Pixel integration #80
ConnorStoneAstro
started this conversation in
Ideas
Replies: 1 comment
-
|
TRapezoid rule requires only N+M extra operations on a sampling which is already N*M so it is probably worth it. Simpsons rule goes a step further and requires evaluating on pixel vertices, center, and midpoint of each edge. That is considerably more evaluations (of order 2-3x more) but the convergence is very fast and so fewer subpixel integrations would be needed. These could be given as options to the user |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Perhaps it would be possible to switch AutoProf over to integration with the Trapezoid rule. This would involve evaluating at pixel edges instead of pixel centers. It would also cost more evaluations per sub gridding, so I'm not sure if the extra convergence is worth it.
Beta Was this translation helpful? Give feedback.
All reactions