Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Genius on Palm / Richard Lang the return

Author: Christophe Theron

Date: 09:55:25 02/08/00

Go up one level in this thread


On February 08, 2000 at 12:09:48, Tony Schleizer wrote:

>
>The page says that it is based on the Amsterdam program.
>Can somebody tell me how fast a Palm Pilot processor
>is compared to the processor the Amsterdam ran on?
>Also, if anyone has tried this out, what are your impressions
>of it's strength.


It is based on the Roma program, not the Amsterdam.

The Amsterdam was a 16 bits program, the Roma is a 32 bits program.

The Roma program is the last program of Richard Lang that did not use hash
tables.

It won the 1987 world championship without opposition actually. Saitek and
Fidelity did not enter the tournament because they were not ready. If they had,
the Roma program would probably have smoked them anyway.

The Mephisto Roma on a 14MHz 68020 has been evaluated by the SSDF at 2030 ELO
(with a good confidence: -26 +27, 95% confidence).

The 14MHz 68020 is evaluated at 2 MIPS.

The 16MHz DragonBall EZ (PalmIIIx and PalmV) is evaluated (by Motorola!) at 2.7
MIPS.

If all this is true, then you can expect the Palm ChessGenius to be rated at
2050 ELO on a standard PalmIIIx or PalmV, and 2080 ELO (approx) on a 26MHz
overclocked Palm.



    Christophe



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.