Blender RigidBody Physics broken #72716

New Issue

Tohny · 2019-12-27T00:30:15+01:00

Tohny commented

2019-12-27 00:30:15 +01:00

System Information
Operating system: Windows-10-10.0.18362 64 Bits
Graphics card: GeForce GTX 750 Ti/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 441.66
CPU: R5 2600
RAM: 16GB DDR4

Blender Version
Broken: version: 2.81 (sub 16), branch: master, commit date: 2019-12-04 11:32, hash: f1aa4d18d4
Worked: (optional)

Short description of error
Objects fly away when I click play{F8245480}

Exact steps for others to reproduce the error
[Please describe the exact steps needed to reproduce the issue]
[Based on the default startup or an attached .blend file (as simple as possible)]

**System Information** Operating system: Windows-10-10.0.18362 64 Bits Graphics card: GeForce GTX 750 Ti/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 441.66 CPU: R5 2600 RAM: 16GB DDR4 **Blender Version** Broken: version: 2.81 (sub 16), branch: master, commit date: 2019-12-04 11:32, hash: `f1aa4d18d4` Worked: (optional) **Short description of error** Objects fly away when I click play{[F8245480](https://archive.blender.org/developer/F8245480/Desk_and_Supplies.blend)} **Exact steps for others to reproduce the error** [Please describe the exact steps needed to reproduce the issue] [Based on the default startup or an attached .blend file (as simple as possible)]

Tohny commented

2019-12-27 00:30:15 +01:00

Added subscriber: @Tohny-3

Philipp Oeser commented

2019-12-28 12:50:21 +01:00

Added subscriber: @lichtwerk

Philipp Oeser commented

2019-12-28 12:50:21 +01:00

Changed status from 'Needs Triage' to: 'Archived'

Philipp Oeser closed this issue

2019-12-28 12:50:21 +01:00

Philipp Oeser self-assigned this 2019-12-28 12:50:21 +01:00

Philipp Oeser commented

2019-12-28 12:50:21 +01:00

I would not call this broken [2.79 behave the same way here]

Those spiky colliders are not always easy to set up, but here are a couple of things to consider:

have an eye on the collision margins [in your case the cup where those pencils drop into is 18cm in diameter but has a 4cm collision margin on each polygon -- which is much to high for a sensible sim]
mass can also play a role here [1kg pencils?]
pencils can get away with convex hull collision shapes [faster to simulate, avoids potential problems]

If I (quickly) adjust these a bit, I can get a sim that is reasonably stable:
Desk and Supplies_repro.blend

to really dig into the "problem", here is where bullet has problems computing a normal for a tri (in your file):

btVector3& btVector3::normalize(): Assertion `!fuzzyZero()' failed.

1   raise                                                                                         0x7ffff1ae4e35 
2   abort                                                                                         0x7ffff1acf895 
3   __assert_fail_base.cold                                                                       0x7ffff1acf769 
4   __assert_fail                                                                                 0x7ffff1add566 
5   btVector3::normalize                                     btVector3.h                     300  0x4ad934e      
6   btPrimitiveTriangle::buildTriPlane                       btTriangleShapeEx.h             90   0x4afa0cc      
7   btGImpactCollisionAlgorithm::collide_sat_triangles       btGImpactCollisionAlgorithm.cpp 427  0x4af72c4      
8   btGImpactCollisionAlgorithm::gimpact_vs_gimpact          btGImpactCollisionAlgorithm.cpp 515  0x4af76de      
9   btGImpactCollisionAlgorithm::gimpact_vs_gimpact          btGImpactCollisionAlgorithm.cpp 489  0x4af7565      
10  btGImpactCollisionAlgorithm::gimpact_vs_gimpact          btGImpactCollisionAlgorithm.cpp 475  0x4af74af      
11  btGImpactCollisionAlgorithm::processCollision            btGImpactCollisionAlgorithm.cpp 887  0x4af89c3      
12  btCollisionDispatcher::defaultNearCallback               btCollisionDispatcher.cpp       279  0x4ad2ea7      
13  btCollisionPairCallback::processOverlap                  btCollisionDispatcher.cpp       230  0x4ad3877      
14  btHashedOverlappingPairCache::processAllOverlappingPairs btOverlappingPairCache.cpp      386  0x4ad0de6      
15  btCollisionDispatcher::dispatchAllCollisionPairs         btCollisionDispatcher.cpp       244  0x4ad2c97      
16  btCollisionWorld::performDiscreteCollisionDetection      btCollisionWorld.cpp            228  0x4ad50d7      
17  btDiscreteDynamicsWorld::internalSingleStepSimulation    btDiscreteDynamicsWorld.cpp     496  0x4b31e8d      
18  btDiscreteDynamicsWorld::stepSimulation                  btDiscreteDynamicsWorld.cpp     456  0x4b31d21      
19  RB_dworld_step_simulation                                rb_bullet_api.cpp               212  0x4702b23      
20  BKE_rigidbody_do_simulation                              rigidbody.c                     2037 0x3122bac

