Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Virtual Memory - More data

Author: Jeremiah Penery

Date: 22:36:34 01/18/00

Go up one level in this thread


Another test Bob did, this time to depth 9 on the same position.  Here are the
results:
-----

[D]3rr1k1/1p2b1pp/p1b1p3/8/PnBBq3/2Q2N2/1PP3PP/R4R1K b - - 0 1

Komputer Korner <korner1@netcom.ca> wrote:
: What we need is a comparison where  the total possible nodes
: calculated (either evaluated or looked up in the hash table) is
: smaller than the number of possible entries. Therefore the experiment
: should be done at a much smaller depth so that we can see the cases
: where the hash table isn't overwritten at all.
: --
: --
: Komputer Korner
: The inkompetent komputer

Here is more data:  This time to depth=9, rather than 11.

time    bytes  entries    nodes searched
-------------------------------------------
1:05     96K       6K        12,139,882
1:00    192K      12K        11,245,725
54.5    384K      24K         9,995,634
55.1    768K      49K        10,131,840
52.3    1.5M      98K         9,567,818
46.3      3M     196K         8,463,339
44.5      6M     392K         8,119,062
43.5     12M     800K         7,932,811
43.0     24M     1.5M         7,833,488
42.9     48M     3.0M         7,802,585
42.7     96M     6.0M         7,779,999
42.7    192M    12.0M         7,779,122
42.7    384M    24.0M         7,778,994

This gives data for a reasonable sized search, but with a hash table that
actually goes way beyond the total size of the tree.  You can draw your own
conclusions, but it is pretty easy to guess, based on the above, how much of my
tree is 'q-search' and is not hashed...
----

Jeremiah



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.