[vtk-developers] vtkImageReslice performance questions/improvemnts

[Sean McBride]

On Thu, 23 Sep 2010 11:06:37 -0600, David Gobbi said:

>On Tue, Sep 21, 2010 at 6:58 PM, David Gobbi <david.gobbi at gmail.com> wrote:
>> On Tue, Sep 21, 2010 at 6:08 PM, Sean McBride <sean at rogue-
>research.com> wrote:
>>
>>> OK thanks.  So I'll test ppc64 and you push the x86_64 case?
>
>I pushed the ifdefs for ppc64 and x86_64.  Nothing broke :)

Thanks David.  I also tested on ppc64 and can confirm that nothing
broke.  But I did not have a chance to profile the change.  However, I
doubt it could be slower since floor() has to check for nan and other
standards-compliance things that we don't need.

For fun, I searched VTK for "i386"... there are 48 occurrences.  Taking
a look through them, I note the following:

1) vtkGridTransform.cxx seems to have copy-pasted vtkResliceFloor and
named it vtkGridFloor.  They were mostly identical, but I suppose
they've now diverged further after your commit.  This duplication
suggests that this code should move to vtkMath (or somewhere).

2) I noticed that vtkMath::Floor() has an "#if defined i386" but no
other special cases.... I wonder if it's worth adding others?

Cheers,

-- 
____________________________________________________________
Sean McBride, B. Eng                 sean at rogue-research.com
Rogue Research                        www.rogue-research.com 
Mac Software Developer              Montréal, Québec, Canada