For many years now the top name in computer chess has been Rybka. The program has previously swept all before it and it has won many, many titles. However, there has always been a certain amount of controversy surrounding Rybka and it looks as though that controversy is about to get ugly. According to the Washington Times, Rybka and Mr Vasik Rajilich have been stripped of their ICGA (International Computer Games Association) World Computer Chess Champion Titles, with Mr Rajilich receiving a lifetime ban from further competition.
From the early days of the Rybka 1.0 beta, computer chess programmers have had concerns about the origins of the Rybka code. Was it an original work of Mister Rajilich or did the code originate elsewhere? Rumours abounded, but with very few of the community having the technical skills to examine the code and arrive at an informed opinion, the debate was often characterised more by emotion rather than logic and reasoned argument or production of factual evidence.
During those times some fans of the program at times behaved in a manner more reminiscent of a fundamentalist religious movement than rational human beings and logical(?) chess players. Chess forums banned discussion of rival programs and posters merely mentioning the rivals names had their accounts banned and posts deleted. It was a period very reminiscent of Stalin’s Russia in many ways, except that the violence was limited to character assassination rather than live ammunition.
The suppression of dissent over the origins of Rybka was one of the reasons that this website was founded. I found the stifling of debate and lack of real effort to find the truth depressing. The repression of debate is never a good thing. The lunatic behaviour of some moderators on a number of chess forums ensured that the ability of the community as a whole to arrive at the truth of the matter was severely hampered.
It was always an object of curiousity to me that Mr Rajilich never sought to prosecute those behind any of the several Rybka clones that were released over the last year or two. Surely a man that has worked hard to produce an original piece of intellectual property that was a world leader in its class would act to protect his product? Now it seems that we have been provided with a possible reason why. David Levy has accused Rybka itself of being the plagiarised work of two separate computer chess programs – Crafty and Fruit. If the accusations prove to be correct then this would, of course, provide ample reason to avoid any public court case involving Rybka. After all, how can you prosecute someone for stealing your intellectual property if your product is itself plagiarised?
As Houdini is currently leading the rankings in terms of playing strength (Elo) the point is somewhat moot anyway, except for one thing: Many assert that Houdini itself is merely a Rybka clone. If this is also true and Houdini gets banned too, then where do the folks that want to play chess versus computer opponents go then?
Original article: Washington Times.
robert babakhani says
play chess
MCris says
Yes, I’ve read on the Internet a lot about Rybka forgery. There were reputable persons there, like Prof. Hyatt (Crafty), and others. The ban seems ok, only that a certain well-known site wishes to sell Rybka further, along with their program. This is not right. And the author of Fruit does not want to sue, as I know. So, lucky Rajilich, you got money and fame, but we know who you are!
Jim C says
I personally don’t believe it. Rybka is so far beyond Crafty and Fruit in its strength that I want to see the code that they are claiming has been stolen. All I’ve heard is vagaries and it doesn’t surprise me that these people making programs want to slay the giant. I own several programs and have run many chess engine tournaments. Rybka is always the winner by a landslide, and the few times in the beginning that I included Crafty and Fruit Rybka crushed them within thirty moves. Whereas Deep Fritz and Hiarcs (single-core Hiarcs unfortunately) are able to go the distance before resigning. There are a few draws, but only rarely would any of those mentioned engines actually win against Rybka. Ver. 3 in my case.
I have just upgraded my computer to eight cores at 4.2 ghz apiece and have been looking for a way to test its strength in the wild and I am amazed at how many people are so eager to ostracize Mr. Rajilich w/o the benefit of a trial. It’s been nothing more than a witch hunt. In fact the way that this whole mess has played out, I would be more suspicious of bribery on the part of the established engines to protect their market share by creating a slur campaign against Rybka. That is not truly an accusation but merely a possibility that would not surprise me one bit. Had there been a true scientific examination and open legal proceeding, we could put this issue to some rest. But the fact that I haven’t seen any impartial adjudication clouds the legitimacy of the accusations laid upon Mr. Rajilich.
Dave says
Hi Jim,
Let me first of all say that I respect the achievement of Vaz and others in building Rybka. I really do. His work spurred others on to do better and so whatever the true origins of Rybka, he helped take computer chess forward.
As to being the victim, Vaz had the chance to take it to court and chose not to. Some say this is because he knew he could not win the case. I guess we will never know the real reasons why.
Any “slur campaign” that occurred, actually happened on some of the public chess forums at the time, but it was hardly Vaz that was on the receiving end. Quite the opposite.
If someone dared to point out the code similarities between the engines and suggest an investigation they would be verbally abused by fan boys of Rybka. If the person suggesting an honest investigation responded in kind, the moderators would ban their account. Shortly thereafter the thread would usually be censored by being later removed from the forum. At one point it got so bad that certain public chess forums banned all discussion of Rybka and the numerous clones.
The suppression of discussion and repeated attacks on anyone that wanted to investigate the matter was truly Stalinist. It sickened me of public chess forums to such an extent I have stayed off them since. I needed some fresh air.
I noticed that you mentioned “impartial adjudication”. Did you follow the link in the article? The information comes from the Washington Times.
I draw your attention to this quote:
http://m.washingtontimes.com/news/2011/jun/30/computer-chess-champ-stripped-of-its-four-titles/
You suggested that Rybka was the strongest engine in your tests “by a landslide” I am happy that you believe Rybka 3.0 to be so good, but I would invite you to look at the test results of CCRL, CEGT or IPON. All three independent testing sites show that other engines are stronger, most notably Houdini.
If you are looking for an engine to beat Rybka 3.0 reliably I suggest looking at Stockfish (free and open source).
Best wishes.