[Paraview] vector plot with multiblock dataset. possible bug or dataset issue?

[Utkarsh Ayachit]

I am not sure how you can avoid the aliasing effect, expect maybe
glyphing all points or using the "Uniform Spatial Distribution" mode
and playing the seed and maximum number of points.

Alternatively, you can try different modes. e.g. attached is a
SurfaceLIC result along with SurfaceLIC + sparser Glyph result.

On Wed, Mar 15, 2017 at 1:17 PM,  <u.utku.turuncoglu at be.itu.edu.tr> wrote:
> Thanks for your help Utkarsh. No, just opposite, I don't want to see the
> borders among blocks. And yes, actually I was talking about those bands
> that you showed in your mail. In my opinion, it is still weird to have
> those bands that exactly appears in the areas that blocks change. Anyway,
> I'll look into deeply to the data (generated by Catalyst) to find any
> problem. BTW, is it possible to remove aliasing artifacts. As I know 5.3.0
> has some improvements about that but in this case I was using 5.3.0RC2 and
> maybe those mods are not default in that particular version. Thanks again,
>
> Regards,
>
> --ufuk
>
>> Ufuk,
>>
>>> I have some weird observation with 5.3RC2 when i try to create
>>> visualisation of vector field that is coming from multiblock dataset
>>> (has
>>> 4x4 decomposition). The problem is that in the simple visualization
>>> pipeline such as Calculator -> Glyph, i could barely see or pick the
>>> boundaries of the individual blocks which is not good.
>>
>> I do not follow. Are you expecting to see the border between blocks
>> highlighted for some reason? There's no reason to expect that, unless
>> that was a characteristic of your data.
>>
>>> Then i tried to add
>>> MergeBlocks filter just after Calculator and try to visualize data with
>>> Glyph. In this case, it fixes the boundary problem (at least it seems
>>> like
>>> that).
>>
>> I don't think I see anything particularly different if I use merge
>> blocks or not. Note, if your "Glyph Mode" is not "All Points" there
>> may be minor differences between which points get selected for
>> glyphing, but that's just coincidental, and any patterns you notice in
>> that case are likely to be aliasing artifacts due to subsampling.
>>
>>> But, when i play with the Glyph Mode property and change it such as
>>> Every Nth Point and set as 3 then i could again see the boundaries with
>>> flat vectors. I just wonder that is it a bug or related with the dataset
>>> itself?
>>
>> If you're referring to the band in the attached image, then this is
>> just an aliasing artifact. Otherwise,  here too I am not sure which
>> flat vectors are you referring to.
>>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screen Shot 2017-03-15 at 1.21.28 PM.png
Type: image/png
Size: 777192 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/paraview/attachments/20170315/f1c4be2e/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screen Shot 2017-03-15 at 1.25.20 PM.png
Type: image/png
Size: 1335254 bytes
Desc: not available
URL: <http://public.kitware.com/pipermail/paraview/attachments/20170315/f1c4be2e/attachment-0003.png>