Ogre:Hardwarebuffer

简介:
  1.      
  2. Ogre中的硬件缓存是指在显卡上的存储,这和在内存上的存储一样是可以访问的。有三种硬件缓存:HardwareVertexBuffer(顶点缓存,存储顶点的各种数据)、HardwareIndexBuffer(索引缓存,存储一个mesh的面片的顶点索引),HardwarePixelBuffer(纹理缓存,存储某个纹理贴图的数据)。这些数据在程序运行时都在显卡的存储上,然而你可以去读和写这些数据,来操控程序中物体的形状、纹理等。这个用处是非常大的。在Ogre中与访问这些硬件缓存有关的类及他们相互间的关系如下图:
  1. hardwarebuffer 
  2. 根据这个图进行解释
  3. 1、最上面的hardwarevertexbuffer
  4. 读写:如果mesh使用的所有子mesh共享buffer的形式,则用mesh的sharedvertexdata,否则用submesh的vertexdata来得到vertexdata结构,vertexdata封装了对该mesh的顶点缓存数据的访问方式,但是却不直接包含这些顶点缓存数据。vertexdata中的vettexbufferbinding可以知道当前的vertexdata对应了确切的硬件上的哪块buffer,可以通过vettexbufferbinding的getBuffer确切的得到该顶点缓存,而vertexdata中的vertexdeclaration则是一个对他对应的buffer进行各种访问的接口,里面有访问的格式等。如果要开始操纵这个buffer,需要将getbuffer得到的hardwarevertexbuffer调用lock,然后将这片缓存上锁,这个lock返回了一个void指针,指向的就是缓存数据。拿着这个指针就可以读取改写等
  5. 创建:使用hardwarebuffermanager的create来创建,创建后利用hardwarevertexbuffer的write写入数据
  6. 2.中间的hardwareindexbuffer
  7. 读写:直接使用submesh的indexdata来得到一个indexdata结构,再调用它的hardwareindexbuffer的来得到这个顶点缓存,童年顶点缓存一样再调用lock来进行读写操作

创建:同顶点缓存

3最下面的hardwarepixelbuffer

读写:从texture中可以直接得到这个hardwarepixelbuffer,然后对它lock后就可以得到一个pixelbox的数据,pixebox封装了所有纹理数据及其各种属性信息

创建:texture是由texturemanager创建的

下面是一些具体的使用硬件缓存的例子

读取顶点和索引缓存

    Ogre::MeshPtr meshPtr=mainEntity->getMesh();
   //假设这里使用的是share的形式
    Ogre::VertexData* vertex_data=meshPtr->sharedVertexData;

   //得到位置数据的信息
    const Ogre::VertexElement* posElem =vertex_data->vertexDeclaration->findElementBySemantic(Ogre::VES_POSITION);

//得到纹理坐标数据的信息
    const Ogre::VertexElement* texcoElem=vertex_data->vertexDeclaration->findElementBySemantic(Ogre::VES_TEXTURE_COORDINATES );
    //得到位置和纹理的缓存
    Ogre::HardwareVertexBufferSharedPtr posBuf =vertex_data->vertexBufferBinding->getBuffer(posElem->getSource());
    Ogre::HardwareVertexBufferSharedPtr texcoBuf =vertex_data->vertexBufferBinding->getBuffer(texcoElem->getSource());
     //顶点位置缓存的lock,读取
    unsigned char* vertexPos =static_cast<unsigned char*>(posBuf->lock(Ogre::HardwareBuffer::HBL_READ_ONLY));

//将第一个点的位置读出
    float* pReal;

//这个函数的作用是将当前vertexPos指向的数据用其他型(这里是float*)的指针指向,这样读出来的数据就是float型的了,或者用float型的数据进行写入
    posElem->baseVertexPointerToElement(vertexPos, &pReal);
   Ogre::Vector3 pt(pReal[0], pReal[1], pReal[2]);
//访问之后要上锁
    posBuf->unlock();
   
    //读取索引信息
          Ogre::SubMesh* submesh = meshPtr->getSubMesh( i );

