<![CDATA[Obi Official Forum

<![CDATA[Obi Official Forum - General]]> https://obi.virtualmethodstudio.com/forum/ Wed, 11 Nov 2026 04:36:31 +0000 MyBB <![CDATA[VR Meta Quest Lag issue]]> https://obi.virtualmethodstudio.com/forum/thread-4659.html Wed, 30 Sep 2026 16:29:11 +0200 kodkuce]]> https://obi.virtualmethodstudio.com/forum/thread-4659.html So when something hits USB connector it moves nice with cable, but my issue now is when i move the cable with controller/hands it lags behind, i tried changing
to Synchronous Fixed mode and setting interpolation to none but it did not help.

i found this old thread where it says to use LateFixedUpdate, that sounds similar to this S.Fixed but like i say it does not fix it for me. Should i try to downgrade
to version 6 so i can try LateFixedUpdate? Any other solution? I wasted full day trying to fix this]]> So when something hits USB connector it moves nice with cable, but my issue now is when i move the cable with controller/hands it lags behind, i tried changing
to Synchronous Fixed mode and setting interpolation to none but it did not help.

i found this old thread where it says to use LateFixedUpdate, that sounds similar to this S.Fixed but like i say it does not fix it for me. Should i try to downgrade
to version 6 so i can try LateFixedUpdate? Any other solution? I wasted full day trying to fix this]]> <![CDATA[designing a tubular structure with compression]]> https://obi.virtualmethodstudio.com/forum/thread-4645.html Tue, 01 Sep 2026 07:57:49 +0200 godwinferin]]> https://obi.virtualmethodstudio.com/forum/thread-4645.html <![CDATA[BurstColliderWorld makes scene dirty all the time]]> https://obi.virtualmethodstudio.com/forum/thread-4637.html Mon, 27 Jul 2026 15:52:13 +0200 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4637.html This is quite annoying when commiting changes in git, so it would be great if that did not happen. Maybe hide flags could help?]]> This is quite annoying when commiting changes in git, so it would be great if that did not happen. Maybe hide flags could help?]]> <![CDATA[GetInstanceID obsolete, can't update to 6.5]]> https://obi.virtualmethodstudio.com/forum/thread-4634.html Fri, 17 Jul 2026 20:23:02 +0200 Jawsarn]]> https://obi.virtualmethodstudio.com/forum/thread-4634.html <![CDATA[Triggers and raycasting SDF]]> https://obi.virtualmethodstudio.com/forum/thread-4620.html Fri, 08 May 2026 17:20:38 +0200 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4620.html I can see in docs that there is example with custom gravity force zone, so I guess I just need to enable isTrigger in unity collider, however I guess this is not optimized in anyway and "contacts" are still generated.

1. Is that the best way to detect object inside some custom zone (trigger collider with SDF)?
2. Is there other way / option to optimize those without much effort?
3. Is it possible to raycast obi SDF? There is option to query particles, but it would be nice to raycast colliders without need of mesh collider.
4. In additon I wonder if there are performance benefits of reusing distance field (besides memory). I assume it needs to find octant and that is main thing to do, assuming bounding boxes overlap and collision layers match. So I guess if there are tow SDF's with the same or slightly different set of nodes there is no much difference.]]> I can see in docs that there is example with custom gravity force zone, so I guess I just need to enable isTrigger in unity collider, however I guess this is not optimized in anyway and "contacts" are still generated.

1. Is that the best way to detect object inside some custom zone (trigger collider with SDF)?
2. Is there other way / option to optimize those without much effort?
3. Is it possible to raycast obi SDF? There is option to query particles, but it would be nice to raycast colliders without need of mesh collider.
4. In additon I wonder if there are performance benefits of reusing distance field (besides memory). I assume it needs to find octant and that is main thing to do, assuming bounding boxes overlap and collision layers match. So I guess if there are tow SDF's with the same or slightly different set of nodes there is no much difference.]]> <![CDATA[ObiFluidRendererFeature compile error]]> https://obi.virtualmethodstudio.com/forum/thread-4618.html Sat, 02 May 2026 17:41:25 +0200 ruitar]]> https://obi.virtualmethodstudio.com/forum/thread-4618.html
Using Unity 6.4 (6000.4.2f1) and I get the following error:

Assets/Obi/Scripts/Fluid/Rendering/URP/ObiFluidRendererFeature.cs(247,30): error CS0115: 'VolumePass.Execute(ScriptableRenderContext, ref RenderingData)': no suitable method found to override

I followed instructions in https://obi.virtualmethodstudio.com/manu...setup.html and have all the dependencies.

