|
@@ -2,59 +2,75 @@
|
|
|
|
|
|
|
|
For breaking API changes see [this document](https://github.com/jrouwe/JoltPhysics/blob/master/Docs/APIChanges.md).
|
|
For breaking API changes see [this document](https://github.com/jrouwe/JoltPhysics/blob/master/Docs/APIChanges.md).
|
|
|
|
|
|
|
|
-## Unreleased changes
|
|
|
|
|
|
|
+## v5.3.0
|
|
|
|
|
|
|
|
### New functionality
|
|
### New functionality
|
|
|
|
|
|
|
|
-* Fixed a number of issues when creating very large `MeshShape`s. `MeshShape`s of up to 110M triangles are possible now, but the actual maximum is very dependent on how the triangles in the mesh are connected.
|
|
|
|
|
|
|
+#### Samples
|
|
|
|
|
+
|
|
|
|
|
+* The Samples and JoltViewer can run on Linux using Vulkan. Make sure to install the Vulkan SDK before compiling (see: `Build/ubuntu24_install_vulkan_sdk.sh`).
|
|
|
|
|
+* The Samples and JoltViewer can run on macOS using Metal.
|
|
|
|
|
+
|
|
|
|
|
+#### MeshShape
|
|
|
|
|
+
|
|
|
|
|
+* `MeshShape`s of up to 110M triangles are possible now, but the actual maximum is very dependent on how the triangles in the mesh are connected.
|
|
|
* Optimized creation of `MeshShape`. Improves build speed by about 25% and reduces number of allocations by a factor of 1000. Allocations caused contention when building meshes from multiple threads.
|
|
* Optimized creation of `MeshShape`. Improves build speed by about 25% and reduces number of allocations by a factor of 1000. Allocations caused contention when building meshes from multiple threads.
|
|
|
* Added `MeshShapeSettings::mBuildQuality` which allows selecting between faster mesh creation or faster run time performance.
|
|
* Added `MeshShapeSettings::mBuildQuality` which allows selecting between faster mesh creation or faster run time performance.
|
|
|
|
|
+
|
|
|
|
|
+#### Character
|
|
|
|
|
+
|
|
|
|
|
+* Added `OnContactPersisted`, `OnContactRemoved`, `OnCharacterContactPersisted` and `OnCharacterContactRemoved` functions on `CharacterContactListener` to better match the interface of `ContactListener`.
|
|
|
|
|
+* Every `CharacterVirtual` now has a `CharacterID`. This ID can be used to identify the character after removal and is used to make the simulation deterministic in case a character collides with multiple other virtual characters.
|
|
|
|
|
+* Added support for `CharacterVirtual` to override the inner rigid body ID. This can be used to make the simulation deterministic in e.g. client/server setups.
|
|
|
|
|
+
|
|
|
|
|
+#### Collision Detection
|
|
|
|
|
+
|
|
|
|
|
+* Added `PhysicsSystem::SetSimShapeFilter`. This allows filtering out collisions between sub shapes within a body and can for example be used to have a single body that contains a low detail simulation shape an a high detail collision query shape. An example of a body that's both a sensor and a rigid body can be found in `ContactListenerTest`.
|
|
|
|
|
+* Added `PhysicsSystem::SetSimCollideBodyVsBody`. This allows overriding the collision detection between two bodies. It can be used to only store the 1st hit for sensor collisions. This makes sensors cheaper if you only need to know if there is an overlap or not. An example can be found in `SimCollideBodyVsBodyTest`.
|
|
|
|
|
+* Added `ClosestHitPerBodyCollisionCollector` which will report the closest / deepest hit per body that the collision query collides with.
|
|
|
|
|
+* Added overridable `CollisionCollector::OnBodyEnd` that is called after all hits for a body have been processed when collecting hits through `NarrowPhaseQuery`.
|
|
|
|
|
+
|
|
|
|
|
+#### New Platforms
|
|
|
|
|
+
|
|
|
|
|
+* Added support for RISC-V, LoongArch and PowerPC (Little Endian) CPUs.
|
|
|
|
|
+* Added support for WASM64.
|
|
|
|
|
+
|
|
|
|
|
+#### Various
|
|
|
|
|
+
|
|
|
* Removed the use of `std::unordered_map` and `std::unordered_set` and replaced them with our own implementation: `UnorderedMap` and `UnorderedSet`.
