Author: Dann Corbit
Date: 13:41:27 07/01/04
Go up one level in this thread
On July 01, 2004 at 15:44:50, Robert Hyatt wrote: >On July 01, 2004 at 15:34:06, Matthew Hull wrote: > >>On July 01, 2004 at 15:11:05, Norm Pollock wrote: >> >>>On July 01, 2004 at 15:01:47, Uri Blass wrote: >>> >>>>On July 01, 2004 at 14:32:18, Norm Pollock wrote: >>>> >>>>>On July 01, 2004 at 14:14:50, Peter Berger wrote: >>>>> >>>>>>On July 01, 2004 at 13:51:27, Norm Pollock wrote: >>>>>> >>>>>>>I hope Dr Hyatt uses 18.12. I think it is stronger than 19.14. 18.12 was >>>>>>>recently matched by SSDF to Shredder8. I suspect the chessbase folks are >>>>>>>expecting 18.12. >>>>>> >>>>>>Given the result of the recent SSDF match I am surprised that you think >>that >>>>18.12 is stronger. It is likely that Dr Hyatt will use version 19.15 >>though, >>>>which is most probably also what the "ChessBase folks" would expect >>:) . >>>> >>>>>>Peter >>>>> >>>>>I think 18.12 got about 25% in the SSDF match. Keep in mind that Shredder8 >>is a >>>stronger engine, the match was in the chessbase GUI with a CB book. I >>think that >>>with a level playing table, 18.12 would have a 33% chance of >>beating shredder8 >>>in a single game. >>>> >>>>I think that you forget that 19.15 has 200% result against every program in >>the >>ssdf list. >>>> >>>>If you do not believe me then take the number of games that 19.15 played in >>the >>ssdf against program X(0 for every X) and multiply by 200%. >>>> >>>>You will find exactly the number of points that Crafty19.15 got in the ssdf >>>>against X(0 for every X). >>>> >>>>Uri >>> >>>Thanks for the math briefing. But isn't 200 * 0 = zero? ;-) >>> >>>Getting back to my equally illogical theory, the CB boys have 3 serious >>>contenders. Only one, Crafty, will have the hardware to pull off an upset. >>They >have checked all the versions of crafty and aside from a new untested >>version >19/.15 that may be buggy, the one they fear the most is 18.12. >>Otherwise why did >SSDF suddenly match it with Shredder8? They burst into the room, laughing maniacally, rubbing their hands together. As we all know the SSDF is just a bunch of chessbase stooges. Yawn. >>19.15 has some fixes. Also, download 19.14 and read "main.c" you will see how >>many bug fixes there have been since 18.12. Even now, 19.15 on ICC dual xeon >>is playing better than 19.14, IMHO. > > >I think it is playing exceptionally well at the present. At the last author's >only event, I saw a horrible endgame mistake. Peter was playing test games >against commercial programs and saw a similar mistake that converted a draw to a >loss. After a lot of studying and testing, this (and a couple of other issues) >has been cleaned up and it is really playing solidly in the endgame now. Better >than any version of the past two years at least... > >Whether that will affect any WCCC games is another subject, of course. But I >expect it to play well, whether it wins or loses every game... 19.12 was a big step forward in strength. I expect that 19.15 will probably be the strongest if it has bug corrections. You will always have those 18.12 fans and 17.14 fans (and probably some would want the Jakarta crafty). ;-) At any rate, I am looking forward to an exciting contest. To bad you didn't get an 8-way box. That would really have put the fear of god into them.
This page took 0.01 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.