Author: John Merlino
Date: 07:41:04 03/27/01
Go up one level in this thread
On March 27, 2001 at 10:26:11, Albert Silver wrote: >On March 27, 2001 at 08:45:32, Bertil Eklund wrote: > >>On March 27, 2001 at 08:29:17, Albert Silver wrote: >> >>>Heh-heh. Fantastic (meaning the inclusion of CM in SSDF), but one question: is >>>the latest patch of CM being used? And are the settings all correct? I ask this >>>last because I know that SSDF's general policy has been to leave the >>>manufacturer's general settings as are, but as CM was designed for a very >>>different audience than that of Fritz and company, they might not be ideal for >>>auto232 matches. Just wanted to be certain things like permanent brain, book >>>learning (very important), etc, are all working. >>> >>> >>Hi! >>The latest patch is used. What do you mean with, leave the genaral settings as >>they are. Of course we try to use best possible settings for every single >>program. In fact I was the one that answered mr de Koning for the best settings >>for CM6000 and he recommended 16mb hash (P200MMX) as "more then enough" .We try >>with 32mb on 450. >> >>Bertil > >No big deal, but when CM6000 was first tested, someone had asked about the >settings and asked in particualr about the famous CM6555 settings everyone >seemed to agree were better. The answer, quite understandably, was that the >CM6555 settings weren't to be used, and the reason given was that it was >presumed that the manufacturer's settings were best. It was unclear how far that >presumption went. Now you cleared that up. > > Albert I don't believe that "everyone seemed to agree" that CM6555 was better than the default CM6000. I also don't believe that it is WORSE either. I followed most of those posts, and many of the individual testing proved inconclusive. Just about the only setting that I would say could POSSIBLY be changed from the default to improve the strength of play would be to increase the "Selective Search" setting. However, while most users appear to believe that the maximum setting here is best, I have also not seen any proof of that. Everybody either uses the default or the maximum, and nothing in between. This is obviously faulty testing, IMHO. jm
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.