|
|
* Removed the use of `std::unordered_map` and `std::unordered_set` and replaced them with our own implementation: `UnorderedMap` and `UnorderedSet`.
|
|
|
* Added `MotionProperties::ScaleToMass`. This lets you easily change the mass and inertia tensor of a body after creation.
|
|
* Added `MotionProperties::ScaleToMass`. This lets you easily change the mass and inertia tensor of a body after creation.
|
|
|
* Split up `Body::ApplyBuoyancyImpulse` into `Body::GetSubmergedVolume` and `Body::ApplyBuoyancyImpulse`. This allows you to use the calculated submerged volume for other purposes.
|
|
* Split up `Body::ApplyBuoyancyImpulse` into `Body::GetSubmergedVolume` and `Body::ApplyBuoyancyImpulse`. This allows you to use the calculated submerged volume for other purposes.
|
|
|
-* Added `PhysicsSystem::SetSimShapeFilter`. This allows filtering out collisions between sub shapes within a body and can for example be used to have a single body that contains a low detail simulation shape an a high detail collision query shape.
|
|
|
|
|
-* Added an example of a body that's both a sensor and a rigid body in `ContactListenerTest`.
|
|
|
|
|
* Added binary serialization to `SkeletalAnimation`.
|
|
* Added binary serialization to `SkeletalAnimation`.
|
|
|
-* Added support for RISC-V, LoongArch and PowerPC (Little Endian) CPUs.
|
|
|
|
|
-* Added support for WASM64.
|
|
|
|
|
* Added the ability to add a sub shape at a specified index in a `MutableCompoundShape` rather than at the end.
|
|
* Added the ability to add a sub shape at a specified index in a `MutableCompoundShape` rather than at the end.
|
|
|
-* The Samples and JoltViewer can run on Linux using Vulkan. Make sure to install the Vulkan SDK before compiling (see: `Build/ubuntu24_install_vulkan_sdk.sh`).
|
|
|
|
|
-* The Samples and JoltViewer can run on macOS using Metal.
|
|
|
|
|
* Added `STLLocalAllocator` which is an allocator that can be used in e.g. the `Array` class. It keeps a fixed size buffer for N elements and only when it runs out of space falls back to the heap.
|
|
* Added `STLLocalAllocator` which is an allocator that can be used in e.g. the `Array` class. It keeps a fixed size buffer for N elements and only when it runs out of space falls back to the heap.
|
|
|
-* Added support for `CharacterVirtual` to override the inner rigid body ID. This can be used to make the simulation deterministic in e.g. client/server setups.
|
|
|
|
|
-* Added `OnContactPersisted`, `OnContactRemoved`, `OnCharacterContactPersisted` and `OnCharacterContactRemoved` functions on `CharacterContactListener` to better match the interface of `ContactListener`.
|
|
|
|
|
-* Every `CharacterVirtual` now has a `CharacterID`. This ID can be used to identify the character after removal and is used to make the simulation deterministic in case a character collides with multiple other virtual characters.
|
|
|
|
|
-* Added overridable `CollisionCollector::OnBodyEnd` that is called after all hits for a body have been processed when collecting hits through `NarrowPhaseQuery`.
|
|
|
|
|
-* Added `ClosestHitPerBodyCollisionCollector` which will report the closest / deepest hit per body that the collision query collides with.
|
|
|
|
|
-* Added `PhysicsSystem::SetSimCollideBodyVsBody`. This allows overriding the collision detection between two bodies. It can be used to only store the 1st hit for sensor collisions. This makes sensors cheaper if you only need to know if there is an overlap or not. An example can be found in `SimCollideBodyVsBodyTest`.
|
|
|
|
|
* Added the following constants on PhysicsSystem: `cMaxBodiesLimit`, `cMaxBodyPairsLimit` and `cMaxContactConstraintsLimit`. These constants are the max allowable values for `PhysicsSystem::Init`. Exceeding these will trigger an assert and the system will clamp the values. Note that on a 32 bit system, you'll run out of memory before you reach these values.
|
|
* Added the following constants on PhysicsSystem: `cMaxBodiesLimit`, `cMaxBodyPairsLimit` and `cMaxContactConstraintsLimit`. These constants are the max allowable values for `PhysicsSystem::Init`. Exceeding these will trigger an assert and the system will clamp the values. Note that on a 32 bit system, you'll run out of memory before you reach these values.
