x
Yes
No
Do you want to visit DriveHQ English website?
Inicio
Características
Precios
Prueba gratuita
Software cliente
Acerca de nosotros
Servidor de archivos
|
Solución de copias de seguridad
|
Servidor FTP
|
Servidor de correo electrónico
|
Alojamiento web
|
Software cliente
Servidor de archivos
Solución de copia de seguridad
Servidor FTP
Servidor de correo electrónico
Alojamiento web
Software cliente
OgreRenderQueueSortingGrouping.h - Hosted on DriveHQ Cloud IT Platform
Arriba
Subir
Descargar
Compartir
Publicar
Nueva carpeta
Nuevo archivo
Copiar
Cortar
Eliminar
Pegar
Clasificación
Actualizar
Ruta de la carpeta: \\game3dprogramming\materials\GameFactory\GameFactoryDemo\references\ogre\include\OgreRenderQueueSortingGrouping.h
Girar
Efecto
Propiedad
Historial
/* ----------------------------------------------------------------------------- This source file is part of OGRE (Object-oriented Graphics Rendering Engine) For the latest info, see http://www.ogre3d.org/ Copyright (c) 2000-2006 Torus Knot Software Ltd Also see acknowledgements in Readme.html This program is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details. You should have received a copy of the GNU Lesser General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA, or go to http://www.gnu.org/copyleft/lesser.txt. You may alternatively use this source under the terms of a specific version of the OGRE Unrestricted License provided you have obtained such a license from Torus Knot Software Ltd. ----------------------------------------------------------------------------- */ #ifndef __RenderQueueSortingGrouping_H__ #define __RenderQueueSortingGrouping_H__ // Precompiler options #include "OgrePrerequisites.h" #include "OgreIteratorWrappers.h" #include "OgreMaterial.h" #include "OgreTechnique.h" #include "OgrePass.h" #include "OgreRadixSort.h" namespace Ogre { /** Struct associating a single Pass with a single Renderable. This is used to for objects sorted by depth and thus not grouped by pass. */ struct RenderablePass { /// Pointer to the Renderable details Renderable* renderable; /// Pointer to the Pass Pass* pass; RenderablePass(Renderable* rend, Pass* p) :renderable(rend), pass(p) {} }; /** Visitor interface for items in a QueuedRenderableCollection. @remarks Those wishing to iterate over the items in a QueuedRenderableCollection should implement this visitor pattern, since internal organisation of the collection depends on the sorting method in use. */ class _OgreExport QueuedRenderableVisitor { public: QueuedRenderableVisitor() {} virtual ~QueuedRenderableVisitor() {} /** Called when visiting a RenderablePass, ie items in a sorted collection where items are not grouped by pass. @remarks If this is called, neither of the other 2 visit methods will be called. */ virtual void visit(const RenderablePass* rp) = 0; /* When visiting a collection grouped by pass, this is called when the grouping pass changes. @remarks If this method is called, the RenderablePass visit method will not be called for this collection. The Renderable visit method will be called for each item underneath the pass grouping level. @returns True to continue, false to skip the Renderables underneath */ virtual bool visit(const Pass* p) = 0; /** Visit method called once per Renderable on a grouped collection. @remarks If this method is called, the RenderablePass visit method will not be called for this collection. */ virtual void visit(const Renderable* r) = 0; }; /** Lowest level collection of renderables. @remarks To iterate over items in this collection, you must call the accept method and supply a QueuedRenderableVisitor. The order of the iteration, and whether that iteration is over a RenderablePass list or a 2-level grouped list which causes a visit call at the Pass level, and a call for each Renderable underneath. */ class _OgreExport QueuedRenderableCollection { public: /** Organisation modes required for this collection. @remarks This affects the internal placement of the items added to this collection; if only one type of sorting / grouping is to be required, then renderables can be stored only once, whilst if multiple types are going to be needed then internally there will be multiple organisations. Changing the organisation needs to be done when the collection is empty. */ enum OrganisationMode { /// Group by pass OM_PASS_GROUP = 1, /// Sort descending camera distance OM_SORT_DESCENDING = 2, /** Sort ascending camera distance Note value overlaps with descending since both use same sort */ OM_SORT_ASCENDING = 6 }; protected: /// Comparator to order pass groups struct PassGroupLess { bool _OgreExport operator()(const Pass* a, const Pass* b) const { // Sort by passHash, which is pass, then texture unit changes uint32 hasha = a->getHash(); uint32 hashb = b->getHash(); if (hasha == hashb) { // Must