Author: Jason Williamson
Date: 04:38:51 04/10/01
Go up one level in this thread
On April 09, 2001 at 16:39:45, Bertil Eklund wrote:
>On April 09, 2001 at 16:20:42, Robert Hyatt wrote:
>
>>On April 09, 2001 at 14:43:32, Tony Hedlund wrote:
>>
>>>On April 09, 2001 at 13:56:05, Mike S. wrote:
>>>
>>>>On April 09, 2001 at 11:38:02, Hans Christian Lykke wrote:
>>>>
>>>>>(...)
>>>>>[White "Deep Fritz K6-2 450 128MB"]
>>>>>[Black "CM8000 K6-2 450 128MB"]
>>>>>(...)
>>>>>
>>>>>{74MB, Deepfr~1.ctg. K6-2 450 128MB}
>>>>
>>>>I must admit that I probably misunderstood the SSDF hash size information until
>>>>now. I always thought, each program uses 128 MB hash. Now, it looks to me as if
>>>>this is just the total RAM memory of the pcs used (?), i.e., here Deep Fritz was
>>>>set to 74 MB Hash. This would mean, the list itself doesn't contain the hash
>>>>size info. With which hash size is Chessmaster 8000 tested?
>>>>
>>>>I apologize if this has already been discussed, and the information given here
>>>>before respectively (which I didn't notice), but: What is the general SSDF
>>>>policy regarding the hash size setting (I hope you aren't testing CM8K in the
>>>>default setting of only 1 MB hash)?
>>>>
>>>>Thanks,
>>>>M.Scheidl
>>>
>>>We are running CM8000 with 32MB hash.
>>>
>>>Tony
>>
>>
>>Since this will obviously be asked soon, why use Fritz with 74mb and CM8000 with
>>only 32mb???
>Hi!
>
>The main reason is that 32mb is max. The second is that mr de Koning told us
>that 16mb was more then enough for P200MMX.
>
>As a general rule, we try to follow the recommendations from the programmers if
>there are no known recommendations, we use as much as possible (no swapping
>allowed).
>
>Bertil
32 megs is NOT the max hash size. I run Chessmaster with 64 megs on my dual
celeron 400 (192 megs total ram), and can go quite a bit higher. Perhaps I am
not understanding what you are saying.
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.