This more-or-less gets rid of the shadow system's only depencency on FFP
stuff. All that remains is it using OSG cameras, which OSG provides a
uniform-based implementation of, too, which we can trivially migrate to.
This should mean we're not eating any of the ~8 FPP-friendly texture
units, which is good as Morrowind models can use all of those on their
(although they very rarely do), and instead use some of the ~160
shader-only texture image units. This just requires not calling
glEnable(GL_TEXTURE_2D), accomplished by changing
setTextureAttributeAndModes to setTextureAttribute.
Also changes from using glTexGen and its eye plane matrices to pass the
shadow space matrix for each light to explicit uniforms. Thankfully, the
maths was a simple combination of the valid region matrix and eye plane
matrix maths.
As of this commit, I believe this kills shadows in one eye for stereo
rendering.
As well as fixing the MacOS build, it meddled with a bunch of files that
already did the right thing.
This reverts commit 8162541a14, reversing
changes made to e16c451d08.
`ciCompareLen(a, b, b.size()) == 0` expression is an equivalent of checking for
equality of `a` prefix with size `b.size()` with `b`.
`ciCompareLen(a, b, a.size()) == 0` is also the same thing but `a` is a prefix
`b` should start with.
Actors may have different collision shapes. Currently there are axis-aligned
bounding boxes and rotating bounding boxes. With AABB it's required to use
bounding cylinder for navmesh agent to avoid providing paths where actor can't
pass. But for rotating bounding boxes cylinder with diameter equal to the front
face width should be used to not reduce of available paths. For example rats
have rotating bounding box as collision shape because of the difference between
front and side faces width.
* Add agent bounds to navmesh tile db cache key. This is required to distinguish
tiles for agents with different bounds.
* Increase navmesh version because navmesh tile db cache key and data has changed.
* Move navmesh version to the code to avoid misconfiguration by users.
* Fix all places where wrong half extents were used for pathfinding.
That are based on half extents used to find path over navmesh which is different
for interior and exterior cells.
Use common functions to get agent height and radius for actor path rendering and
navmesh generation.
Useful when need to find tiles with high number of updates.
Add debug Lua package with new functions to toggle render mode and set navmesh
render mode.
Multiview shaders.
Refactor Frustum management
Rewrite shared shadow map
cull mask should respect stereo
Stereo savegame screencap
LocalMap refactoring
use the vertex buffer hint instead of the display list patch to enable/disable display lists
Character preview fixes
With this PR we refactor `StringUtils::replaceAll` to accept `string_view` as suggested in a code comment. In addition, while we are touching this rebuild happy file, we slim it down a bit by moving a few sparingly used functions elsewhere.
This PR removes dummy serialisers for `StateSetUpdater`, `NodeCallback` and the respective `META` macros that trigger serialisation requirement here.
`StateSetUpdater` and `NodeCallback` are just base classes that can not be used on their own, so there is no need to incorporate them into serialisation. These changes might have minor effects on derived classes that forget to override `className()`, `libraryName()` through `META`, but it makes hardly a difference to now serialise such classes as a dysfunctional `osg::Callback` instead of a dysfunctional `SceneUtil::NodeCallback`.
We currently apply a strange algorithm to `LightManager::mStateSetCache`. For some reason this algorithm inserts hashed keys into `std::map` in a way that fails to handle hash collisions and exhibits worse lookup complexity than `std::unordered_map`. With this PR we just use `std::unordered_map` here.
With this PR we apply `lightMask` to a `Transform` node we create specifically for a light. This mask will allow us to stop traversing such nodes sooner and avoid costly processing associated with `Transform` nodes in the cull visitor.
With this PR we refactor `SceneUtil::KeyframeController` not to require `virtual osg::Callback` inheritance. I suppose such `virtual` overhead is not justified here because it negatively impacts many other classes we derive from `osg::Callback`.
This PR aims to spark the retirement of a questionable pattern I have found all over our code base. I will illustrate how this pattern encourages code duplication, lacks type safety, requires documentation and can be prone to bugs.
```
std::map<std::string, Object> mMap; // Stored in all lowercase for a case-insensitive lookup
std::string lowerKey = Misc::StringUtils::lowerCase(key);
mMap.emplace(lowerKey, object);
std::string lowerKey = Misc::StringUtils::lowerCase(key);
mMap.find(lowerKey);
mMap.find(key); // Not found. Oops!
```
An alternative approach produces no such issues.
```
std::unordered_map<std::string, Object, Misc::StringUtils::CiHash, Misc::StringUtils::CiEqual> mMap;
mMap.emplace(key, object);
mMap.find(key);
```
Of course, such an alternative will work for a `map` as well, but an `unordered_map` is generally preferable over a `map` with these changes because we have moved `lowerCase` into the comparison operator.
