| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
wide default. This allows individual prims to differ in the
underlying linkset implementation.
|
|
|
|
| |
Attempt to fix vehicles being twisted off the ground when they go physical.
|
|
|
|
|
| |
Rebuild compound linkset of any child in the linkset changes shape.
Comments and better detailed logging messages.
|
|
|
|
| |
Not all working yet.
|
|
|
|
| |
working yet.
|
|
|
|
|
|
|
| |
child prims so compound shape can do all collisions.
Don't try to build a compound linkset for non-physical linksets.
Remove and replace root body when compound shape is added so collision
cache is rebuilt.
|
|
|
|
| |
changes.
|
|
|
|
| |
'm_physicsScene' to match coding conventions and reduce confusion.
|
| |
|
|
|
|
| |
from BSShapeCollection. Using BSShape* classes to hold references to shape. Simplified shape dependency callbacks. Remove 'PreferredShape' methods and have each class specify shape type. Disable compound shape linkset for a later commit that will simplify linkset implementation.
|
|
|
|
| |
by default. Like more debugging is needed.
|
|
|
|
|
|
|
|
|
|
|
| |
funny unit that is 100 times real density (default 1000).
Fix avatar drifting slowly when stationary flying.
Fix for physical prims getting corrected for being under terrain when it was
just its geometric center that was below terrain.
Add PreUpdatePropertyAction allowing plugable modifiction of phys
parameters returned from Bullet.
Fix an exception setting GravityMultiplier on initialization.
Update DLLs and SOs for good measure (no functional change).
|
|
|
|
| |
simulator to update info.
|
|
|
|
| |
layers for physical (vs simulator) location displacement and linksets.
|
|
|
|
|
| |
child prim physical positions to get out of sync with the view.
More reliably compute the offset of children in a physical linkset.
|
|
|
|
|
| |
when individual child pos/rot changes were implementated a week or so ago.
Remove some passing of inTaintTime flag when it was never false.
|
|
|
|
|
|
| |
physical by delaying setting until pre-step time. Change vehicle.Refresh()
to schedule the pre-step setting.
Comments and updating of TODO list.
|
| |
|
|
|
|
| |
on the ground. Set defaults for vehicle factors to one. Debug logging changes.
|
|
|
|
|
|
| |
Conflicts:
OpenSim/Region/Physics/BulletSPlugin/BSPrim.cs
|
| |
|
|
|
|
| |
and vehicles and clean up code by removing their kludgyness.
|
|
|
|
|
| |
without rebuilding the whole compound shape. Should make vehicles move
smoother.
|
|
|
|
|
| |
It was a dumb idea to try and do a nose over feature for jumping cars anyway.
Add better logging of native shape creation/reuse so can tell the difference.
|
|
|
|
|
| |
objects have been fetched.
Update TODO list with more work.
|
|
|
|
|
| |
because the computations are wrong.
Add VehicleTorqueImpulse routines.
|
|
|
|
| |
Move physical prim above ground if it is underground. Previously tried to correct by applying and up force but the prim would never go through the ground.
|
|
|
|
|
|
|
| |
Necessitated allowing simulator and physical position of a body to
get out of sync since Bullet assumes that <0,0,0> is the center of mass.
Update DLLs and SOs for the UpdateChildTransform so positions of
individual prim in a linkset can be implemented.
|
|
|
|
| |
Only initialization and debug fuctions left.
|
|
|
|
| |
(.ToString(X)) and replace it with a method on BulletBody, BulletShape, ...
|
|
|
|
| |
functions.
|
|
|
|
| |
BulletSim core to using the BulletSimAPITemplate. Physical object creation and destruction first.
|
|
|
|
|
| |
object is selected.
Update TODO list.
|
|
|
|
|
| |
rebuilt linkset. I was burnt by making get/set methods with side
effects. I should know better.
|
|
|
|
|
|
|
| |
Properly return root mass as mass of just the root prim rather
than the mass of the linkset. SOG has the logic to add the masses
together to get the linkset mass.
Update TODO list.
|
|
|
|
| |
doing unsit, the order of operations on the prims and the vehicle is very chaotic and not in a good order so the root prim was being left physical and thus it fell for a bit. Also changed default of velocity scaling to be closer to the movement standard.
|
|
|
|
| |
changes from the physics engine can be differentiated from changes made by the user. This eliminates a linkset rebuild loop. Also add logic to not rebuild or freak out when the object/linkset crosses a terrain boundry.
|
|
|
|
| |
doesn't need to do that bookkeeping because SOG/SOP already does it.
|
|
|
|
| |
This caused movement problems and large prim vehicles to take up a LOT of simulation time.
|
| |
|
|
|
|
| |
single prim vehicles not working (the surf board now zooms).
|
|
|
|
|
|
|
|
|
| |
While fixing the above, add methods to physical body and shape pointer
wrapper so routines won't have to know that IntPtr.Zero means no
physical instance.
Fix problem with physical linksets failing after a few sits and
unsits by properly restoring child prom positions for compound
linksets after multiple selection and deselections.
|
|
|
|
| |
restore a child's position in the world based on its position in the moved linkset).
|
|
|
|
| |
Linkset.MakeStatic() after call to ForceActivationState2() since linkset might change activation state. Make BSPrim.CreateGeomAndObject public as linkset rebuilding might need access to it. Only rebuild prim if selection state is actually changes -- OpenSimulator calls PhysObject.Selected() multiple times whenever a prim is selected or deselected.
|
| |
|
|
|
|
| |
conflicts with PhysicsShape type defined in later libOMV
|
|
|
|
| |
getting simplified out of existance someday) and update all the references to that enum.
|
| |
|
|
|
|
|
| |
Remove all compilation warnings (mostly 'protected' in sealed classes.)
Add the dynamicAabbEnable parameter to creation of compound shapes.
|
|
|
|
| |
known. This makes sure the correct accounting is done for the particular shape.
|