Age | Commit message (Collapse) | Author |
|
project/gltf_development
|
|
|
|
"possible loss of precision" warnings
|
|
* Migrate GLTF scene rendering to its own shaders
* Add support for ambient occlusion map separate from metallic roughness map (or absent)
* Use UBO's for GLTF joints
* Better error handling of downloading GLTF assets
|
|
LF, and trim trailing whitespaces as needed
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ifdef'd code for potential skinning speed up to avoid lots of int<->float conversions (expensive and static for min space investment)
as updating rigged VBs shows up as a profiling bottleneck for Low rendering (where we actually use CPU skinning).
|
|
|
|
|
|
are enabled.
|
|
get rebuilt after invalid joints are scrubbed.
|
|
|
|
|
|
|
|
|
|
|
|
for rigged meshes in animated objects
|
|
DebugAnimatedObjects output
|
|
constrain encroachment by animated objects
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
If the only use of a variable is within llassert(), have to make the
declaration conditional on SHOW_ASSERT rather than guesswork about release
builds.
|
|
skin weights
|
|
|
|
support a feature we no longer have. This also incidentally fixes any bugs caused by the joint remapping code.
|
|
|
|
|
|
names defined
|
|
|
|
asserts to verify.
|
|
some content.
|
|
|
|
|
|
|
|
a bug somewhere upstream.
|
|
|
|
|
|
|
|
|
|
weights where needed. Warn that related debug settings require restart to take effect.
|