Cineversity

After seeing the breathtaking 2016 AICP Sponsor Reel (https://vimeo.com/169599296), and now that Voronoi fracturing is available in Cinema 4D, I’ve been hard at work trying to reproduce the reverse fracturing-girl effect found at 1:21 into the AICP Sponsor Reel.  Despite approaching this from a variety of angles, I’m not succeeding.  I can’t figure out how to get the chunks to start out in a perfectly still position resting on the ground, then a few seconds later begin building the dancing girl.  My chunks, which are associated with the dancing girl, just want to keep on dancing around on or near the ground no matter what I do.  I’m wondering if anybody has advice to help me accomplish this effect?  I’ve tried to attach my C4D file, but even zipped it is more than the 4 MB limit.

Thanks so much!

Mark

Hi Mark,

It shows that these are obviously done with “Particle Animation”, speaking of the majority.

(Which needs to sort the chunks in y for example)

Beside the tutorials here on Cineversity as well as the Siggraph presentations, I guess you know them well already, let me add a very simple way to it. ;o) Perhaps too simple…! I just created it based on your question.

The Dynamic Tag can be triggered with a MoGraph Selection, which can be changed in XPresso.

The idea is to set the Dynamic “trigger” to velocity, which is introduced by the Plain Effector. To avoid vertically shooting “chunks” I paced a collider on top of that, with a vibrate tag, to make it even invisible as such (change of motion). (Perhaps use a Inheritance to supply progressively (falloff) animation to the chunks, as shown in the Siggraph presentation)

I will take a closer look in to it - if the tutorials and presentations, or my little contribution, will not be the one you are looking for.

All the best

Scene file
https://www.amazon.com/clouddrive/share/mQrvoCX3lup5darZf7CDSzF3RjhmdmqgndpR63inyMG?ref_=cd_ph_share_link_copy

Thanks so much for your reply, Dr. Sassi!  I’ll take a look at your comments and file to see if I (as an advanced beginner) can figure this out.

Mark

Dr. Sassi, I took a look at your file and understand what you did.  It’s similar to what I did with a Random Effector and its Falloff parameter in my animation.  My challenge, however, is having the particles sit still on the ground plane at the beginning of the animation, then fly up and build the dancing girl as the animation progresses.  The fact that her materials make up the particles, and she is in constant motion, are the reasons that I can’t figure out how to get the particles to sit still at the beginning of the animation.  I’m also struggling to figure out how to get the particles to the ground right below her in the first place since she is constantly moving and launching dynamic particles away from her.  Does this make sense?

Thanks!

Mark

Hi Mark,

Sorry I wasn’t aware that the “dancing around part” was the key problem, sorry about that! I hope this one comes closer. (The camera move is just a quick sketch ;o)

If you have the impression that your scene file would give me more insight, then let me know, I will create an upload link for you.

My best wishes

Scene file:
https://www.amazon.com/clouddrive/share/LILUdIDHBzcw7uapJiTYWuaM5uAWsPrtZgD6PHsEMOV?ref_=cd_ph_share_link_copy

Really appreciate your willingness to help, Doc Sassi!  In my case, the dancer is actually in motion and there is no camera move.  I’m trying to recreate an animation very similar to what you see in the 2016 AICP Sponsor Reel (https://vimeo.com/169599296) at the 1:21 mark.  I would love to share my scene file with you.  The part that is missing from it are the particles at rest on the ground in the beginning.  At around 30 frames, the Random Effector successfully assembles the particles into the dancer, although she has ugly fine cracks running through her materials when rendered. 

Yes, please create an upload link for me.  File size is 20 MB.

Thanks!

Mark

Just uploaded the file, Doc.  It’s really nice to know I’m not alone out here.  Thanks so much for helping with this!

Mark,

Thanks for the file. (Normally I mention it: reduced the scene to the minimum of the problem. The HDRI box, which doesn’t matter to me here, took the majority of the file size.)
=====

I think at one point you should just replace the full body, to avoid the “fine cracks”. If the Voronoi works only during the start. In the scene I got for around 120 frames. Perhaps blend it in post…

=====

The “space” that is split into “chunks” is changing with the animation. To have the same stability in the Voronoi chunks, the figure would have to editable, which would be a loss of functionality. Which is an interesting problem. I have created a file that shows the problem and how I would solve it. Since the Skin is a deformer, I would do so, but it needs weighting, and they the Voronoi is not possible anymore. OR: you create a rough mesh around the fractured figure and have the joints/skin work on this mesh. Then use a Mesh-Deformer for the Voronoi. Edit: see post below for an example./edit

Since I have not the T-Pose in this file (no T-pose joints to be precise) to “Bind” the body after fracturing it. (Make the Voronoi editable in T-Pose and then Bind/Weight the character newly)
Another option which needs to be explored is the VAMP manager, but since we produce points inside of the character…

When this is done, the Character can go under a Fracture Object, and the Random Effector can work on it (while Fracture is set to Explode Segments and Connect)

In this way, the amount of “chunks” stays the same and the Inheritance can work as well as all other possible parts stay in sync.

In the moment that is done, and nicely weighted, The Fracture Object can be used in a copy with the chunks on the floor as Inheritance based information and as shown in the scene before blend towards the results you are after.

