Interface to abstract all types of pass definitions (see CompositorPassType):
More...
#include <OgreCompositorPassDef.h>
Interface to abstract all types of pass definitions (see CompositorPassType):
This class doesn't do much on its own. See the derived types for more information A definition is shared by all pass instantiations (i.e. Five CompositorPassScene can share the same CompositorPassSceneDef) and are assumed to remain const throughout their lifetime.
- Modifying a definition while there are active instantiations is undefined. Some implementations may see the change (eg. changing CompositorPassSceneDef::mFirstRQ) immediately while not see others (eg. changing CompositorPassSceneDef::mCameraName) Also crashes could happen depending on the changes being made.
◆ UavDependencyVec
◆ CompositorPassDef()
◆ ~CompositorPassDef()
virtual Ogre::CompositorPassDef::~CompositorPassDef |
( |
| ) |
|
|
virtual |
◆ getParentTargetDef()
◆ getRtIndex()
uint32 Ogre::CompositorPassDef::getRtIndex |
( |
| ) |
const |
◆ getType()
◆ setAllClearColours()
void Ogre::CompositorPassDef::setAllClearColours |
( |
const ColourValue & |
clearValue | ) |
|
◆ setAllLoadActions()
◆ setAllStoreActions()
◆ mClearColour
◆ mClearDepth
float Ogre::CompositorPassDef::mClearDepth |
◆ mClearStencil
uint32 Ogre::CompositorPassDef::mClearStencil |
◆ mColourWrite
bool Ogre::CompositorPassDef::mColourWrite |
When false will not really bind the RenderTarget for rendering and use a null colour buffer instead.
Useful for depth prepass, or if the RTT is actually an UAV. Some passes may ignore this setting (e.g. Clear passes)
◆ mExecutionMask
uint8 Ogre::CompositorPassDef::mExecutionMask |
◆ mExposedTextures
◆ mFlushCommandBuffers
bool Ogre::CompositorPassDef::mFlushCommandBuffers |
Whether to flush the command buffer at the end of the pass.
This can incur in a performance overhead (see OpenGL's glFlush and D3D11' ID3D11DeviceContext::Flush) for info. Usually you want to leave this off. However for VR applications that must meet VSync, profiling may show your workload benefits from submitting earlier so the GPU can start right away executing rendering commands.
The main reason to use this is in CPU-bound scenarios where the GPU starts too late after sitting idle.
◆ mIdentifier
uint32 Ogre::CompositorPassDef::mIdentifier |
Custom value in case there's a listener attached (to identify the pass)
◆ mIncludeOverlays
bool Ogre::CompositorPassDef::mIncludeOverlays |
TODO: Refactor OgreOverlay to remove this design atrocity.
A custom overlay pass is a better alternative (or just use their own RQ)
◆ mLoadActionColour
◆ mLoadActionDepth
◆ mLoadActionStencil
◆ mNumInitialPasses
uint32 Ogre::CompositorPassDef::mNumInitialPasses |
Number of times to perform the pass before stopping. -1 to never stop.
◆ mNumViewports
uint32 Ogre::CompositorPassDef::mNumViewports |
◆ mProfilingId
String Ogre::CompositorPassDef::mProfilingId |
◆ mReadOnlyDepth
bool Ogre::CompositorPassDef::mReadOnlyDepth |
◆ mReadOnlyStencil
bool Ogre::CompositorPassDef::mReadOnlyStencil |
◆ mShadowMapFullViewport
bool Ogre::CompositorPassDef::mShadowMapFullViewport |
Only used if mShadowMapIdx is valid (if pass is owned by Shadow Nodes).
If true, we won't force the viewport to fit the region of the UV atlas on the texture, and respect mVp* settings instead.
◆ mShadowMapIdx
uint32 Ogre::CompositorPassDef::mShadowMapIdx |
Shadow map index it belongs to (only filled in passes owned by Shadow Nodes)
◆ mSkipLoadStoreSemantics
bool Ogre::CompositorPassDef::mSkipLoadStoreSemantics |
Ignore mLoadAction*/mStoreAction*.
Useful when doing multiple passes and you want to continue using the same pass semantics opened by a previous pass
Be careful with this setting. It can silently break a lot of stuff
Only valid for PASS_QUAD and PASS_SCENE
◆ mStoreActionColour
◆ mStoreActionDepth
◆ mStoreActionStencil
◆ mUavDependencies
◆ mViewportModifierMask
uint8 Ogre::CompositorPassDef::mViewportModifierMask |
◆ mVpRect
◆ mWarnIfRtvWasFlushed
bool Ogre::CompositorPassDef::mWarnIfRtvWasFlushed |
Will issue a warning (by raising an exception) if Ogre is forced to flush the RenderTarget, which is very bad for performance on mobile, and can cause serious performance problems in Desktop if using MSAA, and also cause correctness problems (i.e.
bad rendering) if store action is StoreAction::Resolve.
- No warning will be issued if the RenderTargets getting flushed have their LoadAction set to LoadAction::Clear (or LoadAction::ClearOnTilers on tilers).
The documentation for this class was generated from the following file: