Spreading of a dynamic texture (color)

vvvv beta question
1

Hi guys, thanks to joreg i managed to solve my previous problem on spreading of a halo ring animation here: https://discourse.vvvv.org/t/7287 but now i have some more probs on spreading another dynamic texture…

it’s easier than my previous patch but now i am using a dynamic texture (color) and not (value) - by the way i have tried with a dynamic texture (value) too and does not work either.

again…thanks everyone!

dynamic texture spread.v4p (10.1 kB)

2

hard to guess what it’s supposed to look like, but, you did a couple of strange things there…

  • you spreaded the alpha channel and put the texture on a sphere. that will most likely lead to artefacts
  • the topmost linearspread was parameterized in a way that won’t fit an alpha channel well
  • the next linearspread was spreaded to a random number not fitting the upper spread
  • on the dynamictexture you spreaded both width and height, resulting in 256x256 pixels per texture but only fed the node with 512 values.

all those together likely make the output unpredictable. i cleaned up the patch a bit, see attached.

dynamic texture spreaded.v4p (10.3 kB)

3

thanks joreg for helping me again, the main mistake that leaded me to write a post was that i spreaded both width and height of the dynamic texture…i agree with you: difficult to imagine what i wanted to achieve, i dunno either it was just testing out ;)

by the way i cannot understand the first 2 statements you say about the alpha channel…if you pass by and have some time i’d be happy if you say something more about it…

thanks again

4

ao,

ad 1)
transparent textures on double-sided meshes (like a sphere) are problematic since rendering with transparency requires such polygons to be drawn last always which is not automatically handled by a mesh and thus results in artefacts easily. simple workaraound: draw mesh twice with opposing cull modes, ie. first the backfacing polygons, then the frontfacing. i think that does in most cases.

ad 2)
alpha-channels are parametrized from 0 to 1 and you set it some values from -something to >1 which while not causing any harm (since values are clamped there anyway) just doesn’t help in understanding what the result would look like.

hope that helps.

5

thanks that surely has helped, didn’t know the cull node!