SPLOT: integration over log sampled spectra

Frank Valdes wrote on Jul 24, 1998

Q: I have read through the docs and the newsletters and can't find the
simple reassurance I am looking for. Can I be sure that all splot
functions involving integrating over some part of the spectrum are done
correctly if the spectrum has been put into logarithmic wavelength bins
using twodspec.transform?


A: The SPLOT functions involving integrating over a part of the spectrum
are done in pixel space.  The only place the dispersion function enters,
either for logrithmic binning or any other type of non-linear dispersion,
is in converting the end-points to pixel coordinates.  Anyway, for
what it's worth, you can be assured that there is no (known) problem with
logrithmically binned data.

---

Last post on Jul 24, 1998