[Bf-committers] __APPLE__ using doubles sqrt, instead of sqrtf ? Can we optimize Apple OS X Intel?

Ton Roosendaal ton at blender.org
Wed Dec 6 13:45:08 CET 2006


Hi,

This is only supported for gcc4 and later. Our releases were stil gcc3,  
reason why it cannot be added simply.

AFAIK sun cc (and irix cc) do have float.h math for ages. Blender was  
using this all over in the past, but it was removed in the NaN days  
(2000-2001) to comply to linux gcc better.

To make the migration to float math nice again, it should be tackled  
much friendlier... preferably with a wrapper header file redefining all  
sqrtf() and friends to the double counterparts (if not available).

Wouldn't that work better?

-Ton-


On 6 Dec, 2006, at 5:21, erwin at erwincoumans.com wrote:

>
> Hi,
> This is in my Bullet code, for Blender. I want to enable the single  
> precision faster floating point versions.
> #if defined (__sun) || defined (__sun__) || defined (__sparc) ||  
> defined (__APPLE__)
> //use double float precision operation on those platforms for Blender
>
> If this is not possible, how can we optimize at least for Apple OS X  
> Intel?
> I want to add SSE/SIMD in Bullet for some platforms, in particular  
> Apple Intel. Can we use some ifdef for that?
> Thanks,
> Erwin
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-committers
>
>
>
------------------------------------------------------------------------ 
--
Ton Roosendaal  Blender Foundation ton at blender.org  
http://www.blender.org



More information about the Bf-committers mailing list