Latest Threads |
How to dynamically change...
Forum: Obi Rope
Last Post: quent_1982
04-07-2025, 11:29 AM
» Replies: 2
» Views: 92
|
Pipeline that bends
Forum: Obi Softbody
Last Post: josemendez
04-07-2025, 09:52 AM
» Replies: 13
» Views: 794
|
How to implement/sync Obi...
Forum: Obi Rope
Last Post: quent_1982
01-07-2025, 01:48 PM
» Replies: 2
» Views: 182
|
Collisions don't work con...
Forum: Obi Rope
Last Post: chenji
27-06-2025, 03:05 AM
» Replies: 3
» Views: 253
|
Force Zone apply differen...
Forum: Obi Rope
Last Post: chenji
26-06-2025, 11:41 AM
» Replies: 11
» Views: 734
|
Can I blend in and out of...
Forum: Obi Cloth
Last Post: josemendez
24-06-2025, 04:42 PM
» Replies: 3
» Views: 240
|
Using a rigidbody/collide...
Forum: Obi Cloth
Last Post: josemendez
24-06-2025, 09:29 AM
» Replies: 1
» Views: 144
|
Solver is too performance...
Forum: Obi Rope
Last Post: quent_1982
20-06-2025, 08:09 AM
» Replies: 40
» Views: 4,214
|
Obi 7 Model Scaling
Forum: Obi Cloth
Last Post: alkis
19-06-2025, 02:37 PM
» Replies: 2
» Views: 254
|
Obi Softbody instability?
Forum: Obi Softbody
Last Post: Aroosh
18-06-2025, 06:35 PM
» Replies: 0
» Views: 136
|
|
|
New User with a Question |
Posted by: Fearinhell - 25-10-2021, 04:20 PM - Forum: Obi Cloth
- Replies (6)
|
 |
Ok Im trying to develop a game and use obi cloth, I want to change avatar size during game play but this seems to be giving obi cloth fits. Lets use underwear as an example when I adjust character size from 1.0 to .75 the underwear drops to thigh level and starts going through the character. I have adjusted the blueprint of skinned cloth to work at size 1.0 but it doesnt seem to adjust after that. Ive tried toggling the object off and on and for a split second the underwear is in place with the new size but then goes all wonky again. Thanks in advanced.
|
|
|
Obi softbody crashes after setting inverse masses |
Posted by: Jschol - 25-10-2021, 10:42 AM - Forum: Obi Softbody
- Replies (2)
|
 |
Hi all,
I am trying to set the inverse masses of my obi softbody however, it crashes when settting it to a non 0 value with the error:
transform.position assign attempt for 'Cluster0' is not valid. Input position is { NaN, NaN, NaN }.
UnityEngine.Transform et_position (UnityEngine.Vector3)
Obi.ObiSoftbodySkinner:UpdateBones (Obi.ObiActor) (at Assets/Obi/Scripts/Softbody/Rendering/ObiSoftbodySkinner.cs:130)
Obi.ObiActor:Interpolate () (at Assets/Obi/Scripts/Common/Actors/ObiActor.cs:1207)
Obi.ObiSoftbody:Interpolate () (at Assets/Obi/Scripts/Softbody/Actors/ObiSoftbody.cs:289)
I am setting the inverse masses like this:
Code: public void setGroupInvMasses()
{
foreach (int index in obiParticleGroup.particleIndices)
obiActor.solver.invMasses[obiActor.solverIndices[index]] = inverseMasses;
}
I also call:
Code: ObiSoftbody.UpdateParticleProperties()
after setting the inverse masses.
Is there something I miss?
I am not sure if it is relevant but, I also change some of the shape matching constraints of the body.
Kind regards,
Jasper
|
|
|
Sync cloth proxy source cloth over network |
Posted by: KyleJohnOConnor - 25-10-2021, 09:47 AM - Forum: Obi Cloth
- Replies (2)
|
 |
Good day.
I have a high poly Obi cloth proxy that is a dense wire mesh (at about 12000 vertices) with a source cloth that is a very simple plane (at around 110 vertices).
The goal of this is to manipulate the source cloth using Obi particle attachment to place the wire mesh over a section of a wall. There's a good deal of cloth interaction with Obi colliders as well.
