Source code for pbrt, the renderer described in the third edition of "Physically Based Rendering: From Theory To Implementation", by Matt Pharr, Wenzel Jakob, and Greg Humphreys.
The code snippet named "Handle sphere mapping discontinuity for coordinate differentials" in function SphericalMapping2D::Map may not be able to handle the discontinuity of the t texture coordinate in sphere mapping properly.
I think that code snippet assumes the absolute value of (dstdx)[1] or (dstdy)[1] is less than 1, but this assumption does not hold as you set delta as 0.1.
Here is the function:
Point2f SphericalMapping2D::Map(const SurfaceInteraction &si, Vector2f *dstdx,
Vector2f *dstdy) const {
Point2f st = sphere(si.p);
// Compute texture coordinate differentials for sphere $(u,v)$ mapping
const Float delta = .1f; // **DELTA IS 0.1**
Point2f stDeltaX = sphere(si.p + delta * si.dpdx);
*dstdx = (stDeltaX - st) / delta;
Point2f stDeltaY = sphere(si.p + delta * si.dpdy);
*dstdy = (stDeltaY - st) / delta;
// Handle sphere mapping discontinuity for coordinate differentials
if ((*dstdx)[1] > .5) // **THE VALUE HERE COULD BE CLOSE TO 10.0**
(*dstdx)[1] = 1 - (*dstdx)[1];
else if ((*dstdx)[1] < -.5f)
(*dstdx)[1] = -((*dstdx)[1] + 1);
if ((*dstdy)[1] > .5)
(*dstdy)[1] = 1 - (*dstdy)[1];
else if ((*dstdy)[1] < -.5f)
(*dstdy)[1] = -((*dstdy)[1] + 1);
return st;
}
I tested it with a scene using spherical mapping. The log showed the value of (dstdx)[1] could be something like 9.999742. In this case, the final value of (dstdx)[1] was -8.999742 after the code handled the discontinuity. In my opinion, this final value is not desired.
I am a beginner in this field, so I may be wrong here. If I am wrong, hope you could briefly explain why that code snippet could still handle discontinuity properly in the example I mentioned.
Hi,
The code snippet named "Handle sphere mapping discontinuity for coordinate differentials" in function SphericalMapping2D::Map may not be able to handle the discontinuity of the t texture coordinate in sphere mapping properly.
I think that code snippet assumes the absolute value of (dstdx)[1] or (dstdy)[1] is less than 1, but this assumption does not hold as you set delta as 0.1.
Here is the function:
I tested it with a scene using spherical mapping. The log showed the value of (dstdx)[1] could be something like 9.999742. In this case, the final value of (dstdx)[1] was -8.999742 after the code handled the discontinuity. In my opinion, this final value is not desired.
I am a beginner in this field, so I may be wrong here. If I am wrong, hope you could briefly explain why that code snippet could still handle discontinuity properly in the example I mentioned.
Thanks.