Currently i am plotting these data summing $gun_fwhm / $disp pixels around the $current_wav. For chromaTEM, especially, $gun_fwhm is comparable to $disp. This means that sometimes you will integrate 0-1 pixels around central value and will have a lot of noise.
think on what is the best way to solve this. Maybe you can put a new line_edit with integration region. Also, you don't necessarily would need to integrate 1 * FWHM. If you are at chromaTEM, you can integrate a bit more because you are sure you are not overlapping with ZLP. In lumiere, this is more sensitive: if you can increase window too much then you will start to summing up ZLP.
obs: In LUMIERE, this is extremely important for long wavelengths because they more and more close to ZLP. In practice, you need to kill a bit of signal in order to have a less polluted one.
Currently i am plotting these data summing $gun_fwhm / $disp pixels around the $current_wav. For chromaTEM, especially, $gun_fwhm is comparable to $disp. This means that sometimes you will integrate 0-1 pixels around central value and will have a lot of noise.
obs: In LUMIERE, this is extremely important for long wavelengths because they more and more close to ZLP. In practice, you need to kill a bit of signal in order to have a less polluted one.