Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Anything to this hyperbola bitboard URL?

Author: Dann Corbit

Date: 21:56:03 09/26/00

Go up one level in this thread


On September 26, 2000 at 13:51:17, Larry Griffiths wrote:

>I was looking at these URL's.
>
>http://members.tripod.com/~RyanMack/computer.htm
>
>http://members.tripod.com/~RyanMack/hypertech.htm
>
>Does this look promising for bitboards?

A quote:
"The first release of the Hyperbola Project is slated for late 2000 or early
2001. Early estimates predict that the program will run 20-25 times as fast as
Crafty. Without any special enhancements (except a basic hash table, move
ordering, and alpha-beta searching) it is estimated that the program will
analyze nearly 3 million NPS (Nodes Per Second) on a 266 MHz processor. Current
versions of Crafty analyze about 125,000 positions per second on the same
hardware. On the latest Xeon-III 733 MHz processors, it is possible that the NPS
rate might exceed 10,000,000 at times."

Sounds more like the "Hyperbole" project to me.  But the proof of the pudding is
in the eating.  I'll just have to wait and see.

733MHz/10,000,000 NPS = 73 cycles per node, including all program overhead.  I
know I would be impressed if that could be achieved.

The 125K NPS should be close... Here's what I get at 300 MHz with version 17.11:

EPD Kit revision date: 1996.04.21
found computer opening book file [e:/crafty/release/bookc.bin].
hash table memory = 24M bytes.
pawn hash table memory = 4M bytes.
EGTB cache memory = 2M bytes.
draw score set to    0.00 pawns.
choose from book moves randomly (using weights.)
choose from 5 best moves.
book learning enabled
result learning enabled
position learning enabled
threshold set to 9 pawns.
4 piece tablebase files found
1302kb of RAM used for TB indices and decompression tables

Crafty v17.11

White(1): bench
Running benchmark. . .
......
Total nodes: 52426399
Raw nodes per second: 171328
Total elapsed time: 306
SMP time-to-ply measurement: 2.091503
White(1):


Of course, massive hash won't help.  That makes NPS go down.  I would be curious
to see more details about how this is to be worked out.



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.