Re: [NTLK] Einstein Thoughts

From: Sean Luke (sean_at_cs.gmu.edu)
Date: Wed Dec 28 2005 - 17:59:42 PST


I very rarely post to NewtonTalk these days, but what the heck.

The Nokia 770 is an interesting, if odd, device, and certainly would
be a fun thing to port Einstein to -- but it's slow. So would the
Sharp Zaurus SL-C3000, Paul's on personal preference (though I am
quite underwhelmed with the CL-3000's small, ugly screen). Both
devices run Linux great and would be good first choices. But I think
ultimately a better final machine to port to would be the Palm TX.
Wireless, bluetooth, a smaller and much better-designed body,
480x320, 128MB Flash, 312 MHZ PXA270/WMMX. Actually syncs with your
Mac. And cheaper too. Under $300. And the TX's existing body of
software support is huge. The disadvantage: linux on the TX is in
its infancy (see http://hackndev.com/news.php) -- they barely got it
on the LifeDrive -- and we'd want this sitting on top of PalmOS
Garnet anyway, as there's very little available for GPE. So we'll be
waiting a little bit until it gets nice and stable.

However none of these devices is going to cut it until Paul's able to
devote some time to emulation speed issues. At present Einstein is,
I believe, doing straight CPU emulation on a virtual machine. If you
think this was slow on your Powerbook, you've seen nothing yet.
Expect it to run 1/20 the original speed.

There are two ways to handle this. First, if you're not on the same
architecture as what you're emulating, you can do dynamic
recompilation of the machine code, so things start off slow but then
get fast. There exist current ARM CPU emulators which do this (such
as ARMphetamine or QEMU). This is what Paul's got to do to get
emulation running well on the Mac. Second, if you're on the same
architecture -- as is the case for the three machines listed above --
you can run the code directly on the CPU if you can figure out how to
remap all the traps and hard-coded memory access locations. That's
not a task I'd want to have to do.

Be aware that 'tis the season for AI conference paper deadlines --
many of them in the January/February timeframe -- and this, plus the
Newton conference, will probably suck a lot of the time Paul has
right now.

Sean

-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/


This archive was generated by hypermail 2.1.5 : Wed Dec 28 2005 - 20:00:11 PST