How to generate colour smoke from particles with the colour of the particles

[larex39]

I have a problem to generate smoke from particles with same color as particles. (the particles has random colors)

For example I want to generete smoke from 5 particles


Can you write me pls some setup

Than you very much Alex.

[larex39]

I still cant figure out how to solve it, I was looking onto multicolor sample file but there is solution how to generete colour smoke from texture, i need this type of setup but with particles with different colours. I tryed everithing but without succes. In phoenix it is perfect feature, also fume dont have this type of colour generetion from texture, but i dont know how to do it works..
Thanks a lot for answers..

[Ivaylo Katev]

well, how you associate the color to the particle? if the color is script visible, i would recommend to use the scripting, otherwise the only way is to use a map for the uvw.

[larex39]

In particle flow I assign material static operator inside flow and put there multi subobject material with different colour materials in different slots then in material static parameter i click to choose random sub-materials..
If you can write me some steps how to do it, im not good in scripting so the best way for me is to do it without script, if it is possible.

Thank you very much Ivaylo

[larex39]

Any idea???

[Ivaylo Katev]

i made a simple scene , however the mtlid of the particles seems to be hidden, unlike the position, age, speed etc. if you need the mtlid only for the smoke, the material static can be skipped and we can assign the mtlid in the script this is the only solution that i see.
if you want the scene, send me some email

[larex39]

Thank you, for me is imortant the setup of phoenix which can produce smoke with different colour of particles. So if you can send me some scene with some particles with different colours and when i press simulate it will works, i will look of all setups you made, scripts etc.. so it would be perfect for me, because i can reaply it in my enviroment

thank you very much
larex@centrum.cz

[Ivaylo Katev]

ok, i attached the scene in a zip file, didn't know that no-image files can be attached

[wxyz]

Very helpful

Thanks

[larex39]

Thanks a lot!! this is exactly what i wanted to accomplish, now the last thing how can i render it?

[Ivaylo Katev]

ok, here is with the shading. to prevent the darking of the dissipated smoke, the injected uvw is bigger than 1 (actually is 0-255) and then using mapped output texture is clamped to 0-1.

btw - using analytic scattering with mapped diffuse color (like i just made) is wrong, the result is calculated using the fixed diffuse color, not the mapped one this is made because the mapped color makes the analytic calculation impossible.

[wxyz]

Lovely file. How does the material map "a" get transmitted to the simulation? Is it the a_inject in the script? Tried changing the colors in the map, but it doesn't seem to update.

[Ivaylo Katev]

yes, the a_inject function, the last argument is "color", it is obtained by the . in the script are used only the first 6 colors, this is a bit misleading, because the material has more submaterials.

[larex39]

ok, here is with the shading. to prevent the darking of the dissipated smoke, the injected uvw is bigger than 1 (actually is 0-255) and then using mapped output texture is clamped to 0-1.

btw - using analytic scattering with mapped diffuse color (like i just made) is wrong, the result is calculated using the fixed diffuse color, not the mapped one this is made because the mapped color makes the analytic calculation impossible.

Thank you very much, now I dont understand what you mean that the result is calculated using the fixed diffuse color, not the mapped one. It means that if i use for example bitmap file, it wont be working? Can these things be fixed somehowe? My idea was to do explosion with smoke trails from particles, which will have more colours or use bitmap (or gradient) to vary colours over time, this will make it more realistic and i thought phoenix it can do it. Can you please do some very good integraton with materials a uv channels, because it is the most coolest thing in phoenix and i dont know if there is even such fluid simulator which can do it. Thanks a lot

[Ivaylo Katev]

the concept of the analytic scattering is to avoid all the huge GI calculations in the smoke, and to replace them with simple formula that gives the luminance in given point directly from the smoke opacity and color. however, this is not possible when the diffuse color is mapped and because of this the analytic scattering is always calculated from the "simple color", even when texture or gradient is choosen as diffuse source. of course you can use it (like i made) using some averaged color, this will produce nice results despite the mathematical incorectness.