Author: Robert Hyatt
Date: 13:06:54 09/25/02
Go up one level in this thread
On September 25, 2002 at 15:30:39, Roy Eassa wrote: >On September 25, 2002 at 14:36:38, Dann Corbit wrote: > >>On September 25, 2002 at 14:25:33, Roy Eassa wrote: >> >>>On September 25, 2002 at 14:10:22, Robert Hyatt wrote: >>> >>>>On September 25, 2002 at 13:27:10, Roy Eassa wrote: >>>> >>>>> >>>>>Is Ruffian really a completely new chess engine that is very, very strong? >>>>>Christophe estimated the chances of this to be 0.01% in an earlier thread. >>>>> >>>>>(I'm catching up here after a couple weeks of PC problems, but recent Ruffian >>>>>messages seem to be relatively positive.) >>>> >>>> >>>>re-read what he wrote. He said that the chances of a total "unknown" writing >>>>such a strong engine alone is very small. That doesn't preclude the program >>>>being new. And if it _is_ new, that suggests that another possibility we >>>>both hinted at might be the case... namely that the author is not as unknown >>>>as we might think... >>> >>> >>>So the author might be somebody other than that Swedish fellow whose name has >>>been mentioned here several times? What is his role, then, I wonder? And who >>>might the author really be? (Are there top-rank computer chess authors in >>>Sweden?) >> >>I think the point that Christophe was making that a person working in obscurity >>and isolation is not going to make a world-beater chess program. >> >>Then the point that Robert Hyatt made was that this person may not have been >>working in obscurity. Perhaps he has read the JICCA and/or the hundreds of >>computer chess articles on the net. Perhaps he has read Ernst Heinz's book. Or >>in some other way become familiar with the state of the art. >> >>In this respect, I agree fully with Christophe and Robert, unless the author is >>a genius of some perplexing magnitude. > > >I read Christophe's point differently. It seems he and Robert suspect that the >author is really not that Swedish fellow (who I think is quite unknown) but >rather somebody well-known (who has previously produced a very strong program), >or at the very least somebody who has been seen in CC circles asking a lot of >technical questions. I wouldn't personally go that far, yet. Not enough information. After the program has played for a good while, someone will notice if it seems to have characteristics from a particular known engine, which might lead to the person helping the actual author. If that has happened. There are still several different things that _could_ have happened here. Including the possibility that it is really new, really good, which would be really surprising of course. :) > >My interpretation is based on the use of quotes around "unknown" and the use of >the phrase, "the author is not as unknown as we might think." > >If my interpretation is correct, it is a much stronger statement than that the >author has studied computer chess programming extensively (which I think few >would disagree with!). I don't think "studying" would be enough to dive to the front of the pack, but that is just an opinion...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.