differentTransparentQueueItemLessiate by pointer incase 2 passes end up with the same hash return a < b; } else { return hasha < hashb; } } }; /// Comparator to order objects by descending camera distance struct DepthSortDescendingLess { const Camera* camera; DepthSortDescendingLess(const Camera* cam) : camera(cam) { } bool _OgreExport operator()(const RenderablePass& a, const RenderablePass& b) const { if (a.renderable == b.renderable) { // Same renderable, sort by pass hash return a.pass->getHash() < b.pass->getHash(); } else { // Different renderables, sort by depth Real adepth = a.renderable->getSquaredViewDepth(camera); Real bdepth = b.renderable->getSquaredViewDepth(camera); if (Math::RealEqual(adepth, bdepth)) { // Must return deterministic result, doesn't matter what return a.pass < b.pass; } else { // Sort DESCENDING by depth (ie far objects first) return (adepth > bdepth); } } } }; /** Vector of RenderablePass objects, this is built on the assumption that vectors only ever increase in size, so even if we do clear() the memory stays allocated, ie fast */ typedef std::vector
RenderablePassList; typedef std::vector
RenderableList; /** Map of pass to renderable lists, this is a grouping by pass. */ typedef std::map
PassGroupRenderableMap; /// Functor for accessing sort value 1 for radix sort (Pass) struct RadixSortFunctorPass { uint32 operator()(const RenderablePass& p) const { return p.pass->getHash(); } }; /// Radix sorter for accessing sort value 1 (Pass) static RadixSort
msRadixSorter1; /// Functor for descending sort value 2 for radix sort (distance) struct RadixSortFunctorDistance { const Camera* camera; RadixSortFunctorDistance(const Camera* cam) : camera(cam) { } float operator()(const RenderablePass& p) const { // Sort DESCENDING by depth (ie far objects first), use negative distance // here because radix sorter always dealing with accessing sort return static_cast
(- p.renderable->getSquaredViewDepth(camera)); } }; /// Radix sorter for sort value 2 (distance) static RadixSort
msRadixSorter2; /// Bitmask of the organisation modes requested uint8 mOrganisationMode; /// Grouped PassGroupRenderableMap mGrouped; /// Sorted descending (can iterate backwards to get ascending) RenderablePassList mSortedDescending; /// Internal visitor implementation void acceptVisitorGrouped(QueuedRenderableVisitor* visitor) const; /// Internal visitor implementation void acceptVisitorDescending(QueuedRenderableVisitor* visitor) const; /// Internal visitor implementation void acceptVisitorAscending(QueuedRenderableVisitor* visitor) const; public: QueuedRenderableCollection(); ~QueuedRenderableCollection(); /// Empty the collection void clear(void); /** Remove the group entry (if any) for a given Pass. @remarks To be used when a pass is destroyed, such that any grouping level for it becomes useless. */ void removePassGroup(Pass* p); /** Reset the organisation modes required for this collection. @remarks You can only do this when the collection is empty. @see OrganisationMode */ void resetOrganisationModes(void) { mOrganisationMode = 0; } /** Add a required sorting / grouping mode to this collection when next used. @remarks You can only do this when the collection is empty. @see OrganisationMode */ void addOrganisationMode(OrganisationMode om) { mOrganisationMode |= om; } /// Add a renderable to the collection using a given pass void addRenderable(Pass* pass, Renderable* rend); /** Perform any sorting that is required on this collection. @param cam The camera */ void sort(const Camera* cam); /** Accept a visitor over the collection contents. @param visitor Visitor class which should be called back @param om The organisation mode which you want to iterate over. Note that this must have been included in an addOrganisationMode call before any renderables were added. */ void acceptVisitor(QueuedRenderableVisitor* visitor, OrganisationMode om) const; }; /** Collection of renderables by priority. @remarks This class simply groups renderables for rendering. All the renderables contained in this class are destined for the same RenderQueueGroup (coarse groupings like those between the main scene and overlays) and have the same priority (fine groupings for detailed overlap control). @par This class can order solid renderables by a number of criteria; it can optimise them into groups based on pass to reduce render state changes, or can sort them by ascending or descending view depth. Transparent objects are always ordered by descending depth. @par To iterate over items in the collections held by this object you should retrieve the collection in use (e.g. solids, solids with no shadows, transparents) and use the accept() method, providing a class implementing QueuedRenderableVisitor. */ class _OgreExport RenderPriorityGroup { protected: /// Parent queue group RenderQueueGroup* mParent; bool