Author: Robert Hyatt
Date: 18:14:46 11/10/02
Go up one level in this thread
On November 10, 2002 at 04:59:13, Gian-Carlo Pascutto wrote: >On November 09, 2002 at 22:49:13, Robert Hyatt wrote: > >>I'm simply paraphrasing what you said in an email many months ago. >> >>Remember the discussion where you said "my king safety is ridiculous >>and needs to be redone. So that you could get something 'better'?" > >I've been posting here since I was 17 years old and knew very little >about chessprogramming. If you're going to cherrypick pick quotes >from me from the past, you'd have no problem making me look like a >fool, a moron, a total asshole, a thief, and probably a few more things. I don't think I did _any_ of those. Otherwise Bruce did the same thing to himself when he mentioned that he used the original gnuchess 4.0 move generator and later found an improvement for it. Ir I made a fool of myself, because I have _always_ said that my first brute- force attempt came from the article by Slate/Atkin in "Chess skill in man and machine." Stealing code is one thing. Using ideas is another. I don't see a thing wrong with your using ideas from my code, and if you will recall, I mentioned that that was _exactly_ the reason I decided to release the source back in 1995... The only comment I made was that I would _obviously_ have some trouble with your program, since you have looked at what I do, recently, while I haven't seen anything you have done in a long while. I don't see any big deal in that. I put myself in that position knowingly, so... > >If I said that, I was joking. I am very happy with Sjengs current >kingsafety code, thank you very much. And yes, I still think crafties >is no good, but I know that you'll never agree with me on that one :) Not until I see it fail a lot, you are right. :) > >>I didn't make that up. You or someone sent me some sjeng eval code >>a good while back, and the blocked pawn and majority and so forth looked >>identical, as an example, although you have to implement them differently >>without all the bitmap stuff... > >I am positive I never sent you pieces of my eval code (unless they were >trivial). Only Dann Corbit has another copy of the source code, and I'm >farily sure he isn't sending it around either. > >Almost exactly one year ago I emailed you to ask whether you had any >objections to me 'stealing' the idea of your candiate majority system >(I already had outside passers back then). I ended up *debugging* >the relevant code in *crafty* because it didn't work. While doing so, >I sent you several snippets of *crafty* code with corrections, before >deciding that I could probably do much better myself. > >-- >GCP Possibly true. I get "fixes" all the time from all over... :) However, I don't remember any code that "didn't work". I don't generally write code that "doesn't work". I certainly might have a bug here and there, or a special case that causes problems. But "didn't work"???
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.