search

TACC / pvOSPRay

Other
9 stars 4 forks source link

Rendering the surfaces #25

Open mmostajab opened 8 years ago

mmostajab commented 8 years ago

It seems when a cross section of a volume is extracted, the texture on the surface is totally wrong. tmp4 screenshot from 2015-10-19 14 35 00

carsonbrownlee commented 8 years ago

Thanks for the notice on this, we will take a look.

carsonbrownlee commented 8 years ago

For the first image, what are you coloring by and can I reproduce it easily? I was able to clip the wavelet volume in half and the coloration matched, though I do not doubt you were able to find a case where it doesn't work. For the second image, it appears to only be broken for magnitude, we'll look into that, likely related to the fact that those are all 1's.

mmostajab commented 8 years ago

In the first image, the velocity magnitude is mapped to colors. You can load a vector field and extract a cross section out of it and then map the velocity to the colors.

For the second image, actually the magnitude of the normals are used and as you said they are all the same so mapped to the same color.

mmostajab commented 8 years ago

Do you need access to some test datasets?

carsonbrownlee commented 8 years ago

hi, I should have noted that I'm at Vis this week and will be preparing for upcoming SC demos as soon as I get back, so bug fixes for pvOSPRay are currently a bit on hold unless critical. Is this necessary for your application right now? If so we can get someone to work on it, otherwise I will note it on the bug list and fix it after SC.

mmostajab commented 8 years ago

That would be great! It is not critical for me right now :)

Enjoy Vis! :D

carsonbrownlee commented 8 years ago

Great, thanks! We will add it to the bug list for after SC.