[GRLUG] Fwd: Software GR features John Hauck: Cudafy Me - Graphics Processor Programming in C# on Windows 8

[Michael Mol]

OpenCL is the path to take. But CUDA works if you have the closed nVidia
driver.

There are a few programs (mostly graphics processing related) which can use
OpenCL or CUDA 'out of the box'. This works pretty much automagically on
Gentoo...I haven't tried it on other distros.

On Mon, Oct 15, 2012 at 1:57 PM, Casey DuBois <casey at grlug.org> wrote:

> There seems to be a ton of CUDA programming options but why no Linux
> versions lately?
>
> The last Linux article I could find was 2010:
> http://www.linux-mag.com/id/7791/
>
> Has anyone used Linux on GPU's?
>
> Is there a Linux Version that I'm not aware of?
>
> Casey
>
>
>
>
>
> On Mon, Oct 15, 2012 at 1:48 PM, Ben Rousch <brousch at gmail.com> wrote:
>
>> Casey must have read that post about how the best way to get help in
>> Linux is to troll for it. http://bash.org/?152037
>>
>> So what's a good way to approach this on Linux, and do any of you fine
>> folks have experience doing so?
>>
>>
>> On Mon, Oct 15, 2012 at 11:30 AM, Casey DuBois <casey at grlug.org> wrote:
>>
>>> Hey GRLUG,
>>> Please register if you can attend......
>>>
>>> ** **
>>>
>>> ** **
>>>
>>> John Hauck: Cudafy Me - Graphics Processor Programming in C# and Windows
>>> 8
>>>
>>> When: October 23rd, 6:00 pm – 8:00 pm
>>>
>>> Where: Atomic Object, 941 Wealthy Street SE, Grand Rapids, MI 49506
>>>
>>> Please RSVP if you'll be attending<http://softwaregr.us1.list-manage1.com/track/click?u=f3c4b0ffb20d0f119415fffd8&id=c5406ef0c6&e=0ffae3cc08>.
>>> Thanks!****
>>>
>>> The ThreadPool is fine for little dips, but this is the year to dive
>>> head first into the GPGPU - the General Purpose Graphics Processing Unit.
>>> In this talk we will be solving the traveling salesman problem by brute
>>> force. That is, we will compute the length of every possible route
>>> connecting a series of cities - to find the shortest one. We will do this
>>> entirely in Visual Studio C# using three techniques: single-threaded CPU,
>>> multi-threaded CPU, and the massively parallel GPGPU. We will spend most of
>>> the time learning how to use NVidia's CUDA with HybridDSP's Cudafy. After
>>> attending this talk you will see the universe as a massively parallel
>>> phenomenon that is ripe for you to model it.****
>>>
>>>
>>> About the speaker
>>>
>>> John Hauck has been developing software professionally in Michigan since
>>> 1981, and focused on Windows-based development since 1988. For the past 17
>>> years John has been working at LECO, a scientific laboratory instrument
>>> company, where he manages software development. John also served as the
>>> manager of software development at Zenith Data Systems, and as the Vice
>>> President of software development at TechSmith. John loves the Lord, his
>>> wife, their three kids, and sailing on Lake Michigan.****
>>>
>>>
>>>        ****
>>>
>>> Casey DuBois
>>> 616-808-6942
>>> casey at grlug.org
>>>
>>> _______________________________________________
>>> grlug mailing list
>>> grlug at grlug.org
>>> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
>>>
>>
>>
>>
>> --
>>  Ben Rousch
>>    brousch at gmail.com
>>    http://clusterbleep.net/
>>
>> _______________________________________________
>> grlug mailing list
>> grlug at grlug.org
>> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
>>
>
>
>
> --
> Casey DuBois
> 616-808-6942
> casey at grlug.org
>
> _______________________________________________
> grlug mailing list
> grlug at grlug.org
> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
>



-- 
:wq
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://shinobu.grlug.org/pipermail/grlug/attachments/20121015/f25b0bb8/attachment.html>