ISIS - gnuplot

[Andrew Smith]

Hi I am trying to process some Tau DR spectra and have go to the point to plot the result out. However, ISIS V5.1.4 says it has written the file but it is not there! Once I have done it once the polt commands don't seem to work any more. I have gnuplot v 4.6 and win 7 64bit Home Premium. The directories are correct and gnuplot is in c:\gnuplot\ It has worked with earlier versions of ISIS or at least I think it has - old age. Advice welcome.

Thanks Andrew

[Christian Buil]

Andrew,

A very important point, do you have
download the lastest xxx.std files
from the page:

http://www.astrosurf.com/buil/isis/gnuplot/install.htm

?

(note: size of plot is modifiable from interface with this version)

Christian B

[Andrew Smith]

Thanks Christian that was the problem - sorry for wasting your time.

Christian, I have been using ISIS to process some Star Analyser spectra and found getting a good response curve quite difficult. This is mainly due to the mismatch in resolution between the reference spectra and the real spectra. I wonder if it would be possible to add the ability to "filter" the reference spectra to get a better match. This would then reduce the artifacts at the spectra lines and make creating the response curve simpler and more accurate.

Thanks for the software and your continuing efforts.

Regards Andrew

PS I tried this by replacing the Pickles A0V file with one I filtered using the profile filter tool and it worked very well. It would be best to be able to do it ineractivly but I have a work round.

[Robin Leadbeater]

Christian, I have been using ISIS to process some Star Analyser spectra and found getting a good response curve quite difficult. This is mainly due to the mismatch in resolution between the reference spectra and the real spectra. I wonder if it would be possible to add the ability to "filter" the reference spectra to get a better match. This would then reduce the artifacts at the spectra lines and make creating the response curve simpler and more accurate.

Hi Andrew,

I currently still use Visual Spec for this step because of the flexibility to try this sort of thing and also the tools to remove specific parts of the spectrum and pick specific points on the continuum to fit the curve to.

Cheers
Robin

[Christian Buil]

Andrew,

Look the Profile tab, you can use Filter for ajust the spectral resolution
of the reference spectrum (save the result, for a personnal a set of reference
spectrum). You can made the division by using Comparison tool for exemple,
for compute instrumental response.
It is possible to remove specific part of the spectrum from Continuum
tool for remove residual artifact.

Christian

[Andrew Smith]

Thanks Christian - I did try that and it works (see the PS in my 2nd post). I had to do some trial and error to get the right amount of filtering but It did work. I would be nice though to be able to do the filtering during the response creation process where you can see the result in real time. However, I realise you may not agree or wish to spend the time on it.

Regards Andrew

[Peter Somogyi]

Just FYI.

I had to put the gnuplot install really under "C:\gnuplot", otherwise ISIS didn't work with "C:\Program Files (x86)\gnuplot" even though set that long path in ISIS setting tab.
Maybe special chars / long path name issue.
Now it works.