catchyid Posted November 10, 2016 Share Posted November 10, 2016 (edited) Hi, I want to simulate/render a missile smoke trail, I've used initially particles + sprites, but the result was not really satisfactory (see attached). My question here: I want to produce realistic film quality smoke trail, is it possible to do that using particles + sprites or I will just be wasting my time? I am still learning Houdini, but I think simulating smoke physics using particles + expressions/forces won't be possible, also rendering sprites to mimic real smoke is possible but would be challenging. I prefer to use particles+sprites though because I think they will be faster to simulate compared to pyro simulation, specially that the missile is traveling a vast distance and hence the fluid grid will get really big and too slow to simulate!? So in summary, should I keep trying with Particles + Sprites to produce film quality missile smoke trail OR I should switch to Pyro? Thanks Edited November 10, 2016 by catchyid Quote Link to comment Share on other sites More sharing options...
gq_charles Posted November 10, 2016 Share Posted November 10, 2016 I am not sure about the particle way, but if you are switching to pyro, you can use multiple containers along the missile path and trail SOP to help with the substeps. 1 Quote Link to comment Share on other sites More sharing options...
Atom Posted November 10, 2016 Share Posted November 10, 2016 (edited) Here are a couple of example files to check out. One demonstrates copy/instancing of volumes along points. Try out the various switch options. The other one is really cool in that it demonstrates smoke clustering by instancing volume domains. This is how you keep your volumes from getting really large. ap_cluster_smoke_on_path_128.hipnc ap_popnet_missile_trail_1e.hipnc Edited November 12, 2016 by Atom 4 Quote Link to comment Share on other sites More sharing options...
TheDude Posted November 10, 2016 Share Posted November 10, 2016 Another option is to just get nice particle motion from the trail, then vdb rasterize particles, and you can cluster by distance from cam for volume resolution staggering 2 Quote Link to comment Share on other sites More sharing options...
catchyid Posted November 13, 2016 Author Share Posted November 13, 2016 Thanks guys... I think I must use pyro to get nice simulation/rendering. Thanks Atom for your teapot example, breaking down pyro simulation into multiple container is essential here to avoid having one big container, but I have a question though: would these container move fields between them, i.e. if temperature is rising at the border of one container, would the neighbor container catches that temperature or I have to write some code to move fields across the borders? Thanks Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.