Adds Mesh equivalents of `multimesh_get_buffer_rd_rid` so GPU-driven mesh
workflows in a GDExtension can write directly into a surface's vertex,
attribute, skin, or index buffer from a compute pipeline, avoiding the
CPU round-trip through `mesh_surface_update_*_region`.
A new `ARRAY_FLAG_USE_STORAGE_BUFFER` surface format flag opts the backing
buffers into storage-buffer usage so they can be bound in compute shaders.
Vulkan/D3D12/Metal only, OpenGL returns an invalid RID.
The interface SurfaceTool presents for tangents and all its usages work
with tangent frames represented as 4 values: tangent direction and
scalar orientation (+1 or -1).
However, internally it was represented as tangent+bitangent (mistakenly
called binormal), which introduces additional conversions back & forth
and is problematic for replacing the tangent space generation algorithm.
This change converts the internal storage and adjusts the calling code
accordingly. Note that since SurfaceTool::set_tangent is exposed to
scripts, I've kept the Plane interface there even though it's misleading
to use this type here.
A number of headers in the codebase included `rendering_server.h` just for
some enum definitions. This means that any change to `rendering_server.h` or
one of its dependencies would trigger a massive incremental rebuild.
With this change, we decouple a number of classes from `rendering_server.h`,
greatly speeding up incremental rebuilds for that area.
On my machine, this reduces incremental compilation time after an edit of
`rendering_server.h` by 60s (from 2m57s).
We now cache the Node*<>TreeItem* mapping in the SceneTreeEditor. This
allows us to make targeted updates to the Tree used to display the scene
tree in the editor.
Previously on almost all changes to the scene tree the editor would
rebuild the entire widget, causing a large number of deallocations an
allocations. We now carefully manipulate the Tree widget in-situ saving
a large number of these allocations.
In order to know what Nodes need to be updated we add a
editor_state_changed signal to Node, this is a TOOLS_ENABLED,
editor-only signal fired when changes to Node happen that are relevant
to editor state.
We also now make sure that when nodes are moved/renamed we don't check
expensive properties that cannot contain NodePaths. This saves a lot of
time when SceneTreeDock renames a node in a scene with a lot of
MeshInstances. This makes renaming nodes go from ~27 seconds to ~2
seconds on large scenes.
SceneTreeEditor instances will now also not do all of the potentially
expensive update work if they are invisible. This behavior is turned off
by default so it won't affect existing users. This change allows the
editor to only update SceneTreeEditors that actually in view. In
practice this means that for most changes instead of updating 6
SceneTreeEditors we only update 1 instantly, and the others only when
they become visible.
There is definitely more that could be done, but this is already a
massive improvement. In complex scenes we see an improvement of 10x,
things that used to take ~30 seconds now only take 2.
This fixes#83460
I want to thank KoBeWi, TokisanGames, a-johnston, aniel080400 for
their tireless testing. And AeioMuch for their testing and providing a
fix for the hover issue.
* Replaces `find(...) != -1` with `contains` for `String`
* Replaces `find(...) == -1` with `!contains` for `String`
* Replaces `find(...) != -1` with `has` for containers
* Replaces `find(...) == -1` with `!has` for containers
* Disabled 2D navigation tests as they do not work
* Disabled tests for primitive meshes
* Unbound some `Mesh` methods that rely on 3D resources
* Unexposed `World3D` as it depends on physics (and isn't useful)
* Unexposed some `Viewport` vrs related properties that had unexposed
methods
This cleans up a few more cases of uint32_t->uint64_t
Importantly this fixes an edge case in the axis-angle compression by
using the pre-existing Basis methods instead
This allows Godot to automatically compress meshes to save a lot of bandwidth.
In general, this requires no interaction from the user and should result in
no noticable quality loss.
This scheme is not backwards compatible, so we have provided an upgrade
mechanism, and a mesh versioning mechanism.
Existing meshes can still be used as a result, but users can get a
performance boost by reimporting assets.
* This should optimize GDScript function calling _enormously_.
* It also should simplify the GDScript VM considerably.
NOTE: GDExtension calling performance has most likely been affected until going via ptrcall is fixed.
Thaddeus Crews
nikitalita
Kotob M
Arseny Kapoulkine
Rémi Verschelde
kobewi
Aaron Franke
Lukas Tenbrink
あるる / きのもと 結衣
Yufeng Ying
Michael Alexsander
A Thousand Ships
HP van Braam
ze2j
Adam Scott
David Snopek
Radiant
Lyuma
clayjohn
Ninni Pipping
Juan Linietsky
smix8
Chris Hutchinson
RedworkDE
SekoiaTree