Author: Terry McCracken
Date: 06:48:30 11/07/03
Go up one level in this thread
On November 07, 2003 at 08:31:26, Steve Maughan wrote: >Terry, > >>Sir, I'm not trolling at all, and I know what I'm talking about. > >OK - I'll accept that this is a genuinly held view > >>Do you remember Alan Tomalty? Well I know him through his work and and have >>talked with him, he'd agree with me. > >Komputer Korner - what does he have to do with the statement you made? > >>Also, I've tested computers myself and have played in rated tournaments, I >>understand what I'm talking about. > >You're going to have to come up with something a little better than re-stating >that you think you know what you're talking about. > >>So you musn't have a very good grasp what a rating means. > >OK - let's follow this line of reasoning. My idea of a rating is a measure of a >computer's ability to beat its opponents. What's yours? > >>It's _not_ a scientific or absolute measuring of strength. It is statistical >>only and if you understand statistics, you'll understand my meaning of of 50 >>points up or down have little meaning. > >A genuine 50 point difference implies that after an infinite number of games the >program with the highest rating will win more time (I don't have the ELO tables >but probably winning 58% of the time compared to the opponents 42%). Yes this >is an average and will vary from opponent to opponent. 50 ELO is only >irrelevant if the machines are playing against a patzer like me i.e. the >difference between 2700 and 2650 is nothing if the opponent is 1800 ELO - if >they are 2600 ELO it's a different matter. > >He's a question for you - if you throw out statistics and say that it isn't >scientific, what is scientific? Statistics is the basis of all science. > >Regards, > >Steve Steve I'm going to keep it simple, we agree more than we differ. Statistics are probability, which is science, albeit statistics needs a very broad base of data IMHO to reveal the truth, or at least come very close to it. This would have to take into account the quality of play by the engine, the book, bugs, (luck) etc. Is the SSDF accurate insofar as testing which engine is strongest by the parameters they set? Yes, they are, but so much is left out. It needs a human GM or GM's to analyse these matches, to get the best data. It's doubtful we'll see this anytime soon. I use SSDF as a general guidline, and not the gospel truth. Terry Terry
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.