Solitude Posted November 14, 2009 Share Posted November 14, 2009 (edited) Basically I'm trying to get from a particle, shoot a ray at the surface of a given polygonal object... and get some info from it. The main problem I'm having is getting geometry into the vop network. I have an intersect node and it has a file path, but instead of a file, is there no way to just pipe an object into it? ...and on top of that, what would the easiest way to get the closest point on a surface be? Edited November 14, 2009 by Solitude Quote Link to comment Share on other sites More sharing options...
Guest Swann Posted November 14, 2009 Share Posted November 14, 2009 I have an intersect node and it has a file path, but instead of a file, is there no way to just pipe an object into it? Read this Quote Link to comment Share on other sites More sharing options...
pclaes Posted November 14, 2009 Share Posted November 14, 2009 easiest probably ray sop, another way is converting your geometry into an SDF and getting the gradient and distance out in vops. Quote Link to comment Share on other sites More sharing options...
Solitude Posted November 14, 2009 Author Share Posted November 14, 2009 (edited) Read this Sweet, thx... I'll give it a go in a bit. As for the ray sop.. how can I use the info from the raysop within a pop network? ie: I have a pop network, with all of my particle stuff going on, how would I get any info from the ray sop back into my pop netowork since all the points are being piped out of the pop network into the ray sop? As for the sdf... I knew I could do this, but was curious how to do this with geometry before I dove into it. What are the advantages of using the volume gradient vs intersect nodes for getting attributes / info off geometry? ...the main ones I usually would use being pos, normal, and color from a map... Edited November 14, 2009 by Solitude Quote Link to comment Share on other sites More sharing options...
pclaes Posted November 20, 2009 Share Posted November 20, 2009 woops, I meant intersect pop. ... my mind is somewhere else . -I'm currently using sops as if they were pops using a sopsolver. It's slower, but much more control as I tend to do most of my operations in vops or vex (as you can then use pointclouds for attribute transfers/filtering and use multithreading which makes a big difference if you got several cores). Also all the sop operations become "pops" - and you can scrub the timeline . Using an SDF can be faster and lighter for approximations and could provide a level of detail control. When a point is far away, it uses the center of the bbox, when it is closer, it uses a rough sdf, when it is even closer it uses a high detailed sdf. If your mesh is static, it is a matter of looking up the value, rather than performing a ray calculation. The lookup is basically an interpolation between the distances stored on the cornerpoints of the voxel. This interpolation is a bunch of multiplications and additions, which is a lot lighter than a distance calculation in the ray calculation. Probably does not make that much difference on a few thousand points, but more than that should give more obvious speed gains. If you are dealing with maps and textures the sdf's are probably less of an interesting option.. although... you could attribute transfer your color information from the points on the surface to volume fields and get "deep color fields" that way you would be able to read the color values of a surface inside of the object. eg: imagine a round stone, you put 3 different oil paints on the surface of the stone. The stone starts to suck up the colored oil to the center of the stone. Now slice the stone in half and you will see the surface colors have made their way to the center. -not sure if that makes sense to you . 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.