We-all- can see today a true escalade of the strength for a division of chess engines. Twisted Logic 20080620,TerraPi,Bright 0.3d.......+ many commercial engines.....and near every week a new chess engine registered a big and suspect jump in ELO points ( sometimes after a long stagnation).
Original work or not ? This is the question !
Not good to see the same ideas exaggerated multiplied !
If the sources aren't available they are somebody to think we live today o true boom in chess programming ?
I would agree with you that most engines nowadays would not be strong without the open source engines like Fruit/Toga, Glaurung, Scorpio, Crafty, etc..
I have studied the source code of those engines carefully and I have used the ideas from them that I think would benefit my engine. Why reinvent the wheel. If an idea is good, why wouldn't one adapt it, isn't it?
Actually, I only used those ideas to come up with my own ideas. Twisted Logic has some ideas that can never be found on those open source engines, I just don't know if it can be found on the commercial ones.
Those ideas you are talking by the way is not new. It has been there for so long. Only that others have found new ways to improve on the existing idea. I would say almost all engines share the same idea, the difference is in the details.
I think it's true to an extent. A lot of programmers read the source to popular open source engines and take too much out of them. But I think this should only be an argument against closed source. In a completely open source world, I think there would be less emphasis on pure strength and more focus on originality, as paradoxical as it would seem. There's very few strong open source engines, though that number is improving. But all the strong ones don't have that much variety. Crafty and Glaurung are pretty different, but they aren't as different as say, Rebel and Fruit. There aren't that many approaches. There's the super optimized Crafty, the very simple and bug-free Fruit and Glaurung, and that's about it. Where's the super selective and knowledgable programs? Where's the programs that don't use alpha-beta and null move and PVS?
And as a note, I doubt Twisted Logic and Bright are clones in any way, but I am very sure that Alaric is not. Peter's been involved in CC for years, and Alaric is the result of a strong gradual progression in strength.
Hehe... That's true. I seem to remember them having negative Elo though? I don't deny their importance, but I don't know how useful their algorithms will be. I would like to see someone take those concepts and make a competitive chess program. Not necessarily a top-ten, but >2000 or so.
Zach Wegner wrote:Hehe... That's true. I seem to remember them having negative Elo though? I don't deny their importance, but I don't know how useful their algorithms will be. I would like to see someone take those concepts and make a competitive chess program. Not necessarily a top-ten, but >2000 or so.
True, its because of them that I wont calculate my rating list with anything less than offset 2300. If I do, then they get a negetive rating, which I think might disappoint the programmer.
Zach Wegner wrote: Where's the programs that don't use alpha-beta and null move and PVS?
Just an example: Comet by Uli Türke (mtd(f))
Best,
Daniel
Yeah,but unfortunately Uli is not interested in computer chess anymore
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