In this PR I have refactored `Animation::mNodeMap` accordingly. I have reviewed and adapted all direct and indirect usage of `Animation::mNodeMap` to ensure we do not change behaviour with this PR.
This PR fixes a crash caused by the improperly ensured lifetime of RigGeometry::mSourceGeometry. mSourceGeometry was not adequate to ensure mSourceGeometry would outlive mGeometry because we extend mGeometrys lifetime beyond this lifetime by passing mGeometry to the draw traversal instead of this.
In addition,
We add important comments.
We detect and prevent generally unsafe operations in high level code.
We add a sprinkling of const to help clarify intentions.
Currently, we use an `UnrefQueue` which supposedly aims to transfer destruction costs to another thread. The implications of this unusual pattern can not be well understood because some allocators might free resources more efficiently if they are freed by the same thread that allocated them. In addition, `UnrefQueue` complicates the validation of thread safety in our engine. Lastly, our current usage of `UnrefQueue` triggers `ref()`, `unref()` atomic operations as objects are passed into the queue. These operations could be more expensive than the actual destruction.
With this PR we thus remove `UnrefQueue`. We can expect these changes to have a minor impact at most because we free most resources elsewhere in `ResourceSystem::updateCache`.
This variable is only used in ShadowsBin constructor and it's initialized each
time before constructor call so required value can be just passed into
ShadowsBin ctor.
Make ShadowsBin default constructor private because it is required by osg even
it's not actually called.
This PR proposes a simple change to `RigGeometry` `dirtyGLObjects` logic.
1. We will avoid dirtying unmodified arrays.
2. We can drop an osg version guard since `Drawable::dirtyGLObjects` is not nearly as harmful as `Geometry::dirtyGLObjects`.
3. We will avoid crashes in an as yet unfinished future PR concerning `Array` sharing improvements.
osg::NodeVisitor is designed to recursively call virtual apply signatures until we find an implemented signature. Encountered nodes that we do not explicitely handle will trigger additional virtual function calls. With this PR we avoid these additional virtual function calls in the particularly costly ComputeLightSpaceBounds by adding some explicit signatures.
Currently, we always traverse the scene graph an additional time with a ComputeLightSpaceBounds visitor during shadow casting. ComputeLightSpaceBounds is only useful when the shadow casting mask allows us to shrink the bounds of the rendered scene, so we guard its traversal with a check against getCastsShadowTraversalMask. In practice, this guard never works because we build the traversal mask inclusively.
With this PR we limit the getCastsShadowTraversalMask check to relevant masks. This new check allows us to skip a superfluous ComputeLightSpaceBounds traversal with most settings.
We skip this during node path iterations. this is not a node we are interested in.
We avoid allocating a new mGeomToSkelMatrix per frame and avoid a ref_ptr associated with its update.
We speed up a search for the Skeleton node by adding a continue; condition prior to an expensive dynamic_cast.
We can expect marginally improved loading times with this PR. Drawable, Transform and Node counts in stats panels are expected to remain unchanged - this PR does not add new scene graph optimisations, it just increases the speed with which we apply existing ones.
1. We add explicit `NodeVisitor::apply` overrides for commonly encountered node types to avoid additional virtual function calls per node associated with the default `apply` implementation.
2. We skip pushing `StateSet`s when `_mergeAlphaBlending` is enabled or the `StateSet` contains no relevant state.
3. We add a specialised variant of `CollectLowestTransformsVisitor::addTransform` accepting `MatrixTransform` to avoid matrix copies and multiplications.
With this PR we optimise a function that is called quite often when loading new cells.
We remove avoidable dynamic_casts.
We remove an unused pair.second element.
We convert a map to an unordered_map because its ordering is irrelevant in this case.
We avoid adding the root Skeleton node to the bones' node path.
Currently, we create a new ComputeLightSpaceBounds visitor per frame. Within this visitor, we require excessive memory allocations, mainly a new osg::RefMatrix per encountered Transform node.
With this PR we reuse a single ComputeLightSpaceBounds visitor across frames and enable the createOrReuseMatrix functionality to avoid allocating new matrices every frame. osgUtil::CullVisitor internally uses the same approach.
Currently, we run culling tests against all lights in the scene during LightListCallback::pushLightState. We can avoid most of these tests by removing off-screen lights at an earlier stage. We should benchmark the cumulative time spent within LightListCallback::pushLightState before and after this PR.
`handle_stateset` is not needed because `UpdateMatrixTransform` is a `NodeCallback` only allowed to be set on a `Node`. `Geode` and `Drawable` do not need explicit logic because they are both derived from `Node`.