The setup seems to work with older versions but I get the error specifically for 6.4.

Would appreciate any help! Thanks.]]>
Using Unity 6.4 (6000.4.2f1) and I get the following error:

Assets/Obi/Scripts/Fluid/Rendering/URP/ObiFluidRendererFeature.cs(247,30): error CS0115: 'VolumePass.Execute(ScriptableRenderContext, ref RenderingData)': no suitable method found to override

I followed instructions in https://obi.virtualmethodstudio.com/manu...setup.html and have all the dependencies.

The setup seems to work with older versions but I get the error specifically for 6.4.

Would appreciate any help! Thanks.]]> <![CDATA[ECS Physics/Graphics Support]]> https://obi.virtualmethodstudio.com/forum/thread-4615.html Thu, 23 Apr 2026 22:29:29 +0200 Jawsarn]]> https://obi.virtualmethodstudio.com/forum/thread-4615.html
Wondering if there's any plans to support ECS Physics and/or ECS Graphics?
Would be great not to have to add a GO per entity for those we want to add Oi Colliders to, and also if we can keep the rendering on ECS side.]]>
Wondering if there's any plans to support ECS Physics and/or ECS Graphics?
Would be great not to have to add a GO per entity for those we want to add Oi Colliders to, and also if we can keep the rendering on ECS side.]]> <![CDATA[Any upcoming Softbody sales planned?]]> https://obi.virtualmethodstudio.com/forum/thread-4614.html Wed, 22 Apr 2026 16:19:30 +0200 Youxi]]> https://obi.virtualmethodstudio.com/forum/thread-4614.html ]]> ]]> <![CDATA[Collider World Bug]]> https://obi.virtualmethodstudio.com/forum/thread-4608.html Thu, 09 Apr 2026 13:45:42 +0200 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4608.html
Observation - when new prefab is instantiated into scene (and old one is removed), then for some reason ObiColliders are not registered properly and in result there is no collision with them, unless reenabled.

Investigation - it looks like collider world is destroyed (and also solver implementation and everything else) when there are no colliders to manage. However it does not take into account colliders that should be added during next update. I think that Collider becomes enabled so it registers itself in collision world, but because other collider become destroyed/disabled, then it runs various checks, including DestroyIfUnused() and because number of colliders drops to zero it gets destroyed, but it does not care that there are colliders waiting to be registered! Also actor needs to be reset and is disabled for split of second so Teardown() is called and whole simulation dropped for no reason.

Is adding additional check 'collidersToAdd.Count == 0' in the if statement below valid way to fix this problem?

Code:
private void DestroyIfUnused()

{

    // when there is no data and no implementations, the world gets destroyed.

    // don't check materialHandles.Count == 0, as these are scriptable objects and may outlive the world.

    if (colliderHandles.Count == 0 &&

        rigidbodyHandles.Count == 0 &&

        forceZoneHandles.Count == 0 &&

        implementations.Count == 0)

        Destroy();

}

]]>
Observation - when new prefab is instantiated into scene (and old one is removed), then for some reason ObiColliders are not registered properly and in result there is no collision with them, unless reenabled.

Investigation - it looks like collider world is destroyed (and also solver implementation and everything else) when there are no colliders to manage. However it does not take into account colliders that should be added during next update. I think that Collider becomes enabled so it registers itself in collision world, but because other collider become destroyed/disabled, then it runs various checks, including DestroyIfUnused() and because number of colliders drops to zero it gets destroyed, but it does not care that there are colliders waiting to be registered! Also actor needs to be reset and is disabled for split of second so Teardown() is called and whole simulation dropped for no reason.

Is adding additional check 'collidersToAdd.Count == 0' in the if statement below valid way to fix this problem?

Code:
private void DestroyIfUnused()

{

    // when there is no data and no implementations, the world gets destroyed.

