Seti@Home optimized science apps and information
Optimized Seti@Home apps => Discussion Forum => Topic started by: cristipurdel on 20 Nov 2010, 01:28:39 pm
-
Saw this a couple of days ago: http://software.intel.com/en-us/articles/intel-opencl-sdk/
Would it be possible to develop an app using the intel opencl sdk, so that it can run also on their Execution units?
Here it say that they'll support OpenCL 1.1. To have a smooth transition for Larabee maybe :)
Still, if there will be opencl running on intel Sandy Bridge, I would really love a to see a battle with Zacate for the low-end devices :).
-
I downloaded it.
But it is still for testing.
Q – Can I use this version of the Intel® OpenCL SDK to release commercial OpenCL products?
A – No. The current version of the Intel® OpenCL SDK is only a technology preview and can only be used for exploration and validation. You cannot release standalone applications that use this version of the SDK.
~~~~~~~~~~~~~~~~~~
so far I will try and explore it
heinz
-
so far I will try and explore it
heinz
Dreaming of a 17W Sandy Bridge running 2 + 2 + 12 seti tasks :P
-
Would it be possible to develop an app using the intel opencl sdk, so that it can run also on their Execution units?
In theory already released OpenCL apps should work w/o rebuild. But...
-
Dreaming of a 17W Sandy Bridge running 2 + 2 + 12 seti tasks :P
do you know if the integrated GPU of sandy/-ivy bridge will support OpenCL - or only the CPU?
-
Dreaming of a 17W Sandy Bridge running 2 + 2 + 12 seti tasks :P
do you know if the integrated GPU of sandy/-ivy bridge will support OpenCL - or only the CPU?
http://en.wikipedia.org/wiki/GMA_HD#HD_Graphics_.28GMA_HD.29 ;)
-
... Number of CL Platforms : 3 on an intel + nvidia system :P
Apparently OpenCL is the ...future :)
-
1) I would like to try AMDs OpenCL CPU implementation (because of printf(), etc. *g*). Are there any limitations when using the CPU device compared to the GPU device? I think local work groups was limited to 1 for CPU, no?
2) Another question: When I change local work group size, does it have an impact on the algorithm itself (besides performance)?
-
Not right thread for such questions IMO, but :) :
1) max workgroup size remains the same.
Look into stderr of AP task, it liss device parameters.
2) yes, algorithm can change indeed. For example, look my post about why NV MB build should not work with workgroup size of 32.
[more technically, kernel aware of its launch parameters, in particular, about workgroup size, via get_local_size() call.
So, surely it can and sometimes will behave differently for different workgroup sizes]
-
So now OpenCL on Intel EUs....grrrr