supafraud

Hellllllo. I'm working on a project where I need this kind of functionality on my TP setup. I'm still a TP novice but I think I can wrap my head around your suggestions if you don't use too many big words ;-)

https://vimeo.com/8626169

Thanks in advance for any input!

littledevil

the example looks pretty faked. i guess it is just a deflector node, a strong directional
force plus some convex dummy collision mesh (note the huge gaps). an actual 'flow
around surface' compound like you can find it in the xsi or houdini libs is not part of the
c4d tp toolset. you would have to build your own and the task would include some
coding which won't fit the as easy as possible requirements.

the cheating collision node approach is pretty simplistic, it would look something like this.
a low bounce value is the key to get it somewhat realistic.

http://i.imgur.com/872ykx9.jpg

supafraud

Hey, thanks for the reply! Yeah, i've been playing with colliders and and due to the complex nature of the geometry i'm trying to avoid and the animation around the object the particles have a nasty habit of getting stuck to the surface. I also tried using the TP avoid that comes in the presets but it's not very useful because I can't assign it to the complex model. So this is something that would take some coding? Has someone created anything available for download?

littledevil

that is why i wrote the part about convex and concave surfaces in my first posting.
the whole collision approach fails for concave surfaces, as the particles get caught in
those areas. use a convex proxy mesh instead.



first of all another possible fake approach would be using splines - draw the flow with
splines around your surface and then align your particles velocity to the splines tangents.
i think there is also a preset in the tp lib for that. you could mix that method
with the collision method to get better results.

for the flow approach availability - it basically depends on what you are willing to pay
for it. generally speaking a flow around surface force is usually just part of a curl
noise solver which you could think of as a faked fluid sim. so basically you could just
buy one of the two real fluid sims for c4d (dipit or T4d) and you would have an even
better solution.

a public available curl noise / flow around surface solver do not exist for c4d to my
knowledge. the general logic behind a flow around surface force is not very complicated,
but implementing it in pure xpresso can become pretty tricky. the general idea would be
to slowly align the particles velocity in a perpendicular manner to the closest surface
normal while approaching the object. the alignment speed and the trigger distances
will define the smoothness of the flow. to 'unstick' the particle from the surface again
you can either use a simple distance traveled approach or you could define a certain
surface normal variation as the trigger to release the particle again.

fuat

as an addition to littledevils´ post (which was all correct):
technically a "flow along surface" operation take the particles velocity, the closest point/location on the geometry of the current particleposition, and the dot product of the surface normals between those both into account - that´s all.
since we have no geometry queries in XP/TP, it is not that easy to do without scripting/building it yourself from scratch. nevertheless you will have to use a lowpoly geo of your actual mesh, since XP/TP is very slow with these kind of operations (or very slow in general), especially when you have very dense mesh - you´ll have to wait for ages.

hope this hepls,
fuat

__________________
insekt8.de

CGTalk Moderation

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.

__________________
CGTalk Policy/Legalities
Note that as CGTalk Members, you agree to the terms and conditions of using this website.