[vtkusers] Bug in vtkVolumeMappers

[bryan cole]

Hi All,

First up, what's up with the Kitware bug-tracker? I tried to create a
new account there and nothing happened. The create-new-account php-page
just reloaded itself with no acknowledgement and I've not received any
email to indicate I've now got a bug-tracker account.

So here's my bug report:

The vtkVolumeMappers do not position volumes correctly in world-space,
when the lower limits of the Output Whole Extent are anything other than
0,0,0.

Thus, any filters which modify the output whole extent, like
vtkExtractVOI or vtkImageClip cause the volume to move with respect to
any other geometry. I've attached a python script to illustrate the
problem. (Even worse things happen for vtkExtractVOI, if you change the
sample rate, but let's not get into that). 

An Alternative viewpoint is that the VolumeMapper is correct, but the
calculation of the Origin for the Output of the vtkImageClip (or
vtkExtractVOI or any other filter which alters the lower Whole-Extents)
is in error.

The ambiguity here (which would be the underlying cause of this problem)
is that there's no formal design specification for how the position of a
dataset is defined. Is it by the Origin (as returned by
dataset.GetOrigin()), or is it the bounds, or is it the DataExtents (in
the case of image-data)? This duplication of position-information has
led to different vtkObjects taking the data positions from difference
sources such that they are not consistent. We need a definitive design
spec! What IS the fundamental definition of a dataset position in world
space?

I'm using VTK4.2

Bryan







-- 
Bryan Cole
Teraview Ltd., 302-304 Cambridge Science Park, Milton Road, Cambridge CB4 0WG, United Kingdom.
tel: +44 (1223) 435380 / 435386 (direct-dial) fax: +44 (1223) 435382
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BrokenBoundsDemo.py
Type: application/x-python
Size: 1941 bytes
Desc: not available
URL: <http://www.vtk.org/pipermail/vtkusers/attachments/20040803/abbcd81a/attachment.bin>