mSplitPassesByLightingType; bool mSplitNoShadowPasses; bool mShadowCastersNotReceivers; /// Solid pass list, used when no shadows, modulative shadows, or ambient passes for additive QueuedRenderableCollection mSolidsBasic; /// Solid per-light pass list, used with additive shadows QueuedRenderableCollection mSolidsDiffuseSpecular; /// Solid decal (texture) pass list, used with additive shadows QueuedRenderableCollection mSolidsDecal; /// Solid pass list, used when shadows are enabled but shadow receive is turned off for these passes QueuedRenderableCollection mSolidsNoShadowReceive; /// Transparent list QueuedRenderableCollection mTransparents; /// remove a pass entry from all collections void removePassEntry(Pass* p); /// Internal method for adding a solid renderable void addSolidRenderable(Technique* pTech, Renderable* rend, bool toNoShadowMap); /// Internal method for adding a solid renderable void addSolidRenderableSplitByLightType(Technique* pTech, Renderable* rend); /// Internal method for adding a transparent renderable void addTransparentRenderable(Technique* pTech, Renderable* rend); public: RenderPriorityGroup(RenderQueueGroup* parent, bool splitPassesByLightingType, bool splitNoShadowPasses, bool shadowCastersNotReceivers); ~RenderPriorityGroup() { } /** Get the collection of basic solids currently queued, this includes all solids when there are no shadows, or all solids which have shadow receiving enabled when using modulative shadows, or all ambient passes of solids which have shadow receive enabled for additive shadows. */ const QueuedRenderableCollection& getSolidsBasic(void) const { return mSolidsBasic; } /** Get the collection of solids currently queued per light (only applicable in additive shadow modes). */ const QueuedRenderableCollection& getSolidsDiffuseSpecular(void) const { return mSolidsDiffuseSpecular; } /** Get the collection of solids currently queued for decal passes (only applicable in additive shadow modes). */ const QueuedRenderableCollection& getSolidsDecal(void) const { return mSolidsDecal; } /** Get the collection of solids for which shadow receipt is disabled (only applicable when shadows are enabled). */ const QueuedRenderableCollection& getSolidsNoShadowReceive(void) const { return mSolidsNoShadowReceive; } /** Get the collection of transparent objects currently queued */ const QueuedRenderableCollection& getTransparents(void) const { return mTransparents; } /** Reset the organisation modes required for the solids in this group. @remarks You can only do this when the group is empty, ie after clearing the queue. @see QueuedRenderableCollection::OrganisationMode */ void resetOrganisationModes(void); /** Add a required sorting / grouping mode for the solids in this group. @remarks You can only do this when the group is empty, ie after clearing the queue. @see QueuedRenderableCollection::OrganisationMode */ void addOrganisationMode(QueuedRenderableCollection::OrganisationMode om); /** Set the sorting / grouping mode for the solids in this group to the default. @remarks You can only do this when the group is empty, ie after clearing the queue. @see QueuedRenderableCollection::OrganisationMode */ void defaultOrganisationMode(void); /** Add a renderable to this group. */ void addRenderable(Renderable* pRend, Technique* pTech); /** Sorts the objects which have been added to the queue; transparent objects by their depth in relation to the passed in Camera. */ void sort(const Camera* cam); /** Clears this group of renderables. */ void clear(void); /** Sets whether or not the queue will split passes by their lighting type, ie ambient, per-light and decal. */ void setSplitPassesByLightingType(bool split) { mSplitPassesByLightingType = split; } /** Sets whether or not passes which have shadow receive disabled should be separated. */ void setSplitNoShadowPasses(bool split) { mSplitNoShadowPasses = split; } /** Sets whether or not objects which cast shadows should be treated as never receiving shadows. */ void setShadowCastersCannotBeReceivers(bool ind) { mShadowCastersNotReceivers = ind; } }; /** A grouping level underneath RenderQueue which groups renderables to be issued at coarsely the same time to the renderer. @remarks Each instance of this class itself hold RenderPriorityGroup instances, which are the groupings of renderables by priority for fine control of ordering (not required for most instances). */ class _OgreExport RenderQueueGroup { public: typedef std::map
> PriorityMap; typedef MapIterator
