Author: pavel
Date: 22:47:04 08/19/00
Go up one level in this thread
On August 20, 2000 at 00:43:41, Robert Hyatt wrote: >On August 19, 2000 at 19:35:09, Uri Blass wrote: > >>On August 19, 2000 at 19:05:37, Peter Skinner wrote: >> >>>>Ok , here is another one of them :-) >>>> >>>>I never have seen any data suggesting it should be even higher rated than it is >>>>now and I _do_ follow its results. >>>> >>>>That crafty on a Quad-Xeon will be even stronger , sure , but other would >>>>probably be better too on an Athlon 1000 which they don't get either . >>>> >>>>Crafty competes very successful at SSDF currently IMHO and I doubt many will >>>>argue that it won't hurt Crafty getting the Fritz 6 opening book and the >>>>Chessbase book learner ;-) >>>> >>>>If SSDF changed versions every month or so there _never_ would be any reliable >>>>results and they chose the one distributed by Chessbase át some time . This >>>>sounds reasonable ; and in fact it seems it isn't very broken at all . SOS also >>>>wins matches against Crafty17.11 or 17.12 from time to time if you look at other >>>>tournaments . >>>> >>>>I am sure Crafty17.13 will probably soon be tested by SSDF after WMCC but >>>>whatever one can say about SSDF , noone can expect them to switch versions every >>>>other week and play some 150 tournament games with all of them . >>>> >>>>pete >>>> >>>>> >>>>>As for Crafty not using the Winboard environment, that is actually a good thing. >>>>>Under ChessBase GUI, it is getting a well designed book to use. Crafty's book is >>>>>good, but most times very speculative. The Crafty 16.xx series book, was by far >>>>>better than the 17.xx series. >>>>> >>>>>Then again Uli, would you want the SSDF testing a version of Comet that had >>>>>serious bugs in it? I don't think you would, so why should another author have >>>>>to take the same sort of results? >>> >>>I am just stating that the 17.07 version had bugs, serious bugs, so why not use >>>the next version that was more stable, and didn't have these bugs? >> >>What are the serious bugs? >> >>SSDF used in some games Crafty17.04. >>Do you think it was better for them to continue using 17.04 and not use 17.07? >> >>Uri > > >I'm not going to dive into the middle of this... it is too complex. "crafty" >is _different_ from the commercial engines. They release one program a year, >with maybe a bugfix version sometime later. I release 10-20 versions a year. >Which one to test? IMHO _always_ the most recent. But perhaps that is not >practical. I would prefer them to simply list "crafty v17" and upgrade as >new versions come out. That is about the best that could be done, since >Crafty doesn't have an annual upgrade date... So what if an earlier version >had a bug and had a worse result than is reasonable. The 'fixed' version should >make up for the problems, over time. The alternative is really bad IMHO, that >of having _each_ version of Crafty on the SSDF list. In 3 years, 99% of the >entries on their list will be my program. :) on the other hand, to make crafty compatible with chessbase GUIs it is needed to be compiled with additional codes in the engines. Probably nothing complex, probably same as winboard but differant output commands. now tell me how can you tell for sure that chessbase (after adding the neccessary codes) compiled it correctly without making any mistakes? . after all it is not being done by the author of the program himself (eg, bringer, lilgoliath, sos , comet......etc). well you might say the author dont have the interest on compiling each version for chessbase GUI.......in this case why not use the winboard adapter? you might say its not well known?? (great reason).......well NOW you know it ;)) LOL pavel
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.