Monday, October 31, 2011

Pontificating

       As you all know the latest version of Houdini (Houdini 2) is no longer free.  I am sure this probably has something to do with the downfall of Rybka.  No matter. You can now purchase Houdini 2 Aquarium (supports up to 6 cores and 4 GB of hash for about $60 and Houdini 2 Aquarium Pro (supports up to 32 cores and 32 GB of hash for about $95). What does this mean for the average player who just wants to analyze his games? Nothing.  And what about the serious CC player who is playing at the titled level?  Again, I think the answer is nothing.  The reason is because at the titled level they know enough about chess to know when the engine output is suspect.  Still, they will want the best equipment available, so they will make the necessary purchases.  So, who is going to suffer?  Only those CC players who rely solely on engine generated moves but most of them will not get within sniffing distance of a CC title anyway. They will lose a few more games to guys with better equipment and the slightly stronger engine. Both groups will continue to lose to strong IM’s and up who are using engines, even if the engines are a few points weaker.
      I’ve been reading some bulletin boards about various chess engines and the big brouhaha about cloning and who stole what from whom.  Mostly it’s howling at the moon because all engines have relied on previous advances by other engines.  Any time a computer programmer sets out to write any kind of program, he studies the source codes of similar programs. 
       In order to get a new patent on anything all you have to do is make a small improvement on the existing design and you can then market it. Ideas have always been built upon by artists, musicians, philosophers, scientists, engineers, etc.  This is not the same thing as forgery. A copy of a work of art that is claimed to be an original is illegal, but a copy is not. Musical compositions are often given different arrangements.
       So for people to say that for a chess engine program to be legitimate it has to be written from scratch doesn’t make any sense. Even if all the programmer does is take a previous program and tweak 25 or 50 points more out of it, then he has what in any other field would be a patentable improvement for which he would get full credit.  That’s how new discoveries are made in the scientific field so why are chess playing programs any different.
      The programmer of Rybka, Vasik Rajlich, acknowledged that he used many ideas from the open source program Fruit whose source code is easily available on the Internet. Then all of a sudden he was accused of cheating when the International Computer Games Association took the action after an investigation into claims Rybka was a derivative of other chess engines. It was suspicions of cloned engines that led to the formation of the ICGA Clone and Derivative Investigation Panel.
      This whole mess reminds me of the idiot on one of the chess forums a couple years back.  The question came up about a German book of Boris Spassky’s games that Bobby Fischer used in preparation for their match.  A couple of us old timers remembered reading that Fischer did use such a book and I even found a copy of it for sale on e-Bay.  The ad on e-Bay said something to the effect that this is “the book Fischer used in his match preparation...”  So what did the guy go berserk about?  He said that until he read something written personally by Fischer saying he used the book, he would not believe it.  He even went further and criticized the e-Bay ad by claiming it was fraud for the seller to claim it was the book Fischer used to prepare for Spassky.  Why was it fraud?  Because the seller wrote “...it was the book Fischer used...” and this guy was taking it to mean it was the actual book, the one personally owned and fondled by Fischer.   Some people will believe what they want to believe no matter how much proof you offer and no amount of proof will ever be enough.  And then some people just like to nitpick.
      The rule about writing down your move after you play it because to write it down before you make it constitutes consulting notes triggers my gag reflex.  Give me a break!  For years and years we wrote down our move before we played it (or after if you wanted to...nobody cared).  Sometimes, on second thought, a player would scratch out his move then write down another one, but NOBODY ever dreamed of calling it consulting notes and therefore cheating. Heck, I’ve even seen guys tell their opponent, “You forgot to punch your clock.”  It’s a rule violation but I never heard anybody complain.  I wouldn’t be surprised if these days there are those who would summon the TD and try to claim a win because their opponent distracted them.

1 comment:

  1. Well maybe the free Strelka will give Houdini a run for its money

    http://strelkachess.narod.ru/Strelka5.html

    But I am with you. I like to study old games, try and predict the next move, and compare my move to multi-variation analysis. A few elo of one program over another is not as important as trusted analysis. Hiarcs, Shredder, Fritz, or the free Critter and Stockfish seem fairly reliable.

    ReplyDelete