Add minor clarification for "Fisheye Lens Polynomial" section based on the 3.1 update notes.
Also corrected small typo in "Mirror Ball" section
Details
Details
Diff Detail
Diff Detail
- Repository
- rBM Blender Manual
Event Timeline
Comment Actions
I don't think the modified description is an improvement.
"angle of the world ray" is rather vague so I intentionally changed that to "direction with spherical coordinates". Putting both descriptions in a single sentence is confusing.
Comment Actions
@Brecht Van Lommel (brecht) I agree world ray is somewhat vague. When I first read I wasn't quite sure what direction referred to until I read the patch notes with more detail. I have changed the text to be somewhat more clear than what the patch notes say yet still offer slightly more detail than the existing text on the Cameras page.
Let me know if this is still not what you want the description to be.
Comment Actions
It's still confusing to me. We have not defined what "that pixel" is, and "incoming image" is strange terminology since we are rendering to an image, not from one.
Maybe do this:
Match a real world camera by specifying the coordinates of a 4th degree polynomial.
The projection works as follows.
Pixels in the image are mapped to positions :math:`(x, y)` on the camera sensor in mm.
A position on the sensor is mapped to a direction with spherical coordinates
:math:`(1, \theta, \phi)` in radians as follows:
.. math::
& r = \sqrt{x^2 + y^2}\\
& \theta = k_0 + k_1 r + k_2 r^2 + k_3 r^3 + k_4 r^4\\
& \phi = acos(x/r)
Incoming light from this direction is then projected onto the corresponding pixel.Comment Actions
Change to reflect suggested wording from comment from @Brecht Van Lommel (brecht) .
I agree this wording is the most comprehensive and clear.