problem with listpix
lin wrote on Apr 14, 2008
hi everyone,
i type : listpix *.fit wcs=world > *.fit.txt , then draw a graph used gnuplot .
but this *.fit.txt , its first column has some mistake
like 3.5808 53.082e-17
3.5809 53.0933e-17
3.5810 53.1051e-17 ....
i use splot to examine , these wavelengths ought to begin at about 3800
Are there any settings must be change?
i type : listpix *.fit wcs=world > *.fit.txt , then draw a graph used gnuplot .
but this *.fit.txt , its first column has some mistake
like 3.5808 53.082e-17
3.5809 53.0933e-17
3.5810 53.1051e-17 ....
i use splot to examine , these wavelengths ought to begin at about 3800
Are there any settings must be change?
Mao yewei wrote on Apr 14, 2008
Hi Lina,
Allow me to try to describe your problem more simply although I've no idea how to resolve it.
For a spectrum, you used LISTPIX to create a text file of wavelength-flux. But you found that the wavelength data were wrong by checking them using SPLOT. So you may not sure how to use LISTPIX to write right wavelength data.
Is that your case?
Owen
Allow me to try to describe your problem more simply although I've no idea how to resolve it.
For a spectrum, you used LISTPIX to create a text file of wavelength-flux. But you found that the wavelength data were wrong by checking them using SPLOT. So you may not sure how to use LISTPIX to write right wavelength data.
Is that your case?
Owen
lin wrote on Apr 14, 2008
aha..
Yes,it is.
Thank you all the same.you are very kind :D
Yes,it is.
Thank you all the same.you are very kind :D
Francisco Valdes wrote on Apr 14, 2008
My guess is that you have calibrated your spectrum in log wavelength intervals. SPLOT displays this in linear wavelength but the actual world coordinate system in the header is in log(lambda). General tasks, such as listpix or implot, simply output the wcs without knowing that you might want to display in linear units.
The solutions are either to simply plot with SPLOT or use something like DISPCOR or DISPTRAN to create a new version in linear wavelength for use with listpix and your gnuplot.
If you want confirmation either look at the header listing where you will see DC-FLAG=2 or CRVAL1/CD1_1 will be numbers like 3.111 and some small step size. If you want further confirmation you could post the header listing and I can tell you if this is right.
Yours,
Frank Valdes
The solutions are either to simply plot with SPLOT or use something like DISPCOR or DISPTRAN to create a new version in linear wavelength for use with listpix and your gnuplot.
If you want confirmation either look at the header listing where you will see DC-FLAG=2 or CRVAL1/CD1_1 will be numbers like 3.111 and some small step size. If you want further confirmation you could post the header listing and I can tell you if this is right.
Yours,
Frank Valdes
Last post on Apr 14, 2008