Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hash Table Size Versus Performance.

Author: blass uri

Date: 08:14:45 05/30/98

Go up one level in this thread



On May 30, 1998 at 09:56:25, Robert Hyatt wrote:

>On May 30, 1998 at 02:28:02, Steven Juchnowski wrote:
>
>>How much does hash table size effect performance?
>>
>>On one hand we hear that Fritz5 can gain about 40 ELO points by
>>increasing the hash size from 44MB to 100MB.
>>
>>On the other hand there are claims that doubling the hash size will
>>only improve performance by a few ELO points.
>>
>>I assume the above claims are based on tournament time controls.
>>
>>So which is the real picture, or is it perhaps that the sensitivity of
>>hash
>>size on performance varies from program to program.
>>
>>Regards
>
>
>there are at least a couple of things that could make Fritz far more
>sensitive to hash table size than other programs:
>
>(1) a poor replacement strategy.  If this is true, then a larger table
>reduces replacement, which would produce better performance.
>
>(2) using the table for other things besides the normal score/best move/
>etc.  If this is true, replacing *any* entry could be bad, depending on
>what is stored in the table.

I think they use the size of the hash tables to decide about
the selective depth fritz5 analyzes
usually when I give fritz5 big hash tables the selective depth
for the same brute force depth is bigger.

I think this method is not always good because sometimes
when I use big hash tables it takes fritz5 more time to be at the same
brute force depth
and if the brute force depth is important to discover
something big hash tables can be counter productive.

I am not sure about the hash tables I should use if I give
the computer a long time because
I have examples that big hash tables can be counter productive in a long
time(they cared only for tournament time control and it was not
important
for them).

Uri
>
>no one knows what Fritz does, but one of the above reasons is almost
>certain to be correct.  I'd suspect (2) myself, since replacement
>strategies are well-known now.



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.