A nice machine for mining may not exist without your support

[Gareth Nelson]

http://www.kickstarter.com/projects/adapteva/parallella-a-supercomputer-for-everyone

[1715341352]

1715341352

[1715341352]

1715341352

[1715341352]

1715341352

[1715341352]

1715341352

[Graet]

https://bitcointalk.org/index.php?topic=115021.msg1273871#msg1273871
https://bitcointalk.org/index.php?topic=120471.0
both say it wont be good for mining but might be ok as a host for fpga or asic devices

[Stephen Gornick]

https://bitcointalk.org/index.php?topic=115021.msg1273871#msg1273871
https://bitcointalk.org/index.php?topic=120471.0
both say it wont be good for mining but might be ok as a host for fpga or asic devices

also

Can this be used for Bitcoin mining?
 - http://bitcointalk.org/index.php?topic=117615.0

and

Parallella: A Supercomputer For Everyone
 - http://bitcointalk.org/index.php?topic=119140.0

and

Parallella
 - http://bitcointalk.org/index.php?topic=114996.0

and

Parallella: A Supercomputer for Everyone (Ending Saturday Oct 27, 6:00pm EDT)
 - http://bitcointalk.org/index.php?topic=120470.0


[Edit:
Now, if you want a rig that might work as a decent Bitcoin miner:

My Next Mining Rig
 - http://bitcointalk.org/index.php?topic=121154.0 ]

[Zeek_W]

Everyone's too busy pre-ordering ASICs, no spare cash for these computers (although they ARE cool)

[ElectricMucus]

Processor Arrays are the next big thing in computing yes.

But those currently available do not meet the necessary size for some serious mining and are more in the range to be used for gadgets with strict power requirements, low cost which are dsp intensive.
They are scalable and in time will replace current architectures as they are more power efficient and easier to program than fpgas with the same feature size, but it is a long way ahead.
FYI You can actually buy these things right now from the competition, greenarrays. They offer 144cores with 18bit.

I've even done some knee-jerk estimation on their theoretical mining efficiency, its not feasible, at least with these chips. They aim for a much smaller feature size which could make it worthwhile at least from a power efficiency standpoint if they offer 32bit data-width (seems to be 16 bit though). On the other hand if that technology would be actually used to create a 28nm 32 bit version with 9001 processors it should blast fpgas.

[Stephen Gornick]

The $750K goal on Kickstarter was reached (and now over $850K):
 - http://arstechnica.com/information-technology/2012/10/99-raspberry-pi-sized-supercomputer-hits-kickstarter-goal

[psilan]

There are so many of these things now, after the Pi's success. It's amazing they still get funding.

[dust]

Why are their multiple threads in the mining section pumping this device?  Even GPUs will crush it in terms of Mh/J and Mh/$.

[koin]

Why are their multiple threads in the mining section pumping this device?  Even GPUs will crush it in terms of Mh/J and Mh/$.

for a lot of things a gpu does a way better job than this device ever could.  but this is another example of the power of the crowd, and kickstarter seems to be perfecting the art of tapping that force.

[ChipGeek]

I do not believe this will be at all useful for mining.  16 processors at 700MHz each = roughly the same as 4 processors at 2800MHz (2.8GHz) which is any modern PC CPU. 

So would you be excited about a new quad core CPU running at 2.8GHz from Intel or AMD?  I certainly would not be.

[ElectricMucus]

If it costs a buck an burns a watt I'd sure be