Sunday, January 8, 2012


Very recently "Jason" commented on my comments about the "Cheating Rybka Scandal" and mildly taking me to task for not exonerating Rajlich the way Riis did in parts 3 and 4 of the ChessBase article on this mess.

I can't respond part by part because Google doesn't give me the email addresses of those who comment and he (Jason) didn't leave one. That's not all bad because I certainly wouldn't have the time to answer everyone. But I go back to my roots of asking a "reviewer" who panned a book I published to let me answer his comments. Which he wouldn't do. In fact, there were two such people.

When I and a NM who helped me with the book (and the author) responded to the critiques with a point by point rebuttal we couldn't get any traction because "they" were better known, and surprisingly, bigger a-------! There were bigger issues and they wouldn't deal with those instead taking a book they didn't like and in effect, making up complaints (trying to make it look like they were being "fair" when in reality they weren't at all) that held no water. So, Jason has his say and he asks if I would have a change of heart.

In all honesty, I lightly skimmed articles 3 and 4 because they were too long and I would come across points which I felt Riis still wasn't answering in full (but I didn't want to make the article but even longer!) He deluged us with "data." An old lawyer trick.

I invite everyone who is interested to read articles 1-4 in full (not me, too much to consume) and form their own thoughts and reasonings. It was my understanding that initially Rajlich did admit to taking some Fruit code and now Riis says "no" and then adds, and what if he did, "it wasn't important?" Doesn't that call into question two things:
1) Who reported he admitted doing it initially? Was it true?
2) If it wasn't important, then why was it taken in the first place?

And I addressed a question to the panel, which I never saw covered: In effect, "If Fruit was so important, why wasn't it a better and more well known program than Rybka?" Basically, Rybka must've been quantum leaps better at something to maintain it's degree of "sophistication." What was that "something?"

And of course my initial question, way back, "Why was ChessBase saying nothing until now?"

Being the commercialist that I am, I suspect there is a newer Rybka version about to spring forth. Secondly, the money and "prestige" that Rajlich got initially might have been flattering, but if I had "invented" Rybka, I would care little about the Computer organization (just another damn organization of which we don't need more), the small amounts of monetary awards, and instead be more concerned about universal sales and royalties.

Maybe Rybka isn't as good as everyone thinks it is because it works with multiple CPUs (clusters) and the like... but since most of us can't afford all that and are concerned about what works well on OUR duo-core or quad-core computers (PCs) that should have been addressed too.

Personally, my #1 choice for data analysis is Deep Rybka 4. But when I don't like the answers I see, I turn on Fritz or Shredder.

No comments:

Post a Comment