Author: Chris Whittington
Date: 06:20:49 12/19/97
In response to various bits of blather about 'consensual position problem sets', such as the following, but the whole lot of of you are guilty of the same nonsense: >I had a very similar idea a few months ago. As a result I have a >positional test of 24 problems, all from great grandmaster games and all >commented by grandmasters. I discarded positions with solutions not >found by any programs. Since nobody, after two weeks has seen the gaping flaw, I guess I'll just have to arrogantly spell it out for you all. Suppose we have a bunch of one-eyed chess programs. So we take our 1000 positions and GM solutions and we apply them to the bunch of the one-eyes. Then we cull every position 'unsolved' - we call these irrelevant ones; and we cull all those 'solved' very fast - we call these trivial ones. So we get left with some unculled postions, and we apply our chess programs to these positions, and, magic, now we have the king of the one-eyed. Then we take some other chess program(s), and we apply the tests to this(them). It does better ? Magic, we have a new king. It does worse ? Magic, we rank it. Do we begin to see the problem now guys ? What do we do with the program(s) with two eyes ? The programs that see different things ? The ones that aren't clones ? Forget it, I suspect you all of being robots. Happy Christmas. I'm joining the Pope. At least he's in opposition to this kind of exclusive rank-closing and related twaddle. He's mad but he sees more and further than you all put together. Chris Whittington
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.