I'm trying to implement Variance Shadow Mapping. As I searched around I found many sites mentioned that frontface culling could reduce self-shadowing artifacts, but I failed to generate shadow map whenever I turned on frontface culling, and I think it's reasonable, because all the front facing triangle would be culled out. I mean let's imagine your eyes are camera and there's a counter-clock winding(front facing) triangle in front of you, then you do a frontface culling and throw the triangle away, now how could you get the depth of the back face of that triangle? Could someone explain this to me?

another question is: when I render depth to texture, OpenGL would convert x and y from normal device coordinate to texture coordinate(from [-1, 1] to [0, 1]), does the same goes to z axis?

then you do a frontface culling and throw the triangle away, now how could you get the depth of the back face of that triangle? Could someone explain this to me?

You do not disable culling, just redefine what is front face, wheather it clockwise or , counter clockwise.

Just switch the front face identification by winding, this will render the actual "backfaces" now front faces and not render the actual "frontfaces" now backfaces. Do not disable culling, reverse it for the front face, and keep culling backface. Rememeber that vertex attributes such as normals will be bogus by this, so stick to only position and rasterized depth and such.

Thank you guys, I think I got it. Front face culling would fail shadow map generation for flat objects with no thickness.

anyone know answers of my second question?

When the shadow map texture is being generated, the depth is (normally) set to pos.z/pos.w (homogeneous depth).

Thank you Buckeye

I wrote gl_FragCoord.z to texture and it ranges for 0 to 1.