    // don't check materialHandles.Count == 0, as these are scriptable objects and may outlive the world.

    if (colliderHandles.Count == 0 &&

        rigidbodyHandles.Count == 0 &&

        forceZoneHandles.Count == 0 &&

        implementations.Count == 0)

        Destroy();

}

]]> <![CDATA[Inquiry About Future Updates And About Tuanjie Asset Store]]> https://obi.virtualmethodstudio.com/forum/thread-4601.html Thu, 19 Mar 2026 17:40:51 +0100 tecnnt]]> https://obi.virtualmethodstudio.com/forum/thread-4601.html
I recently learned from Unity's official announcement that, by the end of this month, all overseas resources will no longer be accessible to developers in mainland China.

I truly enjoy using your assets and have greatly benefited from them in my projects. However, I’m concerned about how I can continue to receive updates or access your resources after this change takes effect. Could you kindly advise on how to proceed? For instance, will there be alternative platforms or methods to obtain updates for your assets moving forward?

Thank you for your time and support. I really appreciate your work and hope to continue using your resources in the future.

Looking forward to your reply.]]>
I recently learned from Unity's official announcement that, by the end of this month, all overseas resources will no longer be accessible to developers in mainland China.

I truly enjoy using your assets and have greatly benefited from them in my projects. However, I’m concerned about how I can continue to receive updates or access your resources after this change takes effect. Could you kindly advise on how to proceed? For instance, will there be alternative platforms or methods to obtain updates for your assets moving forward?

Thank you for your time and support. I really appreciate your work and hope to continue using your resources in the future.

Looking forward to your reply.]]> <![CDATA[Non destructive blueprint regeneration]]> https://obi.virtualmethodstudio.com/forum/thread-4600.html Fri, 06 Mar 2026 10:03:13 +0100 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4600.html this thread mentioned destructive nature of blueprint generation, I noticed too that iteration based on mesh changes is really slow because of this, adding one vertex to softbody or cloth requires regeneration.

My proposition is to add feature to map existing data to newly generated particles.
To be precise, there would be checkbox next to "Generate" button (or button dropdown similar to lightmap bake) and this mode would keep old blueprint data cached, then after generation it mapsnew particles to old one by finding the closest particle. So things like weight, radius, groups would be copied from previous setup. Also groups - it would keep groups, but replace indices with new one that fit previous positions. In this case adding single vertex to mesh would require minimal changes, if any. In case of meshes with similar surface (but different density of particles) the same would apply, because 99% of data would fit previous setup.

This is not critical feature, but I think it would sagnificantly help many users and make iterations way shorter.]]> this thread mentioned destructive nature of blueprint generation, I noticed too that iteration based on mesh changes is really slow because of this, adding one vertex to softbody or cloth requires regeneration.

My proposition is to add feature to map existing data to newly generated particles.
To be precise, there would be checkbox next to "Generate" button (or button dropdown similar to lightmap bake) and this mode would keep old blueprint data cached, then after generation it mapsnew particles to old one by finding the closest particle. So things like weight, radius, groups would be copied from previous setup. Also groups - it would keep groups, but replace indices with new one that fit previous positions. In this case adding single vertex to mesh would require minimal changes, if any. In case of meshes with similar surface (but different density of particles) the same would apply, because 99% of data would fit previous setup.

This is not critical feature, but I think it would sagnificantly help many users and make iterations way shorter.]]> <![CDATA[Burst job schedule overhead]]> https://obi.virtualmethodstudio.com/forum/thread-4595.html Fri, 20 Feb 2026 16:00:08 +0100 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4595.html First of all I am aware that pushing many large job structs is slow and waking up threads is not free, however I did not expect such a big performance difference.
You can see profiler snapshot of obi simulation scheduling, single frame has 20 substeps and for testing purposes I added callback with job handle to schedule custom job (small blue bottom column).

Observations:
1. Each following substep is longer. For example the first call takes ~0.01ms, while the last one is ~0.1ms. This is huge difference, especially for jobs like particle collision where most of update is taken by them (0.4ms single substep!).
2. Using ScheduleByRef helps a bit, but I think for my custom jobs it's maybe 20% and what is more important I can't see wh huge job structs like friction can take small fraction of frame, while scheduling my job takes several times more time even if it's smaller.
3. I expected that changing Parallel to IJobFor would give me huge boost because it does not need to wake up everything, but it's not true.
4. Calling JobHandle.ScheduleBatchedJobs(); does not change anything, or I can't see visible difference.