PriorityMapIterator; protected: RenderQueue* mParent; bool mSplitPassesByLightingType; bool mSplitNoShadowPasses; bool mShadowCastersNotReceivers; /// Map of RenderPriorityGroup objects PriorityMap mPriorityGroups; /// Whether shadows are enabled for this queue bool mShadowsEnabled; public: RenderQueueGroup(RenderQueue* parent, bool splitPassesByLightingType, bool splitNoShadowPasses, bool shadowCastersNotReceivers) : mParent(parent) , mSplitPassesByLightingType(splitPassesByLightingType) , mSplitNoShadowPasses(splitNoShadowPasses) , mShadowCastersNotReceivers(shadowCastersNotReceivers) , mShadowsEnabled(true) { } ~RenderQueueGroup() { // destroy contents now PriorityMap::iterator i; for (i = mPriorityGroups.begin(); i != mPriorityGroups.end(); ++i) { delete i->second; } } /** Get an iterator for browsing through child contents. */ PriorityMapIterator getIterator(void) { return PriorityMapIterator(mPriorityGroups.begin(), mPriorityGroups.end()); } /** Add a renderable to this group, with the given priority. */ void addRenderable(Renderable* pRend, Technique* pTech, ushort priority) { // Check if priority group is there PriorityMap::iterator i = mPriorityGroups.find(priority); RenderPriorityGroup* pPriorityGrp; if (i == mPriorityGroups.end()) { // Missing, create pPriorityGrp = new RenderPriorityGroup(this, mSplitPassesByLightingType, mSplitNoShadowPasses, mShadowCastersNotReceivers); mPriorityGroups.insert(PriorityMap::value_type(priority, pPriorityGrp)); } else { pPriorityGrp = i->second; } // Add pPriorityGrp->addRenderable(pRend, pTech); } /** Clears this group of renderables. @param destroy If false, doesn't delete any priority groups, just empties them. Saves on memory deallocations since the chances are rougly the same kinds of renderables are going to be sent to the queue again next time. If true, completely destroys. */ void clear(bool destroy = false) { PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { if (destroy) delete i->second; else i->second->clear(); } if (destroy) mPriorityGroups.clear(); } /** Indicate whether a given queue group will be doing any shadow setup. @remarks This method allows you to inform the queue about a queue group, and to indicate whether this group will require shadow processing of any sort. In order to preserve rendering order, OGRE has to treat queue groups as very separate elements of the scene, and this can result in it having to duplicate shadow setup for each group. Therefore, if you know that a group which you are using will never need shadows, you should preregister the group using this method in order to improve the performance. */ void setShadowsEnabled(bool enabled) { mShadowsEnabled = enabled; } /** Are shadows enabled for this queue? */ bool getShadowsEnabled(void) const { return mShadowsEnabled; } /** Sets whether or not the queue will split passes by their lighting type, ie ambient, per-light and decal. */ void setSplitPassesByLightingType(bool split) { mSplitPassesByLightingType = split; PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { i->second->setSplitPassesByLightingType(split); } } /** Sets whether or not the queue will split passes which have shadow receive turned off (in their parent material), which is needed when certain shadow techniques are used. */ void setSplitNoShadowPasses(bool split) { mSplitNoShadowPasses = split; PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { i->second->setSplitNoShadowPasses(split); } } /** Sets whether or not objects which cast shadows should be treated as never receiving shadows. */ void setShadowCastersCannotBeReceivers(bool ind) { mShadowCastersNotReceivers = ind; PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { i->second->setShadowCastersCannotBeReceivers(ind); } } /** Reset the organisation modes required for the solids in this group. @remarks You can only do this when the group is empty, ie after clearing the queue. @see QueuedRenderableCollection::OrganisationMode */ void resetOrganisationModes(void) { PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { i->second->resetOrganisationModes(); } } /** Add a required sorting / grouping mode for the solids in this group. @remarks You can only do this when the group is empty, ie after clearing the queue. @see QueuedRenderableCollection::OrganisationMode */ void addOrganisationMode(QueuedRenderableCollection::OrganisationMode om) { PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { i->second->addOrganisationMode(om); } } /** Setthe sorting / grouping mode for the solids in this group to the default. @remarks You can only do this when the group is empty, ie after clearing the queue. @see QueuedRenderableCollection::OrganisationMode */ void defaultOrganisationMode(void) { PriorityMap::iterator i, iend; iend = mPriorityGroups.end(); for (i = mPriorityGroups.begin(); i != iend; ++i) { i->second->defaultOrganisationMode(); } } }; } #endif
OgreRenderQueueSortingGrouping.h
Dirección de la página
Dirección del archivo
Anterior
149/217
Siguiente
Descargar
( 22 KB )
Comments
Total ratings:
0
Average rating:
No clasificado
of 10
Would you like to comment?
Join now
, or
Logon
if you are already a member.