We are in agreement there.Prima wrote:By your own assessment here, don't you see this applicable in the Fruit-Rybka issue? The same also applies to Rybka 1.0 with regards to Fruit 2.1: Ideas don't result in _identical_ output in a machine with the complexity of a chess engine. Nothing but code does.Jeremy Bernstein wrote:Ideas don't result in _identical_ output in a machine with the complexity of a chess engine. Nothing but code does.
Let's, for one moment, ignore the evidence of Rybka as a Fruit clone provided by the professionals. The damning evidence also came from Vas himself; when he claimed Strelka 2.0 as his own source code. Strelka was and has recently been proven and recently re-examined to have lots of Fruit 2.1 code in it. Since Vas denied using Fruit code in Rybka, stating, and I quote hereHow is Strelka then related to Rybka that supposedly don't have Fruit codes in it?"100% original at the source code level..."
Jeremey, no matter how one (not you in particular) bends this against Ippolits & co., the same logic and predicaments applies to Rybka in a direct fashion.
However, Vas' statement about Strelka was probably not accurate, and he later retracted (whether he did so because he realized that he had inadvertently shown his hand, or because he realized that he was over-reaching, or because he realized that he was wrong isn't clear). That is, Strelka wasn't Rybka, although it apparently had a lot of Rybka in it -- it was a Frankenstein's monster of disassembled Rybka + Fruit.
Jury Osipov claimed that it was easy to get the parts working that he couldn't figure out by patching them with code chunks from Fruit. And claimed that it was easy, because that's how Rybka itself was made. However, the truth is that we need to completely ignore Strelka for the time being, because it's irrelevant to determining whether Rybka is derived from Fruit. Strelka provided an important initial indication, but it's actually a distraction from any real comparison of Rybka and Fruit.
I'm interested in the real comparison of R&F, such as the document published by BB+ the other day, and that document is quite convincing, although I haven't read it closely, and I haven't confirmed any of the data in it (nor can I, in some cases, without getting up to speed in assembly language). But for the same reason that I am willing to give Ippolit the benefit of the doubt, I am willing to give it to Rybka until I can be convinced otherwise. I agree 100% that it's unfair that there are people like Banks or Williamson, who "take the law into their own hands" and ban and censor in the one case and not in the other -- that's clearly a double standard. I don't, though, and I'm the only one I have influence over.
Back to the theme at hand: my issue with Houdini is not that it's based on Ippolit. More power to it. It's simply that Robert Houdart claims that it isn't, which is a lie and a gross misrepresentation of the authorship of the software.
Jeremy
.