← All ideas
  • Published 5 May 2020

Dynamic pinning/goal for cloth and soft body simulations

It would be nice to be able to animate a pickup and drop of cloth. Grabbing a curtain and drag it to the side and then release for instance. I tried to achieve this with vertex wight proximity modifier but the cloth simulation doesn't seem to be able to deal with underlying changes to pin groups.

Skärmavbild 2020-05-05 kl. 23.01.49.png

96.4%

Development Status: Open

Upvote 27 Downvote 1
emil.assarsson
5 May 2020

9 Comments

Vyach

There is free addon «Modelling cloth». Also you can make hooks, forces and collisions.

emil.assarsson Author

* @ Vyacheslav * Can you elaborate a bit more about how that would solve this problem?

Vyach

* @Emil Assarsson * Dynamic parenting hook to hand. Dynamic vertex weight mix from 0 to 1 in the pin group. Thats all. Basics. {attachment 2020-05-10_16-32-46mp4} Do not create proposals before you have learn basics.

1
emil.assarsson Author

* @ Vyacheslav * Thank you. Do you also have a way to do this for soft body simulations? I did not know that it was a prerequisite to know it all, do you have a link to that requirement?

Vyach

* @Emil Assarsson * As far as softbody goal support group weight, it is possible, but Goal is not pinning, it is pretty soft. So you should use full body goal, and probably it will ruin soft deformation. But with high stiffness and damping you can drag softbody. If it is ball, probably the best is to use cloth with pressure.

1
bassem.adel

I guess it is somehow there when playing animation but if you drag a collision object but not the cloth itself , we may say we need something like marvelous designer , it has this feature and it works like charm , but why not having this in blender :-D

lsscpp

Good point. Did you try with obstacle object with animated friction?

emil.assarsson Author

* @ Lsscpp * No, I have not tried that but I doubt it will be acting nice if you do a good pull on the curtain. There might be occasions when those type of solutions may work but my point is that we need to have a solid and predicable way to do this. Thank you for you suggestion :-)

1
emil.assarsson Author

* @ Lsscpp * Tried it. Might be useful in some cases. The collision needs to be big enough. Not entirely reliable.