What I want to do is run the cloth simulation on a single central PC (called it the primary node) and have the obi cloth proxy behave exactly the same way on three other PCs (called them the secondary nodes). These secondary nodes are used for rendering different views(left, front and right); so it's very important that the cloth's positions and behavior is consistent between the secondary nodes (for an extreme example some vertices might have sunk through the wall on the left view rendering secondary node, but are fine on the front view render secondary node).
My first plan was to try use a byte[] to sync the vertex positions of the high poly mesh on the primary node to the instances of the high poly mesh on the secondary nodes. Directly applying the Vector3[] of the mesh vertices from one mesh to another was easy for Unity. Unfortunately it is very costly to convert that same Vector3[] into a byte[] to transfer over a network, and then converting back into the Vector3[] to apply to the high poly mesh on the secondary nodes.
So I was hoping there was a way to apply changes in vertex positions on the source cloth; unfortunately this doesn't seem possible while it's being used as part of the obi cloth system. This might work by removing all the Obi cloth system from the secondary nodes, but then the proxy would no longer work (because it needs the obi solver to function).
TLDR: I want to serialize a cloth from one PC to another so that I know for a certainty that it's visually the same.
Is there a simpler solution that I've missed perhaps?
Any ideas or help would be greatly appreciated!
|
|
|
Extracting mesh from settled fluid |
Posted by: CreepGin - 25-10-2021, 04:50 AM - Forum: Obi Fluid
- Replies (6)
|
 |
My goal is to extract an iso-surface from a resulting fluid simulation. Does Obi Fluid provide any kind of mesh extraction out-of-the-box?
If not, what are the options for me going forward? Do I need to build a SDF of the ellipsoids from scratch?
Thanks for the help!
|
|
|
Optimizing Performance With Multiple Softbodies |
Posted by: relevantname - 22-10-2021, 07:44 PM - Forum: Obi Softbody
- Replies (2)
|
 |
Hi,
I want to have min. 60-70 softbodies in my scene(Little softbody cubes), bu performance issues arise when there are more than 20 softbodies in the scene(FPS dropping to 2-4). Each softbody has 8 particles and ObiSolver's substep amount is 2.
I tried to solve this problem with these solutions;
- I created an Obi Solver for each of my objects. And I unchecked the box 'Simulate When Invisible', but no luck, fps still drops.
- I attached a script that checks the distance between player and softbody to enable(when the player is in range) or disable(when the player outside of the range) the ObiSoftbody component.
This approach worked very well but most of the times softbodies didn't behave correctly to player's actions or there are lots of position mistakes(weird teleportations) happened.
- I attached a script that checks if the softbody is visible to camera or not, and enable or disable the ObiSoftbody component. (Not surprisingly, I got the same result as solution above)
Are there any solid solution for this kind of case, or am I doing something wrong with applying these solutions?
P.S: In solutions 2 and 3, I also tried to AddActor and RemoveActor methods, not only enabling or disabling the ObiSoftbody component.
P.S 2: I'm using Burst mode. (I also tried to switch Obi mode)
P.S 3: I'm not checking collisions on particles.
Thank you.
|
|
|
Particle size different between faucet example scene and my scene |
Posted by: cdauphin - 22-10-2021, 10:42 AM - Forum: Obi Fluid
- Replies (2)
|
 |
Hello !
As the title says, I am facing an issue where the fluid particle size is different between the faucet example scene and my scene.
In my scene, particle are way bigger (like ten times bigger). Solver, emitter and particle renderer components have the exact same value between the two scenes. Emitters uses the same blueprint.
Does someone knows what could cause this ?
Thanks.
|
|
|
Performance again |
Posted by: lacasrac - 22-10-2021, 10:34 AM - Forum: Obi Rope
- Replies (1)
|
 |
In debug mode I get a massive 17 FPS in fullscreen when I use 5 solvers/1 rope/chain with 8 subsets
So basically I have 5 solvers with 3 chains and 2 ropes
Is it normal?
Jobs/Leak detection is off
Burst-> Enable compilation on, Synchronous compilation on, others off
Jobsdebugger off
Any other thing that I can do to increase the FPS?
One more question:
If I have 5 solvers with 1 ropes -> basically 5 ropes with 5 solvers invidiudally that is good or bad for performance?
It will be better 1 solvers with 5 ropes?
|
|
|
|