Loading...
> >

Plycoid

protovuprotovu Posts: 194
edited December 1969 in Carrara Discussion

Hi Fellow Carrararians,

I am very enthused about this new Plycoid plugin for Carrara.

One thing mentioned as possible is:

--- "Align and bank function for particle orientation along motion path" ---

This is of particular interest to me, but I am not quite seeing it in the sample files, nor can I figure it out. I believe we are distinguishing "motion path" from point force, yes? Does anyone have some guidance on this? I would be happy to do the same with the Carrara
Particle emitter, if possible - that is, describe a specific path, and have the particles travel along it.
Basically, I would like to animate some blood cells and other cells tumbling through a vessel. The cells can bump each other, but cannot merge.
I tried doing this with the Carrara particle emitter, using a curving tube, and object collision, but the particles get stuck on the various
bends.

thank you,
Rick

Comments

  • WendyLuvsCatzWendyLuvsCatz Posts: 38,193
    edited December 1969

    have you tried bullet physics for the blood too?
    I did "water" using little spheres going through a bent pipe once using bullet.

  • protovuprotovu Posts: 194
    edited December 1969

    Hi Wendy,
    I have not tried Bullit physics. I will look into it, but one thing I will need to control is the pulse flow. Don't really want a steady stream.
    Additionally, I may want to key frame a few other things in the scene. ....I may need to do this the hard way.

  • DUDUDUDU Posts: 1,945
    edited December 1969

    Hi Rick,
    You speak about “new” the plugin Pycloid…
    Is there a new version?
    I received, a few months ago email of Frederic Ribble, the originator of Pycarrara and Pycloid, and he wrote that he had given up Pycloid for the moment, considering the little of success met by the users.
    I believe that we should perhaps benefit from this topic to state that this marvellous plugin is super interesting and should be continued.
    For my part, I tested it on XP32 then on W7/64, Carrara crashes when I modify the elements of the scenes demos.
    Does somebody on this forum have some experience with Pycloid ?
    About your problem with cells moving in an artery, I believe that I would be rather of agreement with Wendy for the use of the bullet physic, but attention with the adjustments (distance from collision, materials, bouncing, gravity... and forces so necessary).
    Look at this exercise which I carried out in Carrarators forum and you should see if that it's possible.

    http://www.bond3d.byethost18.com/index.php?topic=16.msg87#msg87

  • protovuprotovu Posts: 194
    edited December 1969

    Hi Dudu,
    Sorry...Plycoid is new to me, I guess. The version I am using is 0.2.

    The potential with this Plycoid plugin is very exciting. Yes, I have hit some snags while experimenting with it, but let's hope that Mr. Ribble does not abandon.

    I will look deeper at Bullet, per your and Wendy's recommendation.

    One thing with blood flow, the force needs to follow the lumen (tube). Otherwise, the particles bunch up in bending segments of the tube. Some bunching may
    be OK, as turbulent flow is realistic. This will need to be worked out.

    Your "Stupid Machine" is terrific and inspires me to really dig in further. Thank you,
    Rick

  • DUDUDUDU Posts: 1,945
    edited July 2014

    If you are interested, you can download 2 tests (video AVI ) which I have just carried out, the first with rigid body physics and the second with soft body.
    If you are interrested, I can put the .car files this evening.
    It could be a base of work for you ?

    https://www.dropbox.com/s/vw50mwt7umepdk5/Billes.avi
    https://www.dropbox.com/s/0euytvby9xn6klh/Cholestérol.avi

    Post edited by DUDU on
  • protovuprotovu Posts: 194
    edited December 1969

    Very interested, Dudu.
    That is really generous, thank you.
    I may try to have the motion be upward. Additionally, I may want to bifurcate the flow....
    These could definitely form a base. At the very least, it will be terrific to learn from your file.
    Rick

  • DUDUDUDU Posts: 1,945
    edited December 1969

    protovu said:
    Very interested, Dudu.
    That is really generous, thank you.
    I may try to have the motion be upward. Additionally, I may want to bifurcate the flow....
    These could definitely form a base. At the very least, it will be terrific to learn from your file.
    Rick

    Here the two files .car.
    The first (with Rigid Body Physic) is very light because there are not as many calculations as for the Soft Bodies.
    Indeed, simulations are included in the files and I believe that it will be faster of downloading 144Mo than to remake physical simulation.
    With regard to the rigid balls, I simply used normal gravity, here most important is the friction of the objects (spheres and tube).
    For the soft spheres, I used forces duplicated over the entire length of the tube if not, the spheres remain agglomerated not far from the entry of the tube,
    You can modify the forces which you want for the effect that you research.
    You can also modify the parameters of folding and rigidity in the modifiers, but there, I advise you to erase all the spheres, then to make a test with one or two before duplicating them, it is faster…
    Look also at the physical parameters of the scene.
    I have make a fast test with particles, it's possible to arrive there (with forces as in the soft body example), but it's very difficult to contain them in the tube. I would like to test again when I have more time.

