while not a major issue per-se, the sheer number of times the engine has to jump back in memory and backfill data in a given class can add up. First run of... many.,
Added a check to SceneObject::setProcessTick() to prevent objects that have mounts from being removed from the process list so the processAfter chain of the mounted objects is not broken.
renderPrePassMgr.cpp related:
A) shifting .addFeature( MFT_XYZ); calls from ProcessedShaderMaterial::_determineFeatures to ProcessedPrePassMaterial::_determineFeatures
B) mimicking the "// set the XXX if different" entries from RenderMeshMgr::render in RenderPrePassMgr::render
C) fleshing out ProcessedPrePassMaterial::getNumStages() so that it shares a 1:1 correlation with ProcessedShaderMaterial::getNumStages()
D) causing inline void Swizzle<T, mapLength>::ToBuffer( void *destination, const void *source, const dsize_t size ) to silently fail rather than fatally assert if a source or destination buffer is not yet ready to be filled. (support for #customTarget scripted render targets)
Reflections:
A) removing reflectRenderState.disableAdvancedLightingBins(true); entries. this would otherwise early out from prepass and provide no color data whatsoever.
B) removing the fd.features.addFeature( MFT_ForwardShading ); entry forcing all materials to be forward lit when reflected.
C) 2 things best described bluntly as working hacks:
C1) when reflected, a scattersky is rotated PI along it's z then x axis in order to draw properly.
C2) along similar lines, in terraincellmaterial, we shut off culling if it's a prepass material.
Skies: scattersky is given a pair of rotations for reflection purposes, all sky objects are given a z value for depth testing.
SPECIAL NOTE: highly suggest https://github.com/GarageGames/Torque3D/pull/1441 or a variation thereof to prevent debug spew and false-postives for occlusion results.
With significant research, development and prototyping assistance from both @andr3wmac (shaders and partial hook work), and @LuisAntonRebollo (additional culling)
System operates as follows:
1) materials are given an additional castDynamicShadows boolean entry. (Default at time of writing is true by request. Personal usage at time of writing defaults to false. value is default-initialized in materialDefinition.cpp. script/gui exposed)
2) lights are given a staticRefreshFreq and dynamicRefreshFreq (in milliseconds). script/gui exposed
3) materials are (effectively) sorted into dynamic and static shadowmap render lists based on flag. (see shadowMapPass.cpp)
4) initial shadowmaps are generated for each light and 'list'.
5) as each refreshFreq times out, the relevant shadowmap for a given light is refreshed.
Special notes:
dynamicRefreshFreq for all lights is set to a (script exposed) 8MS refresh timer.
StaticRefreshFreq for the lions share of lights defaults to 250 MS (1/4 of a second)
scattersky's embedded light, which is intended to operate in a mobile manner, defaults to 8
to reiterate, these are all customizable per-light via script/inspector gui in the case of alternate project needs.
Many instances of a function or expression being used repeatedly, which can lower performance.
Fixed it in these cases by creating on local var, reference or pointer that's used instead.
- Updated to work with 0.5.x SDK
- Uses Oculus Rendering rather than PostFX
- Stereo rendering refactored so more rendering info is grabbed from the DisplayDevice
- Implements an Offscreen Canvas for in-game gui with oculus
- Message dialogs and metrics display can now go to the OffScreen Canvas (if oculus demo is setup correctly)
For some reason returning true/false rather that 1/0 from
_getRenderEnabled would cause errors in some comparison cases to see if
it was true or not (would treat it as if it was a string/word rather
than a bool or int).
