...
- Aoppa – according to documentation, this function "Precomputes apparent to observed place parameters required by slaAopqk and slaOapqk". Comparison between PALPY and pyslalib shows that the two routines give consistent outputs to within one part in 10^5. Comparison between PALPY and slalsst.so gives a discrepancy of 17 a few 0.1 arcseconds in the output longitude and latitude parameters. I do not understand the origin of this discrepancy.
- Aopqk – "Quick apparent to observed place." As long as you are considering a location on the sky whose zenith distance is less than 0.4 pi radians, PALPY is consistent with both pyslalib and slalsst.so to within 1 part in 10^10. If the zenith distance is greater than that, the agreement between PALPY and SLALIB quickly becomes unacceptably high. This could be because of nonsense inputs.
- De2h – "Convert equatorial to horizon coordinates." PALPY agrees with both pyslalib and slalsst.so to one part in 10^10
- Dsep – gives the angular separation of two points on the sky. PALPY agrees with both pyslalib and slalsst.so to 3 parts in 10^10
- Epj – "Conversion of modified Julian Date to Julian Epoch." PALPY agrees with pyslalib and slalsst.so exactly.
- Eqgal – "Transformation from J2000.0 coordinates to IAU1985 equatorial coordinates." PALPY agrees with both pyslalib and slalsst.so to 2 parts in 10^10
- Gmsta – Converts from universal to sidereal time. PALPY disagrees with pyslalib and slalsst.so by 0.3 arcseconds. However, this is because PALPY uses the IAU2006 conventions. SLALIB uses the IAU1985 conventions.
- Mappa – "Compute star-independent parameters in preparation for conversions between mean place and geocentric apparent place." PALPY agrees with pyslalib and slalsst.so to within a few 10^-5 arcseconds. Note: PALPY and pyslalib disagree about whether the precession-nutation matrix should be output in row-major or column-major order. PALPY and slalsst.so agree on this ordering.
- Mapqk – "Quick mean to apparent place: transform a star RA,Dec from mean place to geocentric apparent place, given the star-independent parameters." PALPY agrees with pyslalib and slalsst.so to within a few 10^-5 arcseconds.
- Pm – "Apply corrections for proper motion to a star RA,Dec." PALPY agrees with pyslalib and slalsst.so to within a few 10^-4 arcseconds. However, this agreement breaks down if the parallax to the star being considered is very low. The threshold value appears to be a parallax of 0.00045 arcseconds. Below this value, PALPY disagrees with SLALIB on the order of a few hundred arcseconds. Above this value, PALPY agrees with SLALIB at the sub-milliarcsecond level.
- Prenut – calculate the precession-nutation matrix. PALPY agrees with pyslalib and slalsst.so to within a few parts in 10^6.
- Refco – calculate refraction coefficients. PALPY agrees exactly with pyslalib. PALPY agrees with slalsst.so to with a few 10^-3 arcseconds.
- Refz – "Adjust an unrefracted zenith distance to include the effect of atmospheric refraction, using the simple A tan z + B tan3 z model." PALPY agrees exactly with pyslalib. PALPY agrees with slalsst.so to with a few parts in 10^10.
...