Re: Why no OpenCL output on iMac OS X Lion?
- To: mathgroup at smc.vnet.net
- Subject: [mg125309] Re: Why no OpenCL output on iMac OS X Lion?
- From: George Woodrow III <georgevw3 at mac.com>
- Date: Tue, 6 Mar 2012 05:56:00 -0500 (EST)
- Delivered-to: l-mathgroup@mail-archive0.wolfram.com
- References: <201202291221.HAA28305@smc.vnet.net>
OK.
I did the basic examples from both the documentation center and the WTC 2011 talk on using the GPU.
Everything seems to be working just fine.
I am running OS X 10.7.3, Mathematica 8.0.4, and Xcode 4.2.1. I am using a 3 year old MacBook Pro 17 inch, which OpenCLInformation reports as
{"Intel(R) Core(TM)2 Duo CPU T9550 @ 2.66GHz", "GeForce 9600M \
GT", "GeForce 9400M"}
The computer is set up to use the dedicated graphics card.
I hope to try out some non-trivial coding soon, but I am satisfied that the basic functionality works on my system
george woodrow
On Mar 1, 2012, at 5:33 AM, Murray Eisenberg wrote:
> I have XCode installed already, of course: version 4.2.1.
>
> On 2/29/12 8:03 AM, George Woodrow III wrote:
>> I think that you will need to install XCode.
>>
>> I had a recent communication with Wolfram tech support, and they indicated that to use the functionality to generate code and to use OpenCL, you need to have Xcode installed. There are open source options as well, but since Xcode is now free through the app store (for Lion), there is no real reason not to just install the program.
>>
>> Actually *testing* OpenCl within Mathematica is on my list of things to do.
>>
>> george woodrow
>>
>>
>>
>> On Feb 29, 2012, at 7:21 AM, Murray Eisenberg wrote:
>>
>>> I'm running Mathematica 8.0.4 on the latest, 2011 iMac (3.4 GHz Core i7,
>>> 16 GB RAM) under OS X Lion (10.7.3). The graphics adapter is an AMD
>>> Radeon HD 6970M (2048 MB VRAM).
>>>
>>> After loading OpneCLLink, OpenCLQ[] returns True.
>>>
>>> Yet evaluating examples from the ref page for OpenCLFractalRender3D
>>> returns nothing. E.g., OpenCLFractalRender3D[].
>>>
>>> Moreover, it seems as if the OpenCL link is being closed, or the package
>>> unloaded, immediately after I try such an evaluation -- because if I
>>> evaluate it again, now I get the input expression itself returned; and
>>> indeed the $ContextPath no longer includes OpenCLLink (or the SymboliC
>>> and CCompilerDriver automatically added by loading OpenCLLink).
>>>
>>> What could be wrong?
>>>
>>> --
>>> Murray Eisenberg murray at math.umass.edu
>>> Mathematics& Statistics Dept.
>>> Lederle Graduate Research Tower phone 413 549-1020 (H)
>>> University of Massachusetts 413 545-2859 (W)
>>> 710 North Pleasant Street fax 413 545-1801
>>> Amherst, MA 01003-9305
>>>
>>
>
> --
> Murray Eisenberg murray at math.umass.edu
> Mathematics & Statistics Dept.
> Lederle Graduate Research Tower phone 413 549-1020 (H)
> University of Massachusetts 413 545-2859 (W)
> 710 North Pleasant Street fax 413 545-1801
> Amherst, MA 01003-9305
>