On Jun 10, 2009, at 12:44 PM, Gian-Carlo Pascutto wrote:
steve uurtamo wrote:
But here is someting interesting: In the case of computer
chess it has been estimated that the progress in software
has been roughly the same as the progress in hardware.
Modern chess programs are truly amazing, and not just
a result of faster hardware. There is no reason to think
that this won't be true of computer go.
This makes me wonder... so how slow (and RAM starved)
of a computer could you use and still get grandmaster level
chess play?
In other words, how far back could we go in time if we had
today's software and expect a computer to play chess as
well as humans?
Assuming something like Rybka 3 is 3100 human-ELO on a 1 x 3Ghz
Core 2:
3100 - 2500 = 600.
Assuming 70 ELO for a doubling:
8.5 doublings
3Ghz/(2^8.5) = 8Mhz Core2. A Core 2 is a pretty nice CPU, so let's
assume we lose a factor of 2 with a more ancient design:
16Mhz ARM or MIPS
Very roughly, maybe an order of magnitude wrong.
--
GCP
We have evidence against going this low: Rybka and several other
modern engines were ported to the dedicated computers Resurrection
(203 MHz StrongArm) and Revelation (500 MHz XScale). Rybka's rating
in the SSDF pool on these platforms are 2497 and 2634, respectively.
Fruit 2.3.1 on a handheld 400MHz Xscale attained 2656 SSDF, and will
probably soon be surpassed by Pocket Fritz 3 (HIARCS) and Glaurung on
the same platform.
Prior to these systems, the strongest dedicated computer was
considered to be the TASC R40, which ran on a 40MHz ARM and attained
a rating of about 2350 in this same pool.
http://ssdf.bosjo.net/list.htm
To conclude, it appears that 500 MHz (embedded: poor cache
performance) with little memory for transposition tables is the
lowest you can go, while still staying at grandmaster level.
Ian
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
