frobzwiththingz: (waffle off)
[personal profile] frobzwiththingz
I find myself wondering: given a conceptual unit of "cpu-hours" defined in any reasonable way for todays available
hardware, what is the current worldwide ratio of {cpu spent mining Bitcoins} / {cpu spent on things like Folding@Home} ?

Date: 2013-04-11 02:43 pm (UTC)
From: [identity profile] achinhibitor.livejournal.com
It seems like one should be able to calculate it. The total number of Bitcoins must imply the computation needed to have created them. And the figures for any massively parallel project are probably available.

Date: 2013-04-11 06:20 pm (UTC)
From: [identity profile] caladri.livejournal.com
Except complicated by 'CPU hours' implying a general-purpose CPU or at least a GPGPU. With FPGAs there's still some rough analogy, but by the time you look at ASIC miners it breaks down completely. (I've used SHA2 hardware offload on network processors, which is sort of the worst of both worlds: (1) it's not that effective (2) it's really, really unclear how that fits into 'CPU hours'.)

You probably want instead to look at the hash-rate and find some way to compare the unit of generating one hash to that of performing one protein fold or whatever. Which is computationally harder in generalizable terms, etc. Then work backwards from there to construct a ratio. Except that still is probably misleading as far as looking at the computing resources available to Bitcoin vs. other massively-distributed computing projects, but it's probably good enough for a first-approximation, especially for polemical purposes.

Date: 2013-04-11 11:01 pm (UTC)
From: [identity profile] r-ness.livejournal.com
Sad, isn't it?

I imagine the ratio looks even worse if you include all the botted machines enslaved to run Bitcoin mining malware.

July 2017

S M T W T F S
      1
2345678
9 101112131415
16171819202122
23242526272829
3031     

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 26th, 2017 05:28 am
Powered by Dreamwidth Studios