Help Cloth has stretchy behaviour with stiff settings

[Andreia Mendes]

The scale inside Unity reflects the scale you exported the asset with. It's the same value. However, blueprints take a mesh as input, without any scene transform information. So the actual cloth in the scene is smaller than your mesh, and that's what's causing it to shrink.

Export and import settings may affect scale however. Most FBX exporters have options to deal with transform scale, and Unity mesh import settings also allow you to apply an extra scale after import.

Hello,

I have imported the swimsuit at a scale of 1, and now everything seems to be working! Can't believe I missed something so simple...
As a side note, i found where i got that wrong value of 0.01 for the stretch compliance, it was from a previous post that appears when you search "Cloth Jitter"! I'm guessing that one is outdated...

   

it was a bit stretchier than i liked, so I changed stretching scale to 0.8, since it's a swimsuit it makes sense.
Unfortunently a previous problem returned, the cloth edges seem to twist and turn and generally jitter during simulation.


  Screenshot 2025-07-24 132938.png (Size: 46.54 KB / Downloads: 5)

Here is an image of the simulation without the avatar visible:

   

And my current cloth settings:

   

[josemendez]

Hello,

I have imported the swimsuit at a scale of 1, and now everything seems to be working! Can't believe I missed something so simple...
As a side note, i found where i got that wrong value of 0.01 for the stretch compliance, it was from a previous post that appears when you search "Cloth Jitter"! I'm guessing that one is outdated...

Hi!

That post is not outdated and the advice given there still holds, however the post author's case is the exact opposite of yours: he was using very low-poly cloth that was behaving too rigid: increasing compliance is a good way to make it less rigid, which in his case reduce jittering because the cloth was constrained to the character's skeleton using skin constraints.

However in your case cloth is quite high poly, and your problem is that the cloth is too elastic. So you want the compliance to be very low - ideally zero.

it was a bit stretchier than i liked, so I changed stretching scale to 0.8, since it's a swimsuit it makes sense.
Unfortunently a previous problem returned, the cloth edges seem to twist and turn and generally jitter during simulation.

The problem seems to be analogous to the stretching one: bend compliance is also 0.01, so the cloth is free to bend/twist quite a lot. Set it to zero.

kind regards,

[Andreia Mendes]

I see, good to know low and high poly have different behaviours.

As for the bend compliance, when i set it to 0 the behaviour doesn't seem to change.

[josemendez]

I see, good to know low and high poly have different behaviours.

The more triangles your cloth has, the more prone to stretching it will be. The reason is simple: if each triangle is able to stretch 1cm, if you have 10 triangles you end up with less overall stretching than if you have 100.

Same logic applies to any physics engine. For instance if you have a pile of boxes in a rigidbody engine, the taller the pile the more it will compress.

As for the bend compliance, when i set it to 0 the behaviour doesn't seem to change.

What kind of collider are you using for the character's body? also, does it have a collision material assigned? if so, what are the properties of the collision material?

[Andreia Mendes]

The avatar is made of several Obi colliders of different body parts, each attached to each bone to "allow" for animation. Each Obi collider is using a distance field.

   
(The mesh colliders are only used for visualization purposes, not for the Obi colliders)

The Obi collider looks like this:
   

and each Obi collider does have the same collision material attached to it:
   

[josemendez]

Collider setup seems fine. Would it be possible for you to share a scene/project that exhibits this issue? (if so, send it to support(at)virtualmethodstudio.com) I can't really think of anything else that would cause only mesh edges to twist/jitter.

A side note: having higher dynamic friction than static friction doesn't generally make much sense in physical terms. Static friction should usually be higher for it to have any effect.

[Andreia Mendes]

Collider setup seems fine. Would it be possible for you to share a scene/project that exhibits this issue? (if so, send it to support(at)virtualmethodstudio.com) I can't really think of anything else that would cause only mesh edges to twist/jitter.

A side note: having higher dynamic friction than static friction doesn't generally make much sense in physical terms. Static friction should usually be higher for it to have any effect.

Can do, I will make an empty test project and send it over.

I haven't really looked at the collision material in a while, but yes, that does make sense... Something else for me to ajust, thanks for pointing that out!