Ships with free open source Epiphany development tools that include C compiler, multicore debugger, Eclipse IDE, OpenCL SDK/compiler, and run time libraries.
Too bad i was hoping to see ships with openSUSE. What is Ubuntu OS? It is just GNU Linux.All along i have been thinking that it is just another distro.
I am not here to promote Ubuntu, but this ARM based board was brought to my attention via the OpenBSD misc lists and it caught my eye. Obviously if it works with Ubuntu, it might be a good item for OpenSuSE developers to toy with as well. The idea of a “supercomputer” on a card the size of a small book is rather amazing.
On Fri, 19 Apr 2013 14:26:01 +0000, RichardET wrote:
> I am not here to promote Ubuntu, but this ARM based board was brought to
> my attention via the OpenBSD misc lists and it caught my eye. Obviously
> if it works with Ubuntu, it might be a good item for OpenSuSE developers
> to toy with as well. The idea of a “supercomputer” on a card the size
> of a small book is rather amazing.
It’d be interesting to see the specs vs. a GPU-based supercomputer
cluster.
Am 19.04.2013 20:28, schrieb Jim Henderson:
> It’d be interesting to see the specs vs. a GPU-based supercomputer
> cluster.
an average GPU like the one in my PC is much faster, but I do not think
that is the point about this device it is of educational value
–
PC: oS 12.3 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.10.0 | GTX 650 Ti
ThinkPad E320: oS 12.3 x86_64 | i3@2.30GHz | 8GB | KDE 4.10.0 | HD 3000
HannsBook: oS 12.3 x86_64 | SU4100@1.3GHz | 2GB | KDE 4.10.0 | GMA4500
On Fri, 19 Apr 2013 18:52:39 +0000, Martin Helm wrote:
> Am 19.04.2013 20:28, schrieb Jim Henderson:
>> It’d be interesting to see the specs vs. a GPU-based supercomputer
>> cluster.
> an average GPU like the one in my PC is much faster, but I do not think
> that is the point about this device it is of educational value
Yeah, the more I thought about it, the more I thought that was probably
the case - the affordable GPUs that I’ve looked at have 1500 cores,
whereas this has 16 or 64 cores.
Am 19.04.2013 21:22, schrieb Jim Henderson:
> On Fri, 19 Apr 2013 18:52:39 +0000, Martin Helm wrote:
>
>> Am 19.04.2013 20:28, schrieb Jim Henderson:
>>> It’d be interesting to see the specs vs. a GPU-based supercomputer
>>> cluster.
>> an average GPU like the one in my PC is much faster, but I do not think
>> that is the point about this device it is of educational value
>
> Yeah, the more I thought about it, the more I thought that was probably
> the case - the affordable GPUs that I’ve looked at have 1500 cores,
> whereas this has 16 or 64 cores.
>
> Jim
>
you just need to look at the floating point performance (if interested
in number crunching) my card is already in the 1 TFlops range.
The benefit of that Parallella device is to provide a very open piece of
hardware + software stack, while GPU computing with nvidia and amd cards
is more working on a completely closed black box.
–
PC: oS 12.3 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.10.0 | GTX 650 Ti
ThinkPad E320: oS 12.3 x86_64 | i3@2.30GHz | 8GB | KDE 4.10.0 | HD 3000
HannsBook: oS 12.3 x86_64 | SU4100@1.3GHz | 2GB | KDE 4.10.0 | GMA4500
On Fri, 19 Apr 2013 19:35:46 +0000, Martin Helm wrote:
> you just need to look at the floating point performance (if interested
> in number crunching) my card is already in the 1 TFlops range.
Yeah, and that’s what I’m looking at - I like to dabble in 3D rendering a
bit.
> The benefit of that Parallella device is to provide a very open piece of
> hardware + software stack, while GPU computing with nvidia and amd cards
> is more working on a completely closed black box.
Very true. But for GPU rendering with Blender, nVidia is really the only
option to use at this point.
Am 19.04.2013 21:40, schrieb Jim Henderson:
> Very true. But for GPU rendering with Blender, nVidia is really the
> only option to use at this point.
guess why I have the GTX card since I do not play 3d games it is just an
affordable way to get decent performance for numerical calculations for
which the i7 is simply much too slow
–
PC: oS 12.3 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.10.0 | GTX 650 Ti
ThinkPad E320: oS 12.3 x86_64 | i3@2.30GHz | 8GB | KDE 4.10.0 | HD 3000
HannsBook: oS 12.3 x86_64 | SU4100@1.3GHz | 2GB | KDE 4.10.0 | GMA4500
On Fri, 19 Apr 2013 19:46:35 +0000, Martin Helm wrote:
> Am 19.04.2013 21:40, schrieb Jim Henderson:
>> Very true. But for GPU rendering with Blender, nVidia is really the
>> only option to use at this point.
>
> guess why I have the GTX card since I do not play 3d games it is just an
> affordable way to get decent performance for numerical calculations for
> which the i7 is simply much too slow
That’s what I’m thinking, too - I’ve got an i7 in my laptop, but
rendering a simple image with glass and reflective textures at a
reasonable resolution takes 20-30 minutes at a minimum.
I don’t have any real need for this much power at this time but for such a small device, I find it to be amazingly powerful. I am curious as to why ARM technology as opposed to x86 based. It seems like x86 based processors have more punch though most certainly use more power.
Am 20.04.2013 05:46, schrieb futureboy:
> My Core 2 Duo must be a dog to you.
It always depends what you want to do, I used long time an atom netbook
and I am a big friend of using older hardware as long as it does its job
in a reasonable way.
>
> I don’t have any real need for this much power at this time but for
> such a small device, I find it to be amazingly powerful. I am
> curious as to why ARM technology as opposed to x86 based. It seems
> like x86 based processors have more punch though most certainly use
> more power.
>
And that looks like one of the design decisions. If I have not got it
wrong this device uses 5W (!). That is impressive for its performance.
Would make for sure a brilliant “coprocessor” for a mobile device or
some embedded system with high performance signal processing needs or
similar.
I think one reason they chose ARM is due to the power consumption.
–
PC: oS 12.3 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.10.0 | GTX 650 Ti
ThinkPad E320: oS 12.3 x86_64 | i3@2.30GHz | 8GB | KDE 4.10.0 | HD 3000
HannsBook: oS 12.3 x86_64 | SU4100@1.3GHz | 2GB | KDE 4.10.0 | GMA4500
