Cineversity

Hi Dr. Sassi, good day to you sir. I am attempting to upload to dropbox an intersecting problem. My 3D text is intersecting, and I can not figure out why-can you pleas find for me the problem,?, thank you, Craigintersecting-problemintersecting-problem

PS- I probably messed this up ( copy/paste my file, that is) -can you kindly tell me the tutorial that can clarify, so I do not waste your time?, thank you

Hi Craig,

Please use this link:

Edit: the link is in your Cineversity Private Message. /edit

I will have a look into it.

You never waste my time, BTW, any question not asked is a waste of time ;o) ...as any question left, is like a “black hole” in the future development. But yes, you are right, if there is a tutorial, then that is a good idea, of course!

All the best

Hi Dr.Sassi, the file is up in dropbox , thank you-I believe to be a dynamics issue, which is my problem, Best, CZ

Thank you, Craig, I got your file.

I have explored it, and I’m not clear what you like to have. My best estimation is, the letters should move based on the three effectors [including random and delay/blend], but in case of a collision the Dynamic should take over. If the Dynamic takes over, then only for the short amount of an other wise intersection, as the final position must be taken. Certainly only for the two objects in question at that moment. Is that correct? Please advice!

I have no idea how to set this up, as the collisions should be only local and not globally start to work. Do I got this correctly? You have dynamics or not, would be the simple answer.

If the dynamic would bounce the two letters that have a collision apart, then they would move [perhaps] toward other letters. This triggers then another local solution.

NOT A SUGGESTION, just a discussion: With Dynamic Ghost Tags and XPresso, one could determine a collision. This event could go then into the mix and place an object at that position (invisible). this object is then used to create an inverse falloff for all Effectors. After a few frames it needs to be removed. This needs to be set up as a constant iteration with a time trigger. Sounds like a lot of trial and error.
Compared to use your animation as guide, and bake the animation and use the “Reduced Modification Mode” in the timeline to adjust the collisions.

I’m not certain what to suggest here, other than manual animations, or bake and adjust or use a reverse dynamic. Since you set up is based on a parenting Null for each Letter, the Extrude can be moved. Key frame start and end and then around the event, with enough time around - so it doesn’t look artificial. If a mix of both is possible, please let me know.

Let me know if I got your ideas correctly and what is important to you in this set up, I will look into it.

My best wishes

Thanks Dr. Sassi; at about frame 63, the red"o” and the green capital “D” intersect, so I presumed that I have the wrong settings on my dynamic tags. Your suggestions, although efficient in your response, at this point is beyond my expertise. Thank you for time, as I will explore other means. I am just trying to create a short, 5 second “splash page”, so there are many ways and tutorials out there. thank you so much for your time, Craig

I was just looking at a similar frame, when your message came in, Craig, scrub back and see how the delay works on it. This is not a simple scene, and five seconds or not makes no difference.

Well, I have not made it clear obviously. You can’t have both fully, if you like to have only the best from both techniques without setting up a larger analysis what is now the one that should work.

If you like to use Dynamics, then you will have the effect of dynamics in the scene: This will then not lead to the final set up. Final as in: all letters are lined up. Yes, there is an option to switch off and let a key framed animation take over. But then it might intersect again. It would work as baked and reversed, but not if the effectors have 100% control.

I added in the previous post an option. Since you have all letters on an individual Null, you could set up start and end keyframes for that, and then before and after the collision with enough time. Then move both a little bit apart at the peak of the collision and set new key frames. The before and after determines how much it looks in flow. If too far away, then new collisions might be introduced.

The Delay set to Blend will overwrite perhaps.

I would suggest a pure Dynamic solution with a reverse cache.

Craig, if it is 5 seconds or 5 hours, if it is simple or complex, it doesn’t matter if you mix two options and expect the best out of both, and magically excluding the side effects of both.
BTW, It was not my suggestion, just a discussion to put light on the complexity you ask for.
I think the adjust the extrude as described above seems to be the most simple and efficient solution.

With effectors directing the scene and dynamics have only to show up when a collision is there, but then to shut off again and keep things running. This is certainly not simple.

Let me know what direction you would like to go.

All the best

I have uploaded a little clip, Craig, please have a look here:

https://www.amazon.com/clouddrive/share/wMQTwlcs2s4wxJcQR5E5SkzrUwx2ifpeFna2t1IYGMt?ref_=cd_ph_share_link_copy

The files (one zip) contain the scene as in the clip [01], and another one [02] with more adjustments, as the Delay/Blend will change everything of course.

I have deleted all thing not needed. So only the two letters are left and placeholders, as the position under the Fracture counts, if one deletes all in that set up, the animation will differ!

I have set to keyframes [postion/rotation] before and after the problem zone. then where the collision is, I have adjusted the Extrude, as the Null is in question here for the animation. In the clip I moved that adjusted keyframe out, to show the collision, then I moved it in to show its effect. After that I showcase how the tow “neutral” [zero] keyframes will influence the animation by moving them closer or away from the adjusted keyframe.

All the best

Both files are in this zip—link:

https://www.amazon.com/clouddrive/share/Mf3ZLpZmnK4vxI8TfjfNkf76wNXHo68QCXePJJXWfeQ?ref_=cd_ph_share_link_copy

P.S.: A solution with no manual adjustments would be thinkable with Dynamics and Springs, whereby a blend of a fixed manual animation and the blend needs to be established. This would change the character of the animation, but allows to have collision and final positioning.

So, if you ask for a tutorial: Well, Non Linear Animation for Teams contains all the bake and mix options you need for that. It’s a course.
http://www.cineversity.com/vidplaytut/animation_techniques_for_teams_part_01_course

Create one dynamically and one with the Effectors. Then mix both as discussed in this course. Not as simple as you requested, hence why I hesitated to suggest it.

With this method one can easily do a reverse set up as well. Starting with the end point and let the letters drop, or blown away from a “wind instance” to use dynamics.

Let me know if there is anything else. I really like to support your ideas, but I can’t change the way things work.

Attached here is the version of the reverse Dynamics, mentioned above. But then full dynamics.

My best wishes.

Hi Craig,
As usual I do not give up until we find something, if there is a way at all. The following might not work for you. Please advice. I have made a little one minute clip. It is based on the file above [_11]
https://www.amazon.com/clouddrive/share/dxmFWFIO6cSJHMMuhDwvWN4koPAocdpRHCYR8O06luy?ref_=cd_ph_share_link_copy
Perhaps download the file for a better viewing experience.

I have tried to come closer to your set up, but again, if the Effectors need to control 100% then there is no place for Dynamics. To enable a mix, if that is possible:
The Follow Position and Follow Rotation must be adjusted (Dynamic Tag).

I have attached three files [_11 to __13].
https://www.amazon.com/clouddrive/share/5fZ1kwJ38qo2Yoq4m8Lh6L9QVKT4esebvADGtDCa1BM?ref_=cd_ph_share_link_copy
As you can see in the [_13] file, I have set the Follow Pos/Rot to 100 in each: the scale from the Plain would normally “explode” the scene, but it follows the movement, and not at all anymore the Dynamic aspects of the scene as they are 0%.

I hope this exploration made it more clear, what I tried to explain above.

All the best