Author: Sune Fischer
Date: 05:19:41 01/28/04
Go up one level in this thread
On January 28, 2004 at 07:15:47, Mogens Larsen wrote: >On January 27, 2004 at 14:49:12, Thomas Mayer wrote: > >>Hi Mogens, > >Hey Thomas, > >>Again, it was never claimed as a clone -> the claim was only that it might have >>some Crafty in it - which is a) not allowed due to the rules of the ICGA and b) >>not allowed to the Crafty license itself (you may take a look in the Crafty >>Source Code). And as far as I know they did test some positions in a one ply >>search... (Where the search itself does not play such a big role - mainly the >>eval decides about the move choosed in a one ply search...) - and the version of >>List they did test showed almost always the same score and move... > >That sounds like some of the worst nonsense I've heard in a long time. The >evidence was hearsay and a few one ply searches. And that is all you can hope to get if you don't have the code and the clone has been altered slightly. It will simply have to suffice. >In that case it wasn't game >related and could have been determined before start, not as an afterthought >during the championship. I agree that would have been the best option by far, but better late than never. It would be a good idea to have a kind of screening process before the start for all new participants in the future. >>And about those facts given here in the forum -> for every programmer I did talk >>with, those shown facts would have been enough that the Author should be asked >>what is going on... > >Actually, there haven't been any facts so far, only supposition. If you have a better suggestion on how to spot clones, please speak up :) >Would it have >been enough evidence to inspect the source code of a commercial program? Not in >a million years. I don't understand the question, or rather the answer you give to it. >>I doubt it... I still think that most programmers are honest people. At the >>moment I feel only suspicious about two of the better (better means better then >>Gerbil or TSCP) amateur engines... > >It's not an accusation. But fact remains that very few start a 100% percent from >scratch. The endresult - a few versions later - will probably bear very little >resemblance to its origins and therefore perfectly legitimate. However, I >suspect that quite a few engines contain bits and pieces. Take a huge program like Crafty, that's about 50,000 lines of code. Now you change 5,000 lines of code and let's say you improve the program by 50 Elo. The printed output will look completely different, and it won't be easy to see that this originated as Crafty at all. However, the program is still 90% Crafty and although improving it by 50 Elo is impressive, it doesn't really change anything clone-wise. >Your last comment suggests why rules of semi-judicial reviews should have >rigerous rules to prevent sabotage, envy and general incompetence by ICGA and >competitors. Not, "We the people have a few questions based on one ply searches. >Cough up with your secrets or be banned." What about the author whos rights may be getting violated? If the community doesn't try and clens itself from cloners we will soon see 20 "new" Crafty strength engines a year. >>again, the given facts were enough that the ICGA was forced to ask Mr. Reul what >>is going on... Nearly every programmer agrees to that. > >Sure, it's a competitor out of the way. The cynical viewpoint hereby expressed. That's a theory :) >I have no idea if Fritz Reul is guilty or not. But he had every right to refuse >examination of his engine based on the evidence you've given so far. Yes he can refuse to do so, and if he does he will disqualified which is what happened. I would also refuse to send my code to any member of the ICGA :) I could be persuaded to let them quickly glance over it for 20 seconds though, that should be enough. >And now >some regard the engine with suspicion unfairly, which was my original approach >to this thread. I'm not sure about the authenticity of many of the strong amateur engines that have evolved lately. Until an unpartial party, that I trust to make the judgement, has ensured me it is original I will have my doubts. Frankly, given the history of computer chess I would consider it a bit naive to not have at least some doubts. >Based on the current discussions I don't see any need to feel >ashamed about downloading List. I've done it, even if I really hate "UCI-only" >engines ;-). You shouldn't be ashamed, I just wish I knew whether you're downloading a modified Crafty a piece of original work. :) -S. >Regards, >Mogens
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.