What's wrong with my framebuffer?

Question

I'm creating a framebuffer object to be my gbuffer for deferred shading. I mainly learned from http://ogldev.atspace.co.uk/, and modified to be a little more... sane. Here's the source code where I create the framebuffer:

/* Create the FBO */
glGenFramebuffers(1, &fbo);
glBindFramebuffer(GL_FRAMEBUFFER, fbo);

/* Create the gbuffer textures */
glGenTextures(GBUFFER_NUM_TEXTURES, tex);

/* Create the color buffer */
glBindTexture(GL_TEXTURE_2D, tex[GBUFFER_COLOR]);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA8, width, height, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, tex[GBUFFER_COLOR], 0);

/* Create the normal buffer */
glBindTexture(GL_TEXTURE_2D, tex[GBUFFER_NORMAL]);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RG16F, width, height, 0, GL_RG, GL_FLOAT, NULL);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT1, GL_TEXTURE_2D, tex[GBUFFER_NORMAL], 0);

/* Create the depth-stencil buffer */
glBindTexture(GL_TEXTURE_2D, tex[GBUFFER_DEPTH_STENCIL]);
glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH24_STENCIL8, width, height, 0, GL_DEPTH_STENCIL, GL_FLOAT, NULL);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_DEPTH_STENCIL_ATTACHMENT, GL_TEXTURE_2D, tex[GBUFFER_DEPTH_STENCIL], 0);

GLenum drawBuffers[] = {GL_COLOR_ATTACHMENT0, GL_COLOR_ATTACHMENT1};
glDrawBuffers(2, drawBuffers);
glReadBuffer(GL_NONE);

/* Check for errors */
GLenum status = glCheckFramebufferStatus(GL_FRAMEBUFFER);
if (status != GL_FRAMEBUFFER_COMPLETE)
{
    error("In GBuffer::init():\n");
    errormore("Failed to create Framebuffer, status: 0x%x\n", status);
    fbo = 0;
    return;
}

// restore default FBO
glBindFramebuffer(GL_FRAMEBUFFER, 0);

When I run this, however, status returns GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT. If it's not clear, I'm trying to create 3 gbuffers:

• a 32-bit RGBA color buffer (I'd use 24-bit but I'm scared of alignment penalties),
• a 32-bit RG normal buffer (each component using a 16-bit float, but I might get away with a signed short?)
• a 24-bit Depth buffer packed with an 8-bit Stencil buffer
• (total of 96 bits, or 12 bytes)

Possible problem areas that I can see might be using GL_FLOAT for the normal buffer, and GL_FLOAT for the depth-stencil buffer. I'd imagine GL_HALF_FLOAT would be more appropriate for normals, but that's not on the list of types that I can use with glTexImage2D. Similarly, I have no idea what type is most appropriate to use for a depth-stencil buffer.

What am I doing wrong?

Answer 1

Your use of GL_FLOAT is mostly irrelevant, since no pixel transfer actually happens.

You can supply anything you want there as long as it is a meaningful data type. While no pixel transfer happens when you pass NULL for data, GL still validates the pixel transfer data type against the set of valid types and will raise an error if you do something wrong. To that end, if it raises an error the texture will be incomplete and thus cannot be used as an FBO attachment.

Here is where the problem lies, GL_FLOAT is not a meaningful data type for pixel transfer into a packed GL_DEPTH_STENCIL image format... it expects a packed data type such as GL_UNSIGED_INT_24_8 or something really exotic like GL_FLOAT_32_UNSIGNED_INT_24_8_REV​ (64-bit packed Floating-Point Depth + Stencil format).

In any event, there are actually two components that need to be packed into your data type. GL_FLOAT can only describe one of the two components, because floating-point stencil buffers are meaningless.

---

By the way, this whole confusing mess about pixel transfer data type can be completely avoided if you use something like glTexStorage2D (...) to only allocate storage for the texture. glTexImage2D (...) does double-duty, allocating storage for a texture LOD and providing a mechanism to initialize it with data. You really do not care about the later part if you are drawing into the texture with an FBO, since that is the only place it gets any data from.