     //得到这个submesh的indexdata
        Ogre::IndexData* index_data = submesh->indexData;
        int numTris = index_data->indexCount / 3;

//得到indexbuffer
        Ogre::HardwareIndexBufferSharedPtr ibuf = index_data->indexBuffer;
        bool use32bitindexes = (ibuf->getType() == Ogre::HardwareIndexBuffer::IT_32BIT);

//得到具体的索引缓存数据
        unsigned long*  pLong = static_cast<unsigned long*>(ibuf->lock(Ogre::HardwareBuffer::HBL_READ_ONLY));
        unsigned short* pShort = reinterpret_cast<unsigned short*>(pLong);

          … …
          ibuf->unlock();

访问纹理缓存

    Ogre::HardwarePixelBufferSharedPtr crossPixbufferPtr=texture.getPointer()->getBuffer(0,0);       
    crossPixbufferPtr->lock(Ogre::HardwareBuffer::HBL_NORMAL);
    Ogre::PixelBox pb=crossPixbufferPtr->getCurrentLock();
    int height = pb.getHeight();
    int width = pb.getWidth();
    int pitch = pb.rowPitch; // Skip between rows of image   
    uint32* data=static_cast<uint32*>(pb.data);

     ……操纵data……
    crossPixbufferPtr->unlock();