    Good luck! ;-)

    http://mhdproductions.be/Carrarators/RigidBody.car
    http://mhdproductions.be/Carrarators/SoftBody.car

  • protovuprotovu Posts: 194
    edited December 1969

    Thank you, Dudu.
    Downloading now. I will dig into these soon (hopefully today) , and keep you informed of my progress.
    Again, so kind of you to help me learn.
    Rick

  • DUDUDUDU Posts: 1,945
    edited December 1969

    I forgot to say that for simulation “Rigid Body”, if you select all the spheres and that you exchanges “Physic” for “Keyframes”, your simulation will be converted into keyframes for each sphere, therefore éditables individually.
    I am very interested to see your results!

  • protovuprotovu Posts: 194
    edited December 1969

    Hi Dudu,
    Thank you for the additional information.
    I have been experimenting a little bit. For me, while it is amazing what
    Bullet physics can do, waiting for calculations is too slow. This is again why
    I am thinking that some way of using a particle emitter may be a better option, though, as
    you have noted, the particle emitter is not very good at keeping particles within the tube.

    If I am going to need to keyframe the blood cells, then why not just stay with that strategy? As of now,
    I am not sure of how I want to go.

    Additionally, there is the issue of bifurcation and pulsitile flow to be dealt with.
    I need to finish some other issues, so it may be a while before I get to this and post results.
    Thank you again for your help.

  • evilproducerevilproducer Posts: 9,050
    edited December 1969

    protovu said:
    Hi Dudu,
    Thank you for the additional information.
    I have been experimenting a little bit. For me, while it is amazing what
    Bullet physics can do, waiting for calculations is too slow. This is again why
    I am thinking that some way of using a particle emitter may be a better option, though, as
    you have noted, the particle emitter is not very good at keeping particles within the tube.

    If I am going to need to keyframe the blood cells, then why not just stay with that strategy? As of now,
    I am not sure of how I want to go.

    Additionally, there is the issue of bifurcation and pulsitile flow to be dealt with.
    I need to finish some other issues, so it may be a while before I get to this and post results.
    Thank you again for your help.

    For rigid body physics (don't know about softbody) you can run the physics simulation and then if there are problem areas you can convert to keyframes and manually move the problem cells. Could be a real PITA if there are a lot.

    Some scene forces can be keyframed in both intensity and direction, so that could help with bunching. You can also tell the particle emitter to use scene forces.

    Something else to keep in mind when using physics, is the surface properties of not only the objects being directly effected by the forces, but also the objects they collide with. To keep your cells from bunching, you could make the walls of your cell have no friction. It may also be best to use a point force and go into the Scene's Physics tab and eliminate the scene's default gravity.

  • DUDUDUDU Posts: 1,945
    edited December 1969

    If you make compositing in AE for example, you can create a mask which hides the “rebellious” particles…
    I don't know exactly what you want to do, but you can do a multipass rendering and compose the whole in a program of compositing.

  • evilproducerevilproducer Posts: 9,050
    edited July 2014

    Here's an extremely quick and dirty example of using the directional force with the particle emitter to drive particle through a bifurcated vertex tube. I didn't add thickness, so the spheres look like the poke through the walls a bit, but they (mostly) follow the tube! A higher polygon tube with thickness would work better.

    I keyframed the directional force so that it changes direction. I used an oscillating tweener to automate the process. I'll link to the movie (Quicktime h.264) and the scene file for you to look at. Both are on Dropbox, so there's no need to register or anything to get them.
    https://dl.dropboxusercontent.com/u/7370483/particle_test.mov
    https://dl.dropboxusercontent.com/u/7370483/particle-test-scene.car

    Picture_1.png
    637 x 479 - 61K
    Post edited by evilproducer on
  • protovuprotovu Posts: 194
    edited December 1969

    Thanks, Evil.
    Nice of you. I like the wiggling force idea. I have been thinking though, that I would probably have more pulsitile flow control using the particle emitter without a force.

    I tried turning the force on and off......not much luck there. Might need to dig in deeper.

    In that instance, I was thinking that 2 emitters might be good for the bifurcation......but better yet might be 2 polygon particle shader domains on a vertex
    object situated strategically.

    On the particles breaking through: It seems that having an inner, invisible vessel might be the way to go. Still, I might just keyframe......not sure, haven't
    storyboarded the scene.

    Rick

Sign In or Register to comment.