[vtkusers] vtkVolumeTextureMapper3D behaves differently between v5.1 and v6.2

Simon ESNEAULT simon.esneault at gmail.com
Mon Sep 14 10:48:53 EDT 2015


Hello,

Yes we were suffering the same issue.
I believe the change occurred with this commit
https://github.com/Kitware/VTK/commit/0164fa1e25d9c584e7ec60acb7e221f7ea9e5d7c
The 128x256x256 thing is no longer hard coded, it now relies on
the vtkgl::MAX_3D_TEXTURE_SIZE variable to compute a correct reduction
ratio, which will lead to this significant slow down.

Some more explanations :
http://marc.info/?l=vtkusers&m=128869084800678

The Volume Texture Mapper 3D has just been marked as deprecated so I don't
think there will be any sort of fix, but you should be able to restore the
old behavior if you look to the commit, and revert the changes made in the
method
int vtkOpenGLVolumeTextureMapper3D::IsTextureSizeSupported(int size[3], int
components){...}

Best Regards
Simon



2015-09-14 15:55 GMT+02:00 Weiguang guan <guanw at rhpcs.mcmaster.ca>:

> Hi David,
>
> I'll try vtkImageResize. Thanks.
>
> Best,
> Weiguang
>
>
> On 11/09/2015 1:48 PM, David Gobbi wrote:
>
> Hi Weiguang,
>
> You can do the image resizing as a preliminary step with vtkImageResize,
> which does efficient band-limited sinc interpolation and can do a good job
> of shrinking the image without introducing aliasing artifacts.
>
> Hopefully someone else will answer your question about
> vtkVolumeTextureMapper3D (I haven't used it in a while).
>
>  - David
>
> On Fri, Sep 11, 2015 at 11:22 AM, Weiguang guan <guanw at rhpcs.mcmaster.ca>
> wrote:
>
>> Hello VTK users,
>>
>> We develop a viz system using vtkVolumeTextureMapper3D for volume
>> rendering. Recently, we've moved up from v5.1 to v6.2. After that, we
>> notice our viz system has a significant slow-down in rendering speed
>> (compared on the same computer) when volume size is large, say LxWxH ~= 512
>> ^3. It can always render small volumes at interactive rate.
>>
>> I know that vtkVolumeTextureMapper3D needs to resample large volume to
>> fit in 128*256*256 (BTW, our data is regular volume data with a single
>> component). But this down-sampling process is not new in 6.2. Are there
>> other changes that cause the performance issue? Does anyone experience the
>> same issue? Thanks.
>>
>> Best,
>> Weiguang
>>
>
>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the VTK FAQ at:
> http://www.vtk.org/Wiki/VTK_FAQ
>
> Search the list archives at: http://markmail.org/search/?q=vtkusers
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/vtkusers
>
>


-- 
------------------------------------------------------------------
Simon Esneault
Rennes, France
------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtkusers/attachments/20150914/96e5b49b/attachment.html>


More information about the vtkusers mailing list