This looks very worthwhile!  Perhaps, as a first "proof of concept,"
just request access to the cudacluster that nvidia is evidently
running.  Then, if it looks like some high-performance multi-precision
arithmetic could actually be done using the cuda standard, submit a
larger request.

I'll go read up on CUDA.

Jason Worth Martin
Asst. Professor of Mathematics
http://www.math.jmu.edu/~martin



On Sun, Nov 23, 2008 at 3:56 PM, Bill Hart <[EMAIL PROTECTED]> wrote:
>
> Not necessarily. The C code runs on the front end processors, which
> are usually Intel Xeons or something like that. Then one shoves
> specialised routines into the NVIDIA cores. I've no idea whether it is
> there or not. But one would think that given that these things came
> from graphics, XORing should be there. The problem is, even if it is
> there, it isn't part of any FP standard, so it is not guaranteed to be
> supported on other such devices.
>
> Anyhow, I'll look up the CUDA "standard" and see.
>
> Bill.
>
> 2008/11/23 mabshoff <[EMAIL PROTECTED]>:
>>
>>
>>
>> On Nov 23, 12:50 pm, "Bill Hart" <[EMAIL PROTECTED]> wrote:
>>> I'm personally interested in the M4RI type stuff. I think malb could
>>> be as well, but the problem is XOR is not part of the IEEE floating
>>> point standard, and thus it is useless as is for M4RI. That's not to
>>> say there isn't another way....
>>
>> Well, CUDA is also about C code, so I would be very surprised if XOR
>> wasn't there (unless I totally misunderstood your point here :).  "Bit
>> flipping" should be something that a GPU is very good at, but at this
>> point this is mostly my expectations and not backed up by
>> investigation about feasibility yet.
>>
>>> Bill.
>>
>> Cheers,
>>
>> Michael
>> >
>>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"mpir-devel" group.
To post to this group, send email to mpir-devel@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/mpir-devel?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to