Bug / Crash One Particle acts like kinematic

[alex798]

Hello, for some reason part of my rope don't move with ropes end(as shown)- what could be a cause?

And even when i stretch rope stretches only part before this unmoving part...

P.S. and another question- how to make just model from rope(static, notUpdating caple)

[josemendez]

Hello, for some reason part of my rope don't move with ropes end(as shown)- what could be a cause?

Without more info it's difficult to tell. Could be an attachment referencing that particle group, could be a collision, etc. Can you describe how you've setup your rope?

P.S. and another question- how to make just model from rope(static, notUpdating caple)

Simply disable the ObiRope component (just the component, not the whole object), but keep the renderer enabled.

[alex798]

Without more info it's difficult to tell. Could be an attachment referencing that particle group, could be a collision, etc. Can you describe how you've setup your rope?


Simply disable the ObiRope component (just the component, not the whole object), but keep the renderer enabled.

Atachment looks like that. What do u mean by describe setup?

Disabling will prevent any physics calculations?

[josemendez]

Atachment looks like that. What do u mean by describe setup?

I mean:

- Describing where is the rope in the hierarchy in relation to the solver.
- Are you using particle tearing?
- Are you dynamically resizing the rope using cursors?
- Are you creating/destroying attachments at runtime?
- What are your blueprint settings? (resolution, thickness, pool size)
- What are your solver settings? (sleep threshold, constraints, etc)
- How are collisions set up (type and placement of colliders, their thickness, etc)

Just more info on how you're using the system. A simple screenshot of the rope isn't enough to work with.

Disabling will prevent any physics calculations?

Disabling ObiRope will prevent any physics calculations. You'll end up with just a rendered mesh.

[alex798]

I mean:

- Describing where is the rope in the hierarchy in relation to the solver.
- Are you using particle tearing?
- Are you dynamically resizing the rope using cursors?
- Are you creating/destroying attachments at runtime?
- What are your blueprint settings? (resolution, thickness, pool size)
- What are your solver settings? (sleep threshold, constraints, etc)
- How are collisions set up (type and placement of colliders, their thickness, etc)

Just more info on how you're using the system. A simple screenshot of the rope isn't enough to work with.


Disabling ObiRope will prevent any physics calculations. You'll end up with just a rendered mesh.

-Direct child
-no tearing
-no cursor(but stretch because static attachment)
-no
-res=0.3, thick=0.01, pool=0;
-sleep=0.1, constrains: dist:par,20,1; bend:par, 2,1; Coll:par,5,1; Fric:par,3,1; Pin:par,10,1;
-Colliders: table: meshLevelCube+ a little down(to make thicker penetration), Coffeemachine: meshLevelCube; Plug: boxCollider with some space not to touch connected particle collider

[josemendez]

-sleep=0.1, constrains: dist:par,20,1; bend:par, 2,1; Coll:par,5,1; Fric:par,3,1; Pin:par,10,1;
-Colliders: table: meshLevelCube+ a little down(to make thicker penetration), Coffeemachine: meshLevelCube; Plug: boxCollider with some space not to touch connected particle collider

A sleep parameter of 0.1 is huge. That will freeze most particles in place, could be the problem you're having.
Try using something smaller, like 0.0001.

Also, don't use so many constraint iterations unless you've got a good reason. In most cases, using 1 iteration and multiple substeps (4-6) will conserve energy better and result in less stretching. See:
http://obi.virtualmethodstudio.com/manua...gence.html

[alex798]

A sleep parameter of 0.1 is huge. That will freeze most particles in place, could be the problem you're having.
Try using something smaller, like 0.0001.

Also, don't use so many constraint iterations unless you've got a good reason. In most cases, using 1 iteration and multiple substeps (4-6) will conserve energy better and result in less stretching. See:
http://obi.virtualmethodstudio.com/manua...gence.html

Set sleep to suggested(tried this earlier too) but problem persist- one particle is like glued(surrounding particles is OK)...
Used big sleep to prevent any jittering near colliders...

I used the same constraint setting as in sample Rocker scene(if it's wrong- better change example), will look into bigger updater...

[josemendez]

Set sleep to suggested(tried this earlier too) but problem persist- one particle is like glued(surrounding particles is OK)...
Used big sleep to prevent any jittering near colliders...

There should be very little or no jittering near colliders even with zero sleep threshold.
Would it be possible for me to take a look at your scene? You can send it to support(at)virtualmethodstudio.com and I'll diagnose the problem.

I used the same constraint setting as in sample Rocker scene(if it's wrong- better change example), will look into bigger updater...

The sample rocker scene uses many iterations in order to obtain a more dampened simulation, while still reducing stretching. Usually you want more lively results, so it's best to use substeps for adjusting overall look and then fine-tune each constraint type using iterations.

How the simulation works and how iterations/substeps affect it is explained in the manual in great detail:
http://obi.virtualmethodstudio.com/manua...gence.html

[alex798]

There should be very little or no jittering near colliders even with zero sleep threshold.
Would it be possible for me to take a look at your scene? You can send it to support(at)virtualmethodstudio.com and I'll diagnose the problem.


The sample rocker scene uses many iterations in order to obtain a more dampened simulation, while still reducing stretching. Usually you want more lively results, so it's best to use substeps for adjusting overall look and then fine-tune each constraint type using iterations.

How the simulation works and how iterations/substeps affect it is explained in the manual in great detail:
http://obi.virtualmethodstudio.com/manua...gence.html

support@virtualmethodstudio.com - right?- sent part of scene (alex798@inbox.ru - OfficeRope)- hope that there is everything nessassary...

I need sollid connection of cable with plug and coffeeMachine, and good collision with table and floor...

---

There should be very little or no jittering near colliders even with zero sleep threshold.
Would it be possible for me to take a look at your scene? You can send it to support(at)virtualmethodstudio.com and I'll diagnose the problem.


The sample rocker scene uses many iterations in order to obtain a more dampened simulation, while still reducing stretching. Usually you want more lively results, so it's best to use substeps for adjusting overall look and then fine-tune each constraint type using iterations.

How the simulation works and how iterations/substeps affect it is explained in the manual in great detail:
http://obi.virtualmethodstudio.com/manua...gence.html

I know what's the cause- I copied existed rope and just put new transforms into components- and this causing tyhis behaviours- needed create new components and there put parameters...

[josemendez]

support@virtualmethodstudio.com - right?- sent part of scene (alex798@inbox.ru - OfficeRope)- hope that there is everything nessassary...

I need sollid connection of cable with plug and coffeeMachine, and good collision with table and floor...

---

I know what's the cause- I copied existed rope and just put new transforms into components- and this causing tyhis behaviours- needed create new components and there put parameters...

Just downloaded your project and was looking at it, but you beat me at finding the cause

let me know if you need further help. cheers!