.png

profiler.png (Size: 22.86 KB / Downloads: 4)

In short I fail to see what exactly is causing that and on top of that Iam worried about obi8 - if I write custom constraint I guess it will be the same, so writing like 2-3 custom scripts going to kill the frame rate.]]> First of all I am aware that pushing many large job structs is slow and waking up threads is not free, however I did not expect such a big performance difference.
You can see profiler snapshot of obi simulation scheduling, single frame has 20 substeps and for testing purposes I added callback with job handle to schedule custom job (small blue bottom column).

Observations:
1. Each following substep is longer. For example the first call takes ~0.01ms, while the last one is ~0.1ms. This is huge difference, especially for jobs like particle collision where most of update is taken by them (0.4ms single substep!).
2. Using ScheduleByRef helps a bit, but I think for my custom jobs it's maybe 20% and what is more important I can't see wh huge job structs like friction can take small fraction of frame, while scheduling my job takes several times more time even if it's smaller.
3. I expected that changing Parallel to IJobFor would give me huge boost because it does not need to wake up everything, but it's not true.
4. Calling JobHandle.ScheduleBatchedJobs(); does not change anything, or I can't see visible difference.

.png

profiler.png (Size: 22.86 KB / Downloads: 4)

In short I fail to see what exactly is causing that and on top of that Iam worried about obi8 - if I write custom constraint I guess it will be the same, so writing like 2-3 custom scripts going to kill the frame rate.]]> <![CDATA[Managing dynamic constraints]]> https://obi.virtualmethodstudio.com/forum/thread-4582.html Tue, 23 Dec 2025 10:40:40 +0100 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4582.html As far as I know constraints are stored in blueprints or directly in actor (pin).
Then those constraints are pushed to solver and copy is created and merged in solver.

There is mechanism to make constraints dirty, this causes constraints tobe rebuilt, by taking all actors constraints, however what about stitch constraints?
I can't see them stored in actors or blueprints, so is there only runtime representation? What if I marked stitch constraints as dirty, would that mean that all stitch constraints are cleared and cannot be rebuilt, because there is no blueprint/actor (original) copy?

Also what if I wanted to create some pin constraints during runtime via custom script and modify offsets every step? I want to avoid making constraints dirty every frame, but at the same time it does make sense to modify actor copy in case it's rebuilt, so should I modify both? How can I diffrentiate my own batch from all other solver batches to avoid dirtying?]]> As far as I know constraints are stored in blueprints or directly in actor (pin).
Then those constraints are pushed to solver and copy is created and merged in solver.

There is mechanism to make constraints dirty, this causes constraints tobe rebuilt, by taking all actors constraints, however what about stitch constraints?
I can't see them stored in actors or blueprints, so is there only runtime representation? What if I marked stitch constraints as dirty, would that mean that all stitch constraints are cleared and cannot be rebuilt, because there is no blueprint/actor (original) copy?

Also what if I wanted to create some pin constraints during runtime via custom script and modify offsets every step? I want to avoid making constraints dirty every frame, but at the same time it does make sense to modify actor copy in case it's rebuilt, so should I modify both? How can I diffrentiate my own batch from all other solver batches to avoid dirtying?]]> <![CDATA[Distance field default offset]]> https://obi.virtualmethodstudio.com/forum/thread-4570.html Thu, 13 Nov 2025 15:59:39 +0100 Qriva0]]> https://obi.virtualmethodstudio.com/forum/thread-4570.html <![CDATA[Mixing 2D and 3D Physics results in Obi MissingReferenceException:]]> https://obi.virtualmethodstudio.com/forum/thread-4559.html Fri, 17 Oct 2025 16:44:22 +0200 this_guy]]> https://obi.virtualmethodstudio.com/forum/thread-4559.html The Problem

After upgrading to Obi Cloth 7.1, my project suddenly produced the error

"MissingReferenceException: The object of type 'ObiRigidbody2D' has been destroyed but you are still trying to access it. Your script should either check if it is null or you should not destroy the object."

This error was happening everywhere and on every frame!

