我正在使用this lovely tutorial.进行阴影投射。过程是,将场景渲染到帧缓冲区,该缓冲区附加有用于保存深度值的立方体贴图。然后,我们将此立方体贴图传递到片段着色器,对它进行采样并从那里获取深度值。

我与本教程略有不同,因为我没有使用一次几何着色器一次渲染整个立方体贴图,而是渲染了六次场景以获得相同的效果-主要是因为我当前的着色器系统不支持几何着色器,并且现在,我不太担心性能下降。

深度立方体贴图已绘制完成,这是gDEBugger的屏幕截图:

c++ - 尝试采样立方体贴图纹理时为GL_INVALID_OPERATION-LMLPHP

一切似乎都井井有条。

但是,当我尝试对该立方体贴图进行采样时,片段着色器中出现问题。调用glDrawArrays之后,调用glGetError返回GL_INVALID_OPERATION,据我所知,它来自此处:(违规行已被注释)

struct PointLight
{
  vec3 Position;
  float ConstantRolloff;
  float LinearRolloff;
  float QuadraticRolloff;
  vec4 Color;
  samplerCube DepthMap;
  float FarPlane;
};
uniform PointLight PointLights[NUM_POINT_LIGHTS];

[...]

float CalculateShadow(int lindex)
{
  // Calculate vector between fragment and light
  vec3 fragToLight = FragPos - PointLights[lindex].Position;

  // Sample from the depth map (Comment this out and everything works fine!)
  float closestDepth = texture(PointLights[lindex].DepthMap, vec3(1.0, 1.0, 1.0)).r;

  // Transform to original value
  closestDepth *= PointLights[lindex].FarPlane;

  // Get current depth
  float currDepth = length(fragToLight);

  // Test for shadow
  float bias = 0.05;
  float shadow = currDepth - bias > closestDepth ? 1.0 : 0.0;
  return shadow;
}


注释掉上述行似乎可以使所有工作正常-因此,我假设这是导致问题的对纹理采样器的调用。 I saw that this can be attributed to using two textures of different types in the same texture unit-但根据gDEBugger,情况并非如此:

c++ - 尝试采样立方体贴图纹理时为GL_INVALID_OPERATION-LMLPHP

纹理16是深度立方体贴图。

如果相关的话,这是我设置FBO的方法:(仅调用一次)

// Generate frame buffer
glGenFramebuffers(1, &depthMapFrameBuffer);

// Generate depth maps
glGenTextures(1, &depthMap);

// Set up textures
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_CUBE_MAP, depthMap);
for (int i = 0; i < 6; ++i)
  glTexImage2D(GL_TEXTURE_CUBE_MAP_POSITIVE_X + i, 0, GL_DEPTH_COMPONENT,
    ShadowmapSize, ShadowmapSize, 0, GL_DEPTH_COMPONENT, GL_FLOAT, NULL);

// Set texture parameters
glTexParameteri(GL_TEXTURE_CUBE_MAP, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_CUBE_MAP, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_CUBE_MAP, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_CUBE_MAP, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_CUBE_MAP, GL_TEXTURE_WRAP_R, GL_CLAMP_TO_EDGE);

// Attach cubemap to FBO
glBindFramebuffer(GL_FRAMEBUFFER, depthMapFrameBuffer);
glFramebufferTexture(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT, depthMap, 0);
glDrawBuffer(GL_NONE);
glReadBuffer(GL_NONE);
if(glCheckFramebufferStatus(GL_FRAMEBUFFER) != GL_FRAMEBUFFER_COMPLETE)
  ERROR_LOG("PointLight created an incomplete frame buffer!\n");
glBindTexture(GL_TEXTURE_CUBE_MAP, 0);
glBindFramebuffer(GL_FRAMEBUFFER, 0);


这是我的绘画方式:(称为每一帧)

// Set up viewport
glViewport(0, 0, ShadowmapSize, ShadowmapSize);

// Bind frame buffer
glBindFramebuffer(GL_FRAMEBUFFER, depthMapFrameBuffer);

// Clear depth buffer
glClear(GL_DEPTH_BUFFER_BIT);

// Render scene
for(int i = 0; i < 6; ++i)
{
  sh->SetUniform("ShadowMatrix", lightSpaceTransforms[i]);
  glFramebufferTexture2D(GL_FRAMEBUFFER, GL_DEPTH_ATTACHMENT,
    GL_TEXTURE_CUBE_MAP_POSITIVE_X + i, depthMap, 0);
  Space()->Get<Renderer>()->RenderScene(sh);
}

// Unbind frame buffer
glBindFramebuffer(GL_FRAMEBUFFER, 0);


这是我在绘制之前绑定它的方式:

std::stringstream ssD;
ssD << "PointLights[" << i << "].DepthMap";
glActiveTexture(GL_TEXTURE4 + i);
glBindTexture(GL_TEXTURE_CUBE_MAP, pointlights[i]->DepthMap()); // just returns the ID of the light's depth map
shader->SetUniform(ssD.str().c_str(), i + 4); // just a wrapper around glSetUniform1i


感谢您的阅读,如果可以提供更多信息,请告诉我!

最佳答案

这是旧文章,但我认为它可能对搜索中的其他人有用。
您的问题在这里:

glActiveTexture(GL_TEXTURE4 + i);
glBindTexture(GL_TEXTURE_CUBE_MAP, pointlights[i]->DepthMap());


此替换应解决问题:

glActiveTexture(GL_TEXTURE4 + i);
glUniform1i(glGetUniformLocation("programId", "cubMapUniformName"), GL_TEXTURE4 + i);

glBindTexture(GL_TEXTURE_CUBE_MAP, pointlights[i]->DepthMap());


为着色器采样器设置纹理单位编号

关于c++ - 尝试采样立方体贴图纹理时为GL_INVALID_OPERATION,我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/39219908/

10-11 16:25