 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Now that Intel and AMD processors have AVX instruction functionality (Sandy
Bridge, Bulldozer), how well can Povray use these new instructions? I don't have
any of these processors, so I'm unable to test it myself.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 15.10.11 17:38, jhu wrote:
> Now that Intel and AMD processors have AVX instruction functionality (Sandy
> Bridge, Bulldozer), how well can Povray use these new instructions? I don't have
> any of these processors, so I'm unable to test it myself.
That is a good question. You will need to check with your compiler-vendor
and the ability of their compiler to auto-vectorise code for AVX. You will
also need to check with your operating system vendor if the operating system
supports AVX.
As for POV-Ray itself, it (currently) contains no processor specific code,
so any SIMD optimisations depend on the compiler used to compile POV-Ray.
Thorsten
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thorsten Froehlich <tho### [at] trf de> wrote:
> On 15.10.11 17:38, jhu wrote:
> > Now that Intel and AMD processors have AVX instruction functionality (Sandy
> > Bridge, Bulldozer), how well can Povray use these new instructions? I don't have
> > any of these processors, so I'm unable to test it myself.
>
> That is a good question. You will need to check with your compiler-vendor
> and the ability of their compiler to auto-vectorise code for AVX. You will
> also need to check with your operating system vendor if the operating system
> supports AVX.
>
> As for POV-Ray itself, it (currently) contains no processor specific code,
> so any SIMD optimisations depend on the compiler used to compile POV-Ray.
>
> Thorsten
icc 12 and gcc 4.6 supposedly have AVX support. Perhaps someone here with a
processor with AVX support can compile and report back?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
