Author: Vine Smith
Date: 16:17:29 05/05/01
Go up one level in this thread
On May 05, 2001 at 12:52:36, John Merlino wrote:
>On May 05, 2001 at 06:20:19, Vine Smith wrote:
>
>>On May 04, 2001 at 21:22:16, John Merlino wrote:
>>
>>>On May 04, 2001 at 18:32:38, Vine Smith wrote:
>>>
>>>>On May 04, 2001 at 16:18:52, Dann Corbit wrote:
>>>>
>>>>>On May 04, 2001 at 16:15:24, Vine Smith wrote:
>>>>>
>>>>>>Giving this program to Kramnik 3 months ahead of time will be to his
>>>>>>disadvantage. First, he'll waste lots of time figuring out how to get it to work
>>>>>>reasonably ("This isn't like Fritz at all! Where are the hash table settings?").
>>>>>>Then, he'll probably make the mistake of opening an analysis window, maybe the
>>>>>>mentor window too, and base his preparation on the engine running at 1/2 to 1/3
>>>>>>speed due to the 2 or 3 engines spawned by this process, unbeknownst to him.
>>>>>
>>>>>On the other hand, he'd probably post here with his questions under an assumed
>>>>>name and John Merlino would solve his problems in a jiffy.
>>>>>
>>>>>I wonder if the CM folks are going to do a multiple CPU version?
>>>>Nah, Kramnik will be too insulted by being left off the "Personalities" list to
>>>>contact Merlino ("Hmm...Anand, Kasparov...what the? Where am I? Are they saying
>>>>I don't have any identifiable style?!"). Or he'll give himself away posting new
>>>>settings for a Kramnik personality.
>>>>Yeah, Chessmaster SMP -- uses one CPU per window open.
>>>
>>>A Kramnik personality is going to be one of the new GM personalities in the next
>>>version (along with a few other hopefuls, possibly a couple of old-timers such
>>>as Paulsen and Zukertort).
>>>
>>>jm
>>That will be most interesting. Possible descriptive blurb for Kramnik
>>personality: "Does nothing in particular for first 30-40 moves. Then wins
>>endgame by force (all 6-man tablebases REQUIRED)". But please tell me it will be
>>possible to view analysis without incurring a 50% speed penalty -- I spend more
>>time analyzing than playing. The King engine sometimes comes up with interesting
>>approaches, and I'd really like to know what it's thinking about.
>>Regards,
>>Vine Smith
>
>It is very easy to get "infinite analysis" with only one engine operating. Just
>do this:
>
>1) Setup up the position that you want (or go to the proper point in the game)
>2) Set human vs. human (by pressing CTRL+G and choosing the two buttons that
>look like human faces)
>3) Open the Thinking Lines window and select "Chessmaster"
>
>That's it.... You can now move forward and backward through the game, or make
>new moves on the board, and the Thinking Lines window will always reflect the
>current position.
>
>jm
This is actually what I have done in the past when analyzing with CM8000, but I
had misunderstood the comments about this issue that appeared in the context of
the SSDF testing mishap; I thought the first engine never went away under any
circumstances. Also, because the search speeds reported by CM8000 seemed
relatively slow compared to other engines, I assumed this was because of a dual
engine load. All right, I take it all back! (although this reduces your chances
in a hypothetical CM8000-Kramnik match now)
Maybe Henry Bird should be included with other 19th century players whose style
the next version of CM will try to duplicate. But if you can mimic his bizarre
notions successfully (for instance 1.e4 e5 2.Nf3 Nc6 3.Bb5 Nd4 4.Nxd4 exd4 5.O-O
h5!? or 1.e4 e6 2.Bb5!?) it will be a revelation.
Regards,
Vine Smith
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.