Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Even Palms are greater than famous old computers, people don't think

Author: Christophe Theron

Date: 21:23:13 06/10/03

Go up one level in this thread


On June 09, 2003 at 22:06:13, Robert Hyatt wrote:

>On June 09, 2003 at 15:07:00, Christophe Theron wrote:
>
>>On June 09, 2003 at 04:57:05, stuart taylor wrote:
>>
>>> People not involved in chess sometimes look upon a great Palm chess program as
>>>being one of the lowliest little gadgets anyone could think of, and certainly a
>>>most obvious one. But the greatest of handheld programs (soon to be Christophes
>>>upcoming Palm program maximised for about 400Mhz (Arm).Is probably far far
>>>greater in all its aspects than all the great famous old computers of the 1970's
>>>and perhaps of some from the 1980's. More recent too, but of those, the Palm
>>>would CRUSH them in every game.
>>> We should be happy not to be paying ten million dollars for a recent Palm with
>>>best program in a few months from now, maybe now too. Not more than one person
>>>should also have the privelage of having one!
>>> Not to speak of things like Shredder 7.04 on 3Ghz Athlon, which is almost
>>>nothing today.
>>>Who wouldn't have paid alot in the early 1980's for a pocket world champion? ANY
>>>millionaire who understood what it is would have paid $100,000 guranteed!
>>> One who didn't understand what it is would never even agree to keep it out of
>>>the garbage (as far as the chess is concerned) for one second!
>>>It's sometimes hard to explain this to people, I mean to the type who want it in
>>>the garbage. I sound to them like a real baby, and all their opinion of me goes
>>>down the drain!
>>>S.Taylor
>>
>>
>>
>>Some interesting numbers:
>>
>>SSDF elo rating of Chess Tiger 14.9a for PalmOS running on Palm m515 @42MHz:
>>2101 (+/- the usual margin of error that can be found on the SSDF site).
>>
>>Estimated SSDF elo rating of Chess Tiger 15.0 for PalmOS running on Palm
>>Tungsten-C @400MHz: 2290 (explanations and formulas can be found at
>>http://www.chesstiger.com, click on the "FAQ" link).
>>
>>Now here is a list of interesting computers or programs tested by the SSDF:
>>
>>??? Chess Tiger 15.0 TungstenC (estimate)   2290    ?     ?     ?     ?     ?
>> 96 Tasc R30 v. 2.5                         2274   41   -38   343   69%  2136
>>104 Fritz 3.0 Pentium 90 MHz                2242   18   -19  1438   42%  2299
>>111 Mephisto London 68 030 36 MHz           2216   71   -63   122   68%  2080
>>113 Chess Machine 30-32 MHz Schröder 3.1    2205   30   -29   586   64%  2102
>>137 Chess Tiger 14.9 Palm m515 16MB 42MHz   2101   69   -74   100   39%  2180
>>144 Mephisto Atlanta SH7000 20 MHz          2087   29   -27   667   68%  1956
>>154 Sapphire II                             2009   34   -32   464   63%  1918
>>161 Fidelity Mach IV  68020 20 MHz          1974   18   -18  1495   48%  1988
>>184 Chess Genius 1.5 Palm m515 16MB 42MHz   1870   69   -78   100   32%  2002
>>185 Mephisto Roma  68000 12 MHz             1869   18   -18  1519   56%  1827
>>
>>(the last two lines show that a Palm m515 running at 42MHz has approximately the
>>same processing power as a 68000 running at 12MHz because ChessGenius 1.5 for
>>Palm has the Mephisto Roma chess engine, source is the Chess Genius web site)
>>
>>
>>Whe can tell me the SSDF rating of Belle and Cray Blitz?
>>
>
>Neither was ever tested.  Cray Blitz had an official USCF rating of around
>2300 prior to multiple cpus.  Belle had an official USCF rating of 2204 in
>1983 although it went up a bit over the next year or two.  HiTech had a 2450+
>USCF rating in the 1985 time frame.



Then wouldn't it suggest that Chess Tiger 15.0 for PalmOS running on the
Tungsten-C is (at least) in the same ballpark as Cray Blitz before multiple CPUs
and Belle 1983?

Also it suggests that the (future) native ARM version of CT15 running on Palm
Tunsgten-C would be in the same ballpark as Hitech 1985.

Naturally it's a little bit hard to tell because we know that elos can drift
over time. If think it suggests that these computers are probably not very far
in playing strength, at least they are in the same class.



    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.