Tesla Personal Supercomputers

[Kurt Wall]

On Thu, Dec 11, 2008 at 08:20:00AM -0800, Tony Alfrey wrote:
> Kurt Wall wrote:
>> On Thu, Dec 11, 2008 at 10:49:23AM +1100, James McDonald wrote:
>>> I'm curious if our resident NVidia Employees have seen / used one of  
>>> these new "Tesla Personal Supercomputers"?
>>>
>>> http://www.news.com.au/technology/story/0,28348,24758514-5014239,00.html
>>
>> Everyday. Multiple times. They are what they are positioned to be.
>
> So does that mean you think people (universities, research environments,  
> moderate-sized businesses) will buy them?  Of course, probably also  
> making a commitment to some sort of expensive software packages that  
> take advantage of such a platform (otherwise why bother)?

Yes. In fact, people already are buying them. Lots of them. It gets
better, though. If you're running an NVIDIA card purchased in at
least the past year to 18 months, you've already got a small version
of one. In addition, you don't need to buy an expensive software
package. You can download CUDA software, examples, and tools for free
and build your own. You can write programs for them with C. That said,
there is software out there now that uses CUDA that doesn't cost much
money. But, I suck at marketing, so you're better of starting here:
http://www.nvidia.com/object/cuda_home.html.

Kurt
-- 
QOTD:
	"It's hard to tell whether he has an ace up his sleeve or if
	the ace is missing from his deck altogether."