|
|
|
|
|
|
|
|
### Bug fixes
|
|
### Bug fixes
|
|
|
|
|
|
|
|
* Fixed bodies gaining more energy than intended due to restitution. E.g. A restitution of 1 could lead to bodies bouncing ever higher.
|
|
* Fixed bodies gaining more energy than intended due to restitution. E.g. A restitution of 1 could lead to bodies bouncing ever higher.
|
|
|
-* `std::push_heap`/`pop_heap` behave differently on macOS vs Windows/Linux when elements compare equal, this made the cross platform deterministic build not deterministic in some cases.
|
|
|
|
|
-* Added overloads for placement new in the `JPH_OVERRIDE_NEW_DELETE` macro, this means it is no longer needed to do `:: new (address) JPH::class_name(constructor_arguments)` but you can do `new (address) JPH::class_name(constructor_arguments)`.
|
|
|
|
|
-* Fixed a GCC warning `-Wshadow=global`.
|
|
|
|
|
* `BodyInterface::AddForce` applied a force per soft body vertex rather than to the whole body, this resulted in a soft body accelerating much more compared to a rigid body of the same mass.
|
|
* `BodyInterface::AddForce` applied a force per soft body vertex rather than to the whole body, this resulted in a soft body accelerating much more compared to a rigid body of the same mass.
|
|
|
|
|
+* `std::push_heap`/`pop_heap` behave differently on macOS vs Windows/Linux when elements compare equal, this made the cross platform deterministic build not deterministic in some cases.
|
|
|
* Removing a sub shape from a `MutableCompoundShape` would not update the bounding box if the last shape was removed, which can result in a small performance loss.
|
|
* Removing a sub shape from a `MutableCompoundShape` would not update the bounding box if the last shape was removed, which can result in a small performance loss.
|
|
|
|
|
+* An empty `MutableCompoundShape` now returns the same local bounding box as `EmptyShape` (a point at (0, 0, 0)). This prevents floating point overflow exceptions.
|
|
|
* VehicleConstraint would override `Body::SetAllowSleeping` every frame, making it impossible for client code to configure a vehicle that cannot go to sleep.
|
|
* VehicleConstraint would override `Body::SetAllowSleeping` every frame, making it impossible for client code to configure a vehicle that cannot go to sleep.
|
|
|
* Fixed `CharacterVirtual::Contact::mIsSensorB` not being persisted in `SaveState`.
|
|
* Fixed `CharacterVirtual::Contact::mIsSensorB` not being persisted in `SaveState`.
|
|
|
* Fixed `CharacterVirtual::Contact::mHadContact` not being true for collisions with sensors. They will still be marked as `mWasDiscarded` to prevent any further interaction.
|
|
* Fixed `CharacterVirtual::Contact::mHadContact` not being true for collisions with sensors. They will still be marked as `mWasDiscarded` to prevent any further interaction.
|
|
|
* Fixed `Character::SetShape` failing to switch when standing inside a sensor / `Character::PostSimulation` finding a sensor as ground collision.
|
|
* Fixed `Character::SetShape` failing to switch when standing inside a sensor / `Character::PostSimulation` finding a sensor as ground collision.
|
|
|
* Fixed numerical inaccuracy in penetration depth calculation when `CollideShapeSettings::mMaxSeparationDistance` was set to a really high value (e.g. 1000).
|
|
* Fixed numerical inaccuracy in penetration depth calculation when `CollideShapeSettings::mMaxSeparationDistance` was set to a really high value (e.g. 1000).
|
|
|
* Bugfix in `Semaphore::Acquire` for non-windows platform. The count was updated before waiting, meaning that the counter would become -(number of waiting threads) and the semaphore would not wake up until at least the same amount of releases was done. In practice this meant that the main thread had to do the last (number of threads) jobs, slowing down the simulation a bit.
|
|
* Bugfix in `Semaphore::Acquire` for non-windows platform. The count was updated before waiting, meaning that the counter would become -(number of waiting threads) and the semaphore would not wake up until at least the same amount of releases was done. In practice this meant that the main thread had to do the last (number of threads) jobs, slowing down the simulation a bit.
|
|
|
-* An empty `MutableCompoundShape` now returns the same local bounding box as `EmptyShape` (a point at (0, 0, 0)). This prevents floating point overflow exceptions.
