy? Focusing on positional play and neglecting tactics is just plain foolish, chess is a highly tactical game. You have to conform to what works if you want ultimate chess strength, not what pleases you. I think the burden of proof is on you. So far, we've seen that both approaches are very reasonable, there are good slow (but not too slow) programs, but the fast approach is currently working the best. >I think here you are wrong. I know that I will now hurt some people, but >I don't believe writing a fast-searcher program is very difficult >attempt. If this is true, why should I pick the more difficult approach when it will give me a weaker program? This doesn't make sense to me. >I think it was Chrilly himself who said something like this. >Oh no - it was - now I remember it accurate, it was Dieter Steinwender. > >>This leaner code gives >>the program its faster search rate. Giving the program greater search >>depth. With the greater search depth I think the programs sees over the >>board what positional has to be done. > >How can it see what positional stuff has to be done when its knowledge >about positional is very rudimental ? It's called SEARCH, which is a critical skill all chess players must possess. What you do not realize is that search significantly enhances knowledge. You pretend it's completed isolated. >Also - does it even HAVE knoweldge in deep lines ? >Or does it only have piece-square tables from the root ? > >>This may give Junior, Fritz 5, and >>Nimzo 98 a better more flexable positional understanding. > >Big words. How can you prove them ? >I have not seen Fritz showing positional understanding. It must make winning moves purely by accident. How does it beat programs that understand more? Again, you are not taking a wholistic view of things, but pretending chess involves only a subset of skills that YOU consider important. Why don't you just come clean, recognize the facts and then try to learn something from them? Obviously, Fritz is doing SOMETHING right, and whatever it is wins chess games. You seem so very close minded about how things should be done. I have no problem with the approach you advocate, I'm completely open to it or anything that work, or look like it has a chance of working. Like your approach. Why can't you be equally open to other viewpoints? You will notice that Cilkchess is not fast, at least when you consider the hardware. I use what I consider a balanced approach of speed and knowledge. I'm willing to consider that I need more knowledge, and I also consider that I may just need more speed. But there is one thing I know is a sure thing, and it's that more speed will ALWAYS help. In my mind the real trick is getting as much well balanced knowledge in a program with the very least speed penalty. You cannot admit that Fritz and Nimzo have done an amazingly good job at this. Why do you berate them for this? >Maybe Junior, sometimes Nimzo. But Fritz5 ? > > blah blah blah ... - Don
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.