[vtk-developers] Bug in scalar color mapping?

[Utkarsh Ayachit]

BTW, in my script, you can hit "t" to swap the ambient/diffuse coefficients.

On Tue, Sep 9, 2014 at 11:13 AM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> Here's my test script. With the patch applied, I don't see the ambient
> and diffuse colors bleed in anymore.
>
> On Tue, Sep 9, 2014 at 11:11 AM, David Gobbi <david.gobbi at gmail.com> wrote:
>> I added SetScalarMaterialModeToAmbientAndDiffuse() and applied the
>> patch, but it had no effect.  The results were exactly the same as in
>> my first email.
>>
>> I'll have to look through the code to figure out what the intended behaviour is.
>>
>>  - David
>>
>> On Tue, Sep 9, 2014 at 8:56 AM, Utkarsh Ayachit
>> <utkarsh.ayachit at kitware.com> wrote:
>>> Doh! Here you go.
>>>
>>> On Tue, Sep 9, 2014 at 10:53 AM, David Gobbi <david.gobbi at gmail.com> wrote:
>>>> Hi Utkarsh,
>>>>
>>>> The "attached patch" seems to be missing.
>>>>
>>>>  - David
>>>>
>>>> On Tue, Sep 9, 2014 at 8:44 AM, Utkarsh Ayachit
>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>> David,
>>>>>
>>>>> Try the attached patch and then set
>>>>> SetScalarMaterialModeToAmbientAndDiffuse() on the mapper.
>>>>> ScalarMaterialMode is the thing here. When set to
>>>>> SetScalarMaterialModeToDefault(), which is the default, it determines
>>>>> which material component should track the scalar color based on which
>>>>> coefficient is greater (see vtkOpenGLScalarsToColorsPainter.cxx:175).
>>>>> The patch fixes an issue where the painter was not respecting the
>>>>> material mode set on the mapper.
>>>>>
>>>>> Utkarsh
>>>>>
>>>>> On Mon, Sep 8, 2014 at 8:15 PM, David Gobbi <david.gobbi at gmail.com> wrote:
>>>>>> Hi Utkarsh,
>>>>>>
>>>>>> Thanks for the reply.  If I use this:
>>>>>>
>>>>>>   mapper->InterpolateScalarsBeforeMappingOff();
>>>>>>
>>>>>> then the result is exactly the same as it was before I reverted that commit.
>>>>>>
>>>>>>
>>>>>> If I use this:
>>>>>>
>>>>>>   mapper->InterpolateScalarsBeforeMappingOn();
>>>>>>
>>>>>> then the result looks like the attached image, i.e. like it did in my previous
>>>>>> email with property->SetAmbient(0.49), property->SetDiffuse(0.51), but when
>>>>>> I reverse these I don't get the greyscale ambient lighting like in the
>>>>>> right side
>>>>>> of the image from my previous email.
>>>>>>
>>>>>>
>>>>>> Still, I don't get what I expect unless I set one of either the ambient or the
>>>>>> diffuse coefficient to zero.  Whenever they are both nonzero, I get the wrong
>>>>>> result.
>>>>>>
>>>>>>  - David
>>>>>>
>>>>>>
>>>>>> On Mon, Sep 8, 2014 at 5:17 PM, Utkarsh Ayachit
>>>>>> <utkarsh.ayachit at kitware.com> wrote:
>>>>>>> This sounds familiar.
>>>>>>>
>>>>>>> Can you try reverting the following commit and see if that makes any difference:
>>>>>>>
>>>>>>> commit daf8c6e76ec7bb02f56dda17ce260141b7e960a5
>>>>>>> Author: Utkarsh Ayachit <utkarsh.ayachit at kitware.com>
>>>>>>> Date:   Fri Jul 18 15:30:37 2014 -0400
>>>>>>>
>>>>>>>     BUG #14828: Keep surface color from interacting with scalar color.
>>>>>>>
>>>>>>>     When using scalar coloring, if vtkScalarsToColorsPainter decided to use
>>>>>>>     a texture for mapping the scalars to colors, we ended up with the
>>>>>>>     surface color bleeding through the texture. This commit ensures that
>>>>>>>     such bleeding is avoided.
>>>>>>>
>>>>>>>     We don't simply use glTexEnvf (GL_TEXTURE_ENV, GL_TEXTURE_ENV_MODE,
>>>>>>>     GL_REPLACE), since that results in the diffuse highlights being lost.
>>>>>>>     By enabling GL_COLOR_MATERIAL and calling glColor3f(1,1,1), we follow
>>>>>>>     the same lighting/coloring path as we would when not using the texture
>>>>>>>     for scalar coloring.
>>>>>>>
>>>>>>>     Change-Id: I9c4db15dd0db699d5d045fa7ae27696d18828988
>>>>>>>
>>>>>>>
>>>>>>> Utkarsh
>>>>>>>
>>>>>>> On Mon, Sep 8, 2014 at 4:55 PM, David Gobbi <david.gobbi at gmail.com> wrote:
>>>>>>>> Any comments on the "scalar color mapping" problem that I sent to the
>>>>>>>> list last week?
>>>>>>>>
>>>>>>>>  - David
>>>>>>>>
>>>>>>>>
>>>>>>>> On Thu, Sep 4, 2014 at 5:33 PM, David Gobbi <david.gobbi at gmail.com> wrote:
>>>>>>>>> I've run into very odd behavior when I use Ambient and Diffuse
>>>>>>>>> coefficients in combination with vtkDataSetMapper.  This is with
>>>>>>>>> VTK master.
>>>>>>>> [ snip ]
>>>>>>>> _______________________________________________
>>>>>>>> Powered by www.kitware.com
>>>>>>>>
>>>>>>>> Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
>>>>>>>>
>>>>>>>> Follow this link to subscribe/unsubscribe:
>>>>>>>> http://public.kitware.com/mailman/listinfo/vtk-developers
>>>>>>>>