A half hour ago I received an updated and upgraded version of Rainbow! It has not been fully tested- but enough to the point it has shown it should be able to handle Houdini at his best.
But instead of replacing Rainbow 1.0 beta, I was asked to just let that match play out. And to start another 500 game match, but with a faster time control- 5'+5". The feeling is 1.0 beta is getting a shot at the gold ring, and he knows how to handle himself.
However, there should be an appreciable strength increase in this update, which she thought 'what the hell' and decided to send, and liking the idea of 5'+5", so Houdini can be removed from Number 1 in half the time- if 1.0 beta has not already removed him by then.
Updates should be much more often with this match, and it is beginning as we speak. So say hello to the strongest chess engine the world has 'never' seen: "Rainbow Limited- beta 2"
Interesting posts and nice that you want to do an independant test, but I find it a bit sad that the programmer does not want to come forward with his name. I can't protest against the name of the program although that might be confusing to some. This is not the Stockfish fork of similar name, which is not at all private either because 98% Stockfish and of the 2% rest, more than half of that is already released under GPL. There was somebody with the program "Serpent" already. I hope it is not the same person! I had a feeling the programmer might be Gian-Carlo Pascutto because I believe he has been making Stockfish clones for his personal use before, and there he was posting again the next day after a long absence from chessprogramming. So it could be him, I don't know. Look if we want to keep something like the WCCC alive, I think programs like this if it is based on the open source programs, GPL or not, have a right of existence but should not be allowed to compete for the WCCC title against programs that have been developed independantly from the ground up. That would be a sham.
If this is not such a program, then imagine Mark Uniacke making a Houdini clone and adding all the tricks he knows from his own programming experience to it. I think it would give Houdart a run for his money and he would have every right. It would be nice though if he came forward with his real name and explained his reasons, at some time. But he would be allowed to sell it I think if he avoided including any code to which Robert can claim his own copyright. Maybe, if Mark and Harvey were not already diversifying into real good opening books, assisting the world champion of FIDE defending his title against Gelfand and now into a new interface, wonderful stuff, but otherwise Mark Uniacke could do also something like Robert Houdart and compete with him at this other subculture of chessprogramming even at the commercial level. It would maybe force Vas Rajlich and Larry to begin a case, finally, about their own copyright from the Rybka 3 program being breached, but that really is not going to happen I think. The other possibility is more likely.
I would like to point out that the program Rainbow Serpent is a GPL program and as such protected under that license, and that I am in no way involved with a program called Rainbow 1.0 beta 1 or Rainbow 1.0 beta 2, or any programmer who wants to remain anonymous. Thank you!
Eelco
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