Anyways, this does not seem like a real bug in blender, will close...
(feel free though to comment again if you still think you've found a bug...)

I would not call this broken [2.79 behave the same way here] Those spiky colliders are not always easy to set up, but here are a couple of things to consider: - have an eye on the collision margins [in your case the cup where those pencils drop into is 18cm in diameter but has a 4cm collision margin on each polygon -- which is much to high for a sensible sim] - mass can also play a role here [1kg pencils?] - pencils can get away with convex hull collision shapes [faster to simulate, avoids potential problems] If I (quickly) adjust these a bit, I can get a sim that is reasonably stable: [Desk and Supplies_repro.blend](https://archive.blender.org/developer/F8248045/Desk_and_Supplies_repro.blend) to really dig into the "problem", here is where bullet has problems computing a normal for a tri (in your file): ``` btVector3& btVector3::normalize(): Assertion `!fuzzyZero()' failed. ``` ``` 1 raise 0x7ffff1ae4e35 2 abort 0x7ffff1acf895 3 __assert_fail_base.cold 0x7ffff1acf769 4 __assert_fail 0x7ffff1add566 5 btVector3::normalize btVector3.h 300 0x4ad934e 6 btPrimitiveTriangle::buildTriPlane btTriangleShapeEx.h 90 0x4afa0cc 7 btGImpactCollisionAlgorithm::collide_sat_triangles btGImpactCollisionAlgorithm.cpp 427 0x4af72c4 8 btGImpactCollisionAlgorithm::gimpact_vs_gimpact btGImpactCollisionAlgorithm.cpp 515 0x4af76de 9 btGImpactCollisionAlgorithm::gimpact_vs_gimpact btGImpactCollisionAlgorithm.cpp 489 0x4af7565 10 btGImpactCollisionAlgorithm::gimpact_vs_gimpact btGImpactCollisionAlgorithm.cpp 475 0x4af74af 11 btGImpactCollisionAlgorithm::processCollision btGImpactCollisionAlgorithm.cpp 887 0x4af89c3 12 btCollisionDispatcher::defaultNearCallback btCollisionDispatcher.cpp 279 0x4ad2ea7 13 btCollisionPairCallback::processOverlap btCollisionDispatcher.cpp 230 0x4ad3877 14 btHashedOverlappingPairCache::processAllOverlappingPairs btOverlappingPairCache.cpp 386 0x4ad0de6 15 btCollisionDispatcher::dispatchAllCollisionPairs btCollisionDispatcher.cpp 244 0x4ad2c97 16 btCollisionWorld::performDiscreteCollisionDetection btCollisionWorld.cpp 228 0x4ad50d7 17 btDiscreteDynamicsWorld::internalSingleStepSimulation btDiscreteDynamicsWorld.cpp 496 0x4b31e8d 18 btDiscreteDynamicsWorld::stepSimulation btDiscreteDynamicsWorld.cpp 456 0x4b31d21 19 RB_dworld_step_simulation rb_bullet_api.cpp 212 0x4702b23 20 BKE_rigidbody_do_simulation rigidbody.c 2037 0x3122bac ``` Anyways, this does not seem like a real bug in blender, will close... (feel free though to comment again if you still think you've found a bug...)

Tohny commented

2020-01-02 00:09:05 +01:00

Thank you for your support! I keep these tweaks in mind

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

Blender RigidBody Physics broken #72716