Author: Sune Fischer
Date: 10:20:02 10/22/03
Go up one level in this thread
On October 22, 2003 at 12:50:01, Mike S. wrote: >>(...) >>I would assume the default settings were intended to be the allround strongest >>by the author, hence by default the default settings should be used (heh), >>unless the author has a strong request that they use something else (which will >>probably default in the next release then!?). > >You obviously have no idea what you are talking about, when you talk The King >settings. - It*is* true what you say though, for most engines. But *not* for The >King, at least certainly not since CM8000, the previous version (King 3.12). All I'm saying is, if you have a case present it to the author and persuade him to make the changes into the default ones. If you can't persuade him you can't persuade me, I always trust the programmer to know more about the program than anyone else, no offence. >What puzzles me is: > >Didn't you read all the CCC threads (and there were *many* of them), which have >discussed, explained, and virtualy proven this (with lots of results)? > >Don't you know any of the tournaments and ratings lists, which show how strong >several custom personalities are, compared to the default? I'm not convinced. >For example: >http://www.grailmaster.com/misc/chess/comp/cm.html >best-ranking CM9000 setting rated 2725, >CM9000 default personality rated 2612, rank #56 > >Actually it would be *very unfair* to test King with default settings against >other top programs, because it would limit it way below it's potential, on fast >comps (estim. 500 MHz...3 GHz) typically used nowadays among computerchess fans. The great thing about default settings is that even your everyday idiot will get the strongest program playing out of the box. No need to scan the internet for tricks and tweaks and spend hours becomming an expert. In most cases the thing will be running on default settings so those should be the one that gets tested, as a matter of principle. >Has it come so far, that something which I thought is *common knowledge* in CCC, >must be explained and proven in every posting, and even long time regulars >question well-known facts?? You could be right but you could also be wrong, I say we make the author the judge of that :) -S. >Regards, >M.Scheidl
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.