Author: Will Singleton
Date: 20:00:36 08/15/02
Go up one level in this thread
On August 15, 2002 at 21:43:14, Russell Reagan wrote: >On August 15, 2002 at 18:04:56, Will Singleton wrote: > >>With regard to one of Hyatt's statements, "The commercial >>programs are using some form of forward pruning that they won't discuss", I have >>seen hints of such in dicussions here. > >There have been vague hints, yes. But nothing meaningful. When I think of their >hints, it makes me think of a rocket scientist who discovered a way to travel to >Mars and back in a matter of minutes. When asked about how he did it, he >replied, "I made a space ship that travels at the speed of light." Gee, thanks. >That's not really helpful now is it? The commercial author's will give "hints" >like, "I use forward pruning". Neither the rocket scientist or the commercial >engine author's tell you how they accomplished what they did, only that they did >it, which isn't very helpful. Certainly not anywhere close to as helpful as >anything Bob has done. Then again, the fact that they are "commercial" means >they are in it for the money, not for educational purposes. It would be nice to >see them give something back though, since they surely learned something along >the way from a program like Crafty, or the other amateur engines out there. > >>But is it logical to assume that only >>the commercials are using secret techniques? > >Not *only* the commercials, but _only_ the commercials are using "big" secrets. >See below... > >>There are several strong amateur >>programs whose authors also do not discuss their techniques. > >While this is true that there are amateur engines that keep secrets, I seriously >doubt that any of those secrets are "big" secrets. I think it is 100% safe to >conclude that the commercial engines are keeping some "big" secrets. Probably >what Bob said, forward pruning methods. Here is my reasoning. > >1. Crafty makes use of techniques and methods that are well known, and in >addition to that, Bob has always been willing to explain how any part of Crafty >works. > >2. No amateur engine has drastically surpassed Crafty. Yes, some may be slightly >better, but they are still much closer to Crafty's playing level than they are >to the level of Fritz and friends. > >This leads me to believe that no amateur is using any "big" secret, or else they >wouldn't be on par with Crafty, which has no secrets. I think there might be >small secrets within some of the strong amateur engines, which would account for >the slight increase in playing ability compared to Crafty, at times. > >I think the fact that there are only one or two amateur engines that can even >score greater than 40% against the commercial engines some of the time says >something. I think it's a little niave to believe that the commercial engines >aren't keeping any secrets. > >>The commercials don't have a corner on secrecy, nor should they. > >They have a corner on the cutting edge secrets. IE the "big" secrets. > >Russell You make some good points, and I have no first-hand knowledge to refute them. So I can only guess, and my guess is that there are no "big" secrets that the commercials are using. They are just better at every aspect of what goes into making programs play well. Nevertheless, I also believe that both commercial and amateur authors are using some unknown techniques. Let's assume that I discovered some magic bullet pruning method, and it shrank my tree in half. Sure, it would play better, but the opening book would still suck, the crappy eval would still be there, and the code would still be riddled with all kinds of bugs. In short, it would still lose to the commercials. So, I don't believe one can assert that the fact no amateur program has surpassed Crafty by a significant margin is evidence that new techniques aren't being used by amateurs. You may well be right, but that argument isn't persuasive. Will
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.