======

Since this is a new problem, Voronoi on a Skin Deformer based object, I have to research this a little bit. When I find something, I let you know.

I have my sketch like “New Bind”, with the problems as link via PM.

The file below as mentioned above.

All the best

Voronoi Deformer Effector examples:
https://www.amazon.com/clouddrive/share/jrAIMo1K6Kh0oLYRFL3d1jReYMgAThvsvMcayyVFSmd?ref_=cd_ph_share_link_copy

P.S.: This is an idea to keep the animation as well the Voronoi active, and have the option to influence the “chunks” of the Voronoi with Effectors.
I’m not certain if that is the most elegant solution, but for today it is all I can share. ;o)

Note that the Mesh Deformer is initiated and any change in the Voronoi requires a new intiation.

Scene sketch:
https://www.amazon.com/clouddrive/share/MIPiVzCmkNMZdMTiwZaCCN9nvf1qN8w105wdw5Wcau2?ref_=cd_ph_share_link_copy

Doc, thanks for the insight about minimizing the scene to just the problem.  I look forward to investigating your files and ideas.  If I succeed (or fail), I’ll let ya know.

Mark

Mark, thanks for the nice feedback.

So far I can see, no one has discussed the problem of an object moving through a Voronoi fracture space, or how it can be prevented to be newly set into different fragments each frame. Sorry that I have no link to a solution for you.

My best wishes!

Much closer, and close to be much more perfect ;o)

Mark, here is another one, I have send you a link to the file via PM. I would love to share it here, but I guess that is not really possible. I have to set up an example tomorrow, then :o)

The idea is simple. The Matrix Object uses the Object Mode and here the joint-animated object itself.
With this, and set to Surface (Matrix), the information is related to the animated body, and not to the space in the scene only. With this,the information (here the Matrix) can be fed into the Voronoi Object.

The Matrix Object needs to be below the Joint/Skin animated object—priority wise.

(Note that the whole Voronoi Effect is based on lines between points. Always half way between points, all points. So, se the matrix “points” away from the surface, which makes it a little bit more stable. Close to be perfect, but points create an effect….)

The “chunks” are not flickering or look like randomly new with each frame. Of course, when the arms move and have a matrix point, the body of the dancer will get affected and the arms from the body based points.
Anyway, it looks way more harmonic and stable than any other option I have explored. Yet, it is not as precise as making the Voronoi editable and animate the polygon chunks via joints/skin. No surprise here.

I have also added a comparison among the three options discussed so far, how a Voronoi works with Character animation, as well if the Matrix is used with the animated object as source, and of course the Voronoi editable and animated. All with joint/skin as animation source.

My best wishes

Screen-capture (still): Object manager set up
https://www.amazon.com/clouddrive/share/0qWYfwrDE6tc3njir9OBhbVIZ6JmxI5gpSFhrfDt0si?ref_=cd_ph_share_link_copy

Comparison of three methods - Scene: (switch the Formula effector on and off to see even better what is going on!)
https://www.amazon.com/clouddrive/share/ynUI5zRpxrwLgzlNJ8Iughc8zUJmHXVfxTX38Ecd8JG?ref_=cd_ph_share_link_copy
Comparison of three methods - Screen image: (the huge difference is only really visible while the animation is played!)
https://www.amazon.com/clouddrive/share/lM17fA3pY604lLND12KFiP6ETOIRQFOIGUWjgysSwG6?ref_=cd_ph_share_link_copy

I’m excited about the progress you made in the “Voronoi_dancer_01” file, Doc, but I’m confused by all of the effectors and how each one works.  I’ll have to do some studying to see if I can gain a better understanding of these mysterious creatures.  I look forward to solving this challenge at some point in time.  There’s got to be a way.  Thanks for all of your help and getting me this far!

Mark

Hi Mark,

You’re welcome. :o)

I think the “comparison” of the three versions shows clearly what and how it works.

From there it is more or less a question of clone manipulation, as shown in some earlier files above.

(BTW: The only stable version would be to place the surface of the character itself into the Sources.)

As a side note: The video clip you had linked used motion and editing to improve the impression, a static camera with no edit is always harder and often less convincing.

All the best

P.S.: to give the “chunks” a little nudge to move on and have a nice independent dynamic after that initial nudge is the key of this file _31.
The second file uses this and reverse the cache so it assembles itself. Pretty much the motion (in a nutshell) that you were after.

The null called “Move-me!” is a way to change direction (before cached of course). You might explore these, as even using only a few parts, there is a lot to “play” with.

ENJOY

Both files in one zip:
https://www.amazon.com/clouddrive/share/pZxfRClvTHUar8NVU4uP0NOtbs5wT9qHUBXtUnQRb9q?ref_=cd_ph_share_link_copy

Preview clip-3sec:
https://www.amazon.com/clouddrive/share/OcC0bA2mqn3vmX5lSyJGKW1wsQWLNqlA6qqp7Ovn1FN?ref_=cd_ph_share_link_copy

Man, do I appreciate your persistence, Doc!  As soon as I can set aside a few moments, I’m super excited to examine all of your files and hints.  I only hope I can comprehend what you’ve put together.  I’m a bit more green with Cinema than you.