The Solution

My project mixes 2D and 3D physics on purpose. The cause of the error was this snippet of gameobject hierarchy

> Rigidbody 2D
> Box Collider 2D
...
> Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

Notice the 2D Unity Rigidbody mixed with the 3D Unity Box Collider and 3D Obi Collider.

The fix is to insert a 3D Unity Rigidbody below the 2D Unity Rigidbody like

> Rigidbody 2D
> Box Collider 2D
...
> Rigidbody (i.e. Unity's 3D Rigidbody component)
Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

I also marked the (3D) Rigidbody as "Is Kinematic" so that the 2D Box Collider and the 3D Box Collider would behave as if they are part of the same Rigidbody (which is the behavior I'm looking for).

Explanation

In the editor, Obi thought that everything was fine with the original hierarchy. ObiCollider was happy because there was a 3D Unity Collider on the same gameobject that the ObiCollider was attached to. And at runtime, just as the online documentation says, Obi "look[ed] for the first Rigidbody component up its hierarchy and attach[ed] an ObiRigidbody component to it." Since the first Rigidbody component Obi encountered was the Rigidbody2D, Obi added an ObiRigidbody2D component to the hierarchy like this

> Rigidbody 2D
Obi Rigidbody 2D

> Box Collider2D
...
> Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

But then Obi threw the above MissingReferenceException.

I traced the Obi code a bit. The error is thrown from ObiRigidbody2D.CacheVelocities() where it turns out that this is null!! I didn't know that was possible Sonrisa

. I think it has something to do with parallelization. One thread deleted the ObiRigidbody2D (I'm not exactly sure where in the code the deletion happens), but there was still a reference to the obiRigidbody2D in memory that is able to be called??

At any rate, by removing more and more of the gameobjects in my scene I eventually whittled the object hierarchy down to the above original tree and noticed the mixing of 2D and 3D components and then remembered that it was an ObiRigidbody2D that was being deleted and throwing the error. On a whim, I modified the hierarchy to be:

> Rigidbody 2D
> Box Collider 2D
...
> Rigidbody (i.e. Unity's 3D Rigidbody component)
Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

At runtime, Obi now adds an ObiRigidbody(2D) to both the upper and the lower gameobject. But I guess that's just what is supposed to happen(?)

> Rigidbody2D
ObiRigidBody2D
> Box Collider 2D
...
> Rigidbody (i.e. Unity's 3D Rigidbody component)
ObiRigidbody (i.e. the 3D version of ObiRigidbody)
Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

But most importantly... All the errors go away!]]> The Problem

After upgrading to Obi Cloth 7.1, my project suddenly produced the error

"MissingReferenceException: The object of type 'ObiRigidbody2D' has been destroyed but you are still trying to access it. Your script should either check if it is null or you should not destroy the object."

This error was happening everywhere and on every frame!

The Solution

My project mixes 2D and 3D physics on purpose. The cause of the error was this snippet of gameobject hierarchy

> Rigidbody 2D
> Box Collider 2D
...
> Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

Notice the 2D Unity Rigidbody mixed with the 3D Unity Box Collider and 3D Obi Collider.

The fix is to insert a 3D Unity Rigidbody below the 2D Unity Rigidbody like

> Rigidbody 2D
> Box Collider 2D
...
> Rigidbody (i.e. Unity's 3D Rigidbody component)
Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

I also marked the (3D) Rigidbody as "Is Kinematic" so that the 2D Box Collider and the 3D Box Collider would behave as if they are part of the same Rigidbody (which is the behavior I'm looking for).

Explanation

In the editor, Obi thought that everything was fine with the original hierarchy. ObiCollider was happy because there was a 3D Unity Collider on the same gameobject that the ObiCollider was attached to. And at runtime, just as the online documentation says, Obi "look[ed] for the first Rigidbody component up its hierarchy and attach[ed] an ObiRigidbody component to it." Since the first Rigidbody component Obi encountered was the Rigidbody2D, Obi added an ObiRigidbody2D component to the hierarchy like this

> Rigidbody 2D
Obi Rigidbody 2D

> Box Collider2D
...
> Box Collider (i.e. a 3D collider)
Obi Collider (i.e. the 3D version of ObiCollider

But then Obi threw the above MissingReferenceException.

I traced the Obi code a bit. The error is thrown from ObiRigidbody2D.CacheVelocities() where it turns out that this is null!! I didn't know that was possible Sonrisa