|
|
|
|
|
-* Fixed a bug in `ManifoldBetweenTwoFaces` that led to incorrect `ContactManifold::mRelativeContactPointsOn2` when the contact normal and the face normal were not roughly parallel. Also it led to possibly jitter in the simulation in that case.
|
|
|
|
|
|
|
+* Fixed a bug in `ManifoldBetweenTwoFaces` that led to incorrect `ContactManifold::mRelativeContactPointsOn2` when the contact normal and the face normal were not roughly parallel. Also it possibly led to jitter in the simulation in that case.
|
|
|
* Fixed `InternalEdgeRemovingCollector` not working when colliding with a very dense triangle grid because it ran out of internal space. Now falling back to memory allocations when this happens to avoid ghost collisions.
|
|
* Fixed `InternalEdgeRemovingCollector` not working when colliding with a very dense triangle grid because it ran out of internal space. Now falling back to memory allocations when this happens to avoid ghost collisions.
|
|
|
* Fixed running out of stack space when simulating a really high number of active rigid bodies.
|
|
* Fixed running out of stack space when simulating a really high number of active rigid bodies.
|
|
|
* Moved the 'broad phase bit' to the highest bit in `BodyID` to avoid running out of `NodeID`s in `BroadPhaseQuadTree` when calling `PhysicsSystem::OptimizeBroadPhase` on a tree with a very high body count.
|
|
* Moved the 'broad phase bit' to the highest bit in `BodyID` to avoid running out of `NodeID`s in `BroadPhaseQuadTree` when calling `PhysicsSystem::OptimizeBroadPhase` on a tree with a very high body count.
|
|
|
* `TempAllocatorImpl` uses 64 bit integers internally to allow for a higher max contact constraint count.
|
|
* `TempAllocatorImpl` uses 64 bit integers internally to allow for a higher max contact constraint count.
|
|
|
* When inserting lots of bodies without using batching, a broad phase tree of depth > 128 can be created. If the `PhysicsSystem` was destructed in this situation, a stack overflow would cause a crash.
|
|
* When inserting lots of bodies without using batching, a broad phase tree of depth > 128 can be created. If the `PhysicsSystem` was destructed in this situation, a stack overflow would cause a crash.
|
|
|
* When calling `PhysicsSystem::Update` with a delta time of 0, contact remove callbacks were triggered by accident for all existing contacts.
|
|
* When calling `PhysicsSystem::Update` with a delta time of 0, contact remove callbacks were triggered by accident for all existing contacts.
|
|
|
-* Fixed 'HingeConstraint' not having limits if `LimitsMin` was set to `-JPH_PI` or `LimitsMax` was set to `JPH_PI`. It should only be turned off if both are.
|
|
|
|
|
-* Fixed 'CylinderShape::GetSupportingFace' returning the wrong face. When the height of a cylinder was small compared to its radius, it would sink more into the ground than needed during simulation.
|
|
|
|
|
|
|
+* Fixed `HingeConstraint` not having limits if `LimitsMin` was set to `-JPH_PI` or `LimitsMax` was set to `JPH_PI`. It should only be turned off if both are.
|
|
|
|
|
+* Fixed `CylinderShape::GetSupportingFace` returning the wrong face. When the height of a cylinder was small compared to its radius, it would sink more into the ground than needed during simulation.
|
|
|
* When there were no active bodies, the step listeners weren't called. This meant they couldn't wake up bodies. The step listeners are now only skipped if the physics system is updated with zero delta time.
|
|
* When there were no active bodies, the step listeners weren't called. This meant they couldn't wake up bodies. The step listeners are now only skipped if the physics system is updated with zero delta time.
|
|
|
* Fixed a race condition in soft body simulation that could break determinism.
|
|
* Fixed a race condition in soft body simulation that could break determinism.
|
|
|
|
|
+* Added overloads for placement new in the `JPH_OVERRIDE_NEW_DELETE` macro, this means it is no longer needed to do `:: new (address) JPH::class_name(constructor_arguments)` but you can do `new (address) JPH::class_name(constructor_arguments)`.
|
|
|
|
|
+* Fixed a GCC warning `-Wshadow=global`.
|
|
|
|
|
|
|
|
## v5.2.0
|
|
## v5.2.0
|
|
|
|
|
|
Jorrit Rouwe