创建顶点缓存和索引缓存,进而根据其创建一个自定义的mesh

void createColourCube()
    {
    /// Create the mesh via the MeshManager
    Ogre::MeshPtr msh = MeshManager::getSingleton().createManual("ColourCube", "General");
    /// Create one submesh
    SubMesh* sub = msh->createSubMesh();
    const float sqrt13 = 0.577350269f; /* sqrt(1/3) */
    /// Define the vertices (8 vertices, each consisting of 2 groups of 3 floats
    const size_t nVertices = 8;
    const size_t vbufCount = 3*2*nVertices;
    float vertices[vbufCount] = {
            -100.0,100.0,-100.0,        //0 position
            -sqrt13,sqrt13,-sqrt13,     //0 normal
            100.0,100.0,-100.0,         //1 position
            sqrt13,sqrt13,-sqrt13,      //1 normal
            100.0,-100.0,-100.0,        //2 position
            sqrt13,-sqrt13,-sqrt13,     //2 normal
            -100.0,-100.0,-100.0,       //3 position
            -sqrt13,-sqrt13,-sqrt13,    //3 normal
            -100.0,100.0,100.0,         //4 position
            -sqrt13,sqrt13,sqrt13,      //4 normal
            100.0,100.0,100.0,          //5 position
            sqrt13,sqrt13,sqrt13,       //5 normal
            100.0,-100.0,100.0,         //6 position
            sqrt13,-sqrt13,sqrt13,      //6 normal
            -100.0,-100.0,100.0,        //7 position
            -sqrt13,-sqrt13,sqrt13,     //7 normal
    };
    RenderSystem* rs = Root::getSingleton().getRenderSystem();
    RGBA colours[nVertices];
    RGBA *pColour = colours;
    // Use render system to convert colour value since colour packing varies
    rs->convertColourValue(ColourValue(1.0,0.0,0.0), pColour++); //0 colour
    rs->convertColourValue(ColourValue(1.0,1.0,0.0), pColour++); //1 colour
    rs->convertColourValue(ColourValue(0.0,1.0,0.0), pColour++); //2 colour
    rs->convertColourValue(ColourValue(0.0,0.0,0.0), pColour++); //3 colour
    rs->convertColourValue(ColourValue(1.0,0.0,1.0), pColour++); //4 colour
    rs->convertColourValue(ColourValue(1.0,1.0,1.0), pColour++); //5 colour
    rs->convertColourValue(ColourValue(0.0,1.0,1.0), pColour++); //6 colour
    rs->convertColourValue(ColourValue(0.0,0.0,1.0), pColour++); //7 colour
    /// Define 12 triangles (two triangles per cube face)
    /// The values in this table refer to vertices in the above table
    const size_t ibufCount = 36;
    unsigned short faces[ibufCount] = {
            0,2,3,
            0,1,2,
            1,6,2,
            1,5,6,
            4,6,5,
            4,7,6,
            0,7,4,
            0,3,7,
            0,5,1,
            0,4,5,
            2,7,3,
            2,6,7
    };
    /// Create vertex data structure for 8 vertices shared between submeshes
    msh->sharedVertexData = new VertexData();
    msh->sharedVertexData->vertexCount = nVertices;
    /// Create declaration (memory format) of vertex data
    VertexDeclaration* decl = msh->sharedVertexData->vertexDeclaration;
    size_t offset = 0;
    // 1st buffer
    decl->addElement(0, offset, VET_FLOAT3, VES_POSITION);
    offset += VertexElement::getTypeSize(VET_FLOAT3);
    decl->addElement(0, offset, VET_FLOAT3, VES_NORMAL);
    offset += VertexElement::getTypeSize(VET_FLOAT3);
    /// Allocate vertex buffer of the requested number of vertices (vertexCount)
    /// and bytes per vertex (offset)
    HardwareVertexBufferSharedPtr vbuf =
        HardwareBufferManager::getSingleton().createVertexBuffer(
        offset, msh->sharedVertexData->vertexCount, HardwareBuffer::HBU_STATIC_WRITE_ONLY);
    /// Upload the vertex data to the card
    vbuf->writeData(0, vbuf->getSizeInBytes(), vertices, true);
    /// Set vertex buffer binding so buffer 0 is bound to our vertex buffer
    VertexBufferBinding* bind = msh->sharedVertexData->vertexBufferBinding;
    bind->setBinding(0, vbuf);
    // 2nd buffer
    offset = 0;
    decl->addElement(1, offset, VET_COLOUR, VES_DIFFUSE);
    offset += VertexElement::getTypeSize(VET_COLOUR);
    /// Allocate vertex buffer of the requested number of vertices (vertexCount)
    /// and bytes per vertex (offset)
    vbuf = HardwareBufferManager::getSingleton().createVertexBuffer(
        offset, msh->sharedVertexData->vertexCount, HardwareBuffer::HBU_STATIC_WRITE_ONLY);
    /// Upload the vertex data to the card
    vbuf->writeData(0, vbuf->getSizeInBytes(), colours, true);
    /// Set vertex buffer binding so buffer 1 is bound to our colour buffer
    bind->setBinding(1, vbuf);
    /// Allocate index buffer of the requested number of vertices (ibufCount)
    HardwareIndexBufferSharedPtr ibuf = HardwareBufferManager::getSingleton().
        createIndexBuffer(
        HardwareIndexBuffer::IT_16BIT,
        ibufCount,
        HardwareBuffer::HBU_STATIC_WRITE_ONLY);
    /// Upload the index data to the card
    ibuf->writeData(0, ibuf->getSizeInBytes(), faces, true);
    /// Set parameters of the submesh
    sub->useSharedVertices = true;
    sub->indexData->indexBuffer = ibuf;
    sub->indexData->indexCount = ibufCount;
    sub->indexData->indexStart = 0;
    /// Set bounding information (for culling)
    msh->_setBounds(AxisAlignedBox(-100,-100,-100,100,100,100));
    msh->_setBoundingSphereRadius(Math::Sqrt(3*100*100));
    /// Notify -Mesh object that it has been loaded
    msh->load();
    }

然后可以从mesh直接创建entity放在场景中

Entity*thisEntity = sceneManager->createEntity("cc", "ColourCube");

目录
相关文章
|
8月前
|
数据可视化 C++
msvc编译opencascade和vtk
1.opencascade源码 我下载的时候最新版本是7.7.0
108 0
|
异构计算
好工具推荐系列:OpenGL Extensions Viewer
好工具推荐系列:OpenGL Extensions Viewer
174 0
|
XML 算法 数据挖掘
|
C++
OGRE学习笔记之一
非常之简单,就是把mRoot->showConfigDialog()修改成了mRoot->restoreConfig(),直接加载配置文件.啊,你应该已经发现了我的修改以后的代码和之前的代码不一样,嗯,是的.因为这个是在cpp里面的实现.
1791 0