Best way to stack different exposures?

[russellhq]

Very interesting thread I've been debating whether to bin RGB 2x2 or not. With the MN190 and 214L+ I find the seeing generally limits me to 2-3 arcseconds or about 2 pixels and 2x2 binned subs seem pretty much as good as unbinned. So I've been using unbinned L and 2x2 binned RGB plus unbinned Ha, for galaxies. Then again would I be better with 2x2 binning everything in these conditions? Would this give me better S/N ratio for a given total time than unbinned?

If seeing limits you to 2 pixels, then don't you need to sample at 1 pixel anyway? Nyquist Theorem:

http://www.stark-lab...5_Sampling2.pdf

[opticalpath]

Very interesting thread I've been debating whether to bin RGB 2x2 or not. With the MN190 and 214L+ I find the seeing generally limits me to 2-3 arcseconds or about 2 pixels and 2x2 binned subs seem pretty much as good as unbinned. So I've been using unbinned L and 2x2 binned RGB plus unbinned Ha, for galaxies. Then again would I be better with 2x2 binning everything in these conditions? Would this give me better S/N ratio for a given total time than unbinned?

Well, you might be able to test the two methods with the data you've already got, Gina. If you have software that has a 'binning' math function (does AstroArt?), you could bin your full resolution L and Ha images in software and try combining everything in 2x2.

Now I know software binning is not the same as on-chip binning, but it's very nearly the same unless your images are read-noise limited, which is untypical unless you have excellent dark sky conditions and full exposures. Software binning will give you a gain in S/N (and a loss of resolution) that is similar to real binning under typical conditions. The supposed 'increased sensitivity' of binning is a myth; the benefit is increased S/N resulting from the (arithmetically) increased signal level, and the reduced read noise arising from fewer read events after on-chip binning. You get the first of those from software binning; the read noise benefit may or may not be significant depending on your conditions.

Adrian

[Gina]

If seeing limits you to 2 pixels, then don't you need to sample at 1 pixel anyway? Nyquist Theorem:

http://www.stark-lab...5_Sampling2.pdf

Yes, that's what my maths theory told me.

[Gina]

Putting figures onto my findings, the best FWHM I get with the MN190 and 314L+ unbinned varies from just under 2 to around 3. Binned 2x2 this figure roughly halves as expected.

I've read that PDF and reconsidered the Nyquist Theory which states that you need to sample at twice the highest frequency you want to retain. So I guess to retain 2 or 3 FWHM spacial resolution I need to samlpe at 1 to 1.5 pixels. In other words I WILL lose data by binning 2x2.

QED

[russellhq]

Yes, but will losing that data make a visible difference to your images? You already stated you could not discern the difference between binned and unbinned images for certain cases. So it may the case you lose data but that data does not make a contribution to which you can perceive.

If it's quicker binning, and you can't tell the difference when you don't, is it then worth sticking with binning? Maybe a more fruitful use of imaging time would be spent on capturing Lum?