#593 closed defect (fixed)
psProject / psDeproject
| Reported by: | eugene | Owned by: | |
|---|---|---|---|
| Priority: | high | Milestone: | |
| Component: | astro | Version: | 0.8.0 |
| Severity: | minor | Keywords: | |
| Cc: | robert.desonia@… |
Description
psProject and psDeproject are inefficiently implemented, and can have
out-of-range errors. these functions calculate phi and theta via, eg, atan of a
polynomial combination of the input angles, then calculate their sines and
cosines. this is better done by writing the sines and cosines directly as
combinations of the arguments used to calculate the atan. I am attaching
example code which calculates the projections this way, but is not very PS-lib
compliant.
Attachments (2)
Change History (10)
by , 21 years ago
comment:1 by , 21 years ago
| Owner: | changed from to |
|---|
comment:2 by , 21 years ago
| blocked: | → 592 |
|---|
comment:3 by , 21 years ago
| blocked: | 592 |
|---|---|
| dependson: | → 592 |
comment:4 by , 21 years ago
| Cc: | added |
|---|
comment:5 by , 21 years ago
There are also errors in the projection implementation: it turns out that the
values of phi are being returned in the wrong quadrant due to not using atan2 to
calculate phi in psDeproject. I needed to implement the sin/cos version of the
code to catch this. The new implementation is in cvs under tag eam_rel8_b0
comment:6 by , 21 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
branch has been joined with cvs trunk
example of psDeproject code without so many sines and cosines