22-11-2022, 08:34 AM
(This post was last modified: 22-11-2022, 08:35 AM by josemendez.)
(22-11-2022, 12:46 AM)Milionario Wrote: Increasing an ObiCollider thickness value from 0 to 0.1f for example, gives better results, but the visual part has to be remediated by increasing the extruded renderer thickness scale
Setting it to a negative value like -0.2f will have the effect of not even colliding at low speeds
Negative values make the collider thinner (positive values make it thicker, and 0 leaves it its its default thickness) so using a negative values will of course make the problem worse.
(22-11-2022, 12:46 AM)Milionario Wrote: https://gyazo.com/142551ed697eea873cbd61a5b1af2ec6
this gif shows my ragdoll hitting a really thin cube (simulating a rope shape), it doesnt go through
This is a static cube, so it doesn't apply any forces to the ragdoll: it just kinematically projects any penetrating colliders outside of it. It doesn't relate in any way to a rope. I would expect the ragdoll to collide correctly against any static geometry.
(22-11-2022, 12:46 AM)Milionario Wrote: Just tried with a falling cube and the rope never lets it go through and it bounces accordingly.. So its something with the character, but what could i try to pinpoint the problem?
Yes, I suspect the problem lies in the way the active ragdoll is implemented, and how it controls the rigidbodies. From your first video, it's clear that the ragdoll is largely ignoring the forces applied by the rope: it's as if only a small percentage of the force imparted by the rope is actually affecting the rigidbodies. Obviously, the larger the ragdol collider and/or the particles, the more time they spend overlapping each other so these small force is applied during more frames, which amounts to a larger net force, hence the feeling that it works better when you make colliders thicker.
If possible, could you share a small scene/project that reproduces this issue? I think that's the fastest way for us to sort this problem out.
kind regards,