OpenGL ES：一个 VBO - 几个精灵 - 每个单独翻译

Posted 2023-02-22

【中文标题】OpenGL ES：一个 VBO - 几个精灵 - 每个单独翻译

【英文标题】：OpenGL ES: one VBO - several sprites - translate each individually

【发布时间】：2017-05-25 12:06:17

【问题描述】：

我想知道当这些精灵的顶点存储在同一个 VBO 中时，如何单独访问和操作每个精灵。出于性能原因，我为我的精灵使用纹理图集，然后将它们映射到顶点数组。我尝试将精灵拆分为单独的顶点数组并将它们放在单独的 VBO 中。但是，如果我绘制 100 个精灵，这会影响性能，因为每次调用都必须进行很多单独的绘制。相反，我想使用一个相同的 VBO，然后分别翻译每个精灵。这可能吗？

源代码

     @Override
public void onSurfaceCreated(GL10 unused, EGLConfig config) 

    vertexBuffer = GLData.createVertices(nSprites, vertices);
    GLData.createUvsData(nSprites, alienUvs, uvBufferAlien);
    //drawListBuffer = GLData.createIndices(indices, nSprites);
    GLData.createVertexBufferObject(vertexBuffer,nSprites, uvBufferAlien, bufferId.length, bufferId);

    createCamera();

    GLES20.glEnable(GLES20.GL_CULL_FACE);

    GLES20.glEnable(GLES20.GL_BLEND);
    GLES20.glBlendFunc(GLES20.GL_ONE, GLES20.GL_ONE_MINUS_SRC_ALPHA);

    // get shadersource
    final String vertexShader = getVertexShader();
    final String fragmentShader = getFragmentShader();

    final int vertexShaderHandle = ShaderHelper.compileShader(GLES20.GL_VERTEX_SHADER, vertexShader);
    final int fragmentShaderHandle = ShaderHelper.compileShader(GLES20.GL_FRAGMENT_SHADER, fragmentShader);

    mProgramHandle = ShaderHelper.createAndLinkProgram(vertexShaderHandle, fragmentShaderHandle,
            new String[] "a_Position",  "a_Color", "a_Normal", "a_TexCoordinate");

    //create texture
    textureHandle = TextureHelper.loadTexture(context);

    GLES20.glUseProgram(mProgramHandle);

    mMVPMatrixHandle = GLES20.glGetUniformLocation(mProgramHandle, "u_MVPMatrix");
    mMVMatrixHandle = GLES20.glGetUniformLocation(mProgramHandle, "u_MVMatrix");
    //mColorHandle = GLES20.glGetAttribLocation(mProgramHandle, "a_Color");
    mTextureCoordinateHandle = GLES20.glGetAttribLocation(mProgramHandle, "a_TexCoordinate");

    mPositionHandle = GLES20.glGetAttribLocation(mProgramHandle, "a_Position");



@Override
public void onSurfaceChanged(GL10 unused, int width, int height) 

    game_width = width;
    game_height = height;

    GLES20.glViewport(0, 0, width, height);


    // Create a new perspective projection matrix. The height will stay the same
    // while the width will vary as per aspect ratio.
    final float ratio = (float) width / height;
    final float left = -ratio;
    final float right = ratio;
    final float bottom = -1.0f;
    final float top = 1.0f;
    final float near = 1f;
    final float far = 20.0f;

    Matrix.frustumM(mProjectionMatrix, 0, left, right, bottom, top, near, far);




@Override
public void onDrawFrame(GL10 unused) 

    Matrix.setIdentityM(mModelMatrix, 0);
    Matrix.translateM(mModelMatrix, 0, 2, 0f, -7f);

    draw();



private void draw() 

    //uvs
    GLES20.glBindBuffer(GLES20.GL_ARRAY_BUFFER, bufferId[0]);
    GLES20.glEnableVertexAttribArray(mTextureCoordinateHandle);
    GLES20.glVertexAttribPointer(mTextureCoordinateHandle, mTextureCoordinateDataSize, GLES20.GL_FLOAT, false, 0, 0);


    //vertices
    GLES20.glBindBuffer(GLES20.GL_ARRAY_BUFFER, bufferId[1]);
    GLES20.glEnableVertexAttribArray(mPositionHandle);
    GLES20.glVertexAttribPointer(mPositionHandle, mPositionDataSize, GLES20.GL_FLOAT, false, 0, 0);


    // Clear the currently bound buffer (so future OpenGL calls do not use this buffer).
    GLES20.glBindBuffer(GLES20.GL_ARRAY_BUFFER, 0);



                         /* MATRIX */

    // This multiplies the view matrix by the model matrix, and stores the result in the MVP matrix
    // (which currently contains model * view).
    Matrix.multiplyMM(mMVPMatrix, 0, mViewMatrix, 0, mModelMatrix, 0);

    // Pass in the modelview matrix.
    GLES20.glUniformMatrix4fv(mMVMatrixHandle, 1, false, mMVPMatrix, 0);

    // This multiplies the modelview matrix by the projection matrix, and stores the result in the MVP matrix
    // (which now contains model * view * projection).
    Matrix.multiplyMM(mMVPMatrix, 0, mProjectionMatrix, 0, mMVPMatrix, 0);

    // Pass in the combined matrix.
    GLES20.glUniformMatrix4fv(mMVPMatrixHandle, 1, false, mMVPMatrix, 0);

    // Draw the sprites

    GLES20.glDrawArrays(GLES20.GL_TRIANGLES, 0, 6 * nSprites);

顶点着色器

 attribute vec4 a_Position;
 uniform mat4 u_MVMatrix;
 uniform mat4 u_MVPMatrix;
 void main()       

    gl_Position = u_MVPMatrix * a_Position;

 

【参考方案1】：

您可以上传一个统一的矩阵数组，并将索引作为每个精灵的每个顶点属性，为每个精灵提供一个唯一的矩阵（用于蒙皮骨骼动画模型的技术，但也适用于此）。

但是，对于简单的 sprite，上传矩阵的成本几乎与更新位置数组一样昂贵，而且您需要担心统一数组的大小有限制。您最好只在软件中为它们设置动画，并为每帧批次中的每个顶点上传一个新位置。将位置拆分为一个单独的缓冲区，例如您的纹理坐标，以最小化您上传每帧的带宽。

【讨论】：

我为好的答案投了赞成票 - 但问题没有解决。事实上 - 不敢相信 - 当我在 Galaxy S3（从 S7 切换）上使用我的应用程序时，帧时间下降到 20。我整晚都在考虑这个问题，但它可能是 1920 X 1856 的纹理?是因为它不是 POT 还是它太大了？？？我会试试这个 - 如果它不起作用，我会在堆栈上提出一个新问题并给你链接。谢谢！！！！！！！

您是否正在修改纹理图集或现有缓冲区而不使用流水线？您可能只是触发了大量资源副本。见community.arm.com/graphics/b/blog/posts/…