I checked the forums to see if there was any mention of a topic such as this, so please dont flame me if I missed it..

Some friends and I were discussing the fact of the millions of mobile devices out there sitting idle. Most newer pda's are approaching near PC computing power, several upcoming cellular devices with television capability, 100-400mhz processors (roughly - for some of the pda's i've seen)..all spending 99% of the day sitting idle on your belt loop.

With obvious issues such as memory constraints (and that wont be long before your phone has fat storage/ram) aside..Thats a lot of computing power when you multiply it times the millions of phones/pda's out there. My nokia, for example, spends all day showing a clock with the minutes ticking by.

The question I bring up is: Practicality aside, What sort of magnitude of computing power would we be talking about, if there was some form of Seti for J2ME/OTA devices? (Think about the cellular hardware 2 years from now - its putting a lot of power in a small space.)

Distributed Computing for Mobile Devices - Silly, or a possible major souce of cpu cycles? Discuss? :D

There's only one thing holding such an idea back: Battery. You'll need more amp-hours than what the regular LiIons are able to supply.

From my understanding, cell phones usually haven't got more than a meg or so on the cheap-o models, and most of the hardware/processor calls are rudimentary at best. I heard a rumour (probably about two years ago) that Motorola cells use PowerPC 603s. If that's the case, then it's plausible to get some crunching done on a cell phone.

I doubt that a Nokia will crank out FFTs like a 486 can, but who knows?