Imaging with oversampling? Are there any Techniques to it? Advice and discussion

5 replies164 views
Menelaos avatar

Hey guys! This is general discussion and if you have any advice on the subject! I was wondering if there are any people out there that are imaging in particular deep sky objects (not planetary or lunar photography) with oversampling, either with long focal length or small pixels camera (if there is a difference between the two). I know there is lucky imaging but it’s hard to stack a large number of images for deep sky objects and if not mistaken there isn’t a software that actually does the job for deepsky objects. How do we battle seeing? Are there any Techniques for it? Any equipment in particular that is more capable in this area? I know i can sharpen the images with Blur exterminator but this is not real details right? Thank you.

andrea tasselli avatar
Mine are often richly oversampled but I wouldn't call them at high resolution. You can't beat seeing and that is that, you need to move to where the seeing is uniformly good. You also need a fairly large aperture, think 16" or larger.
Kevin Morefield avatar

I wouldn’t be too concerned about being oversampled. Virtually everyone with a scope beyond 1500mm is oversampled to some degree these days. With my CDK17 my typical FWHM range is 5-7 pixels which is quite oversampled.

You can resample down to something like a 3 to 4 pixel FWHM and gain SNR from quadrature. Aside from that, I wouldn’t worry about it.

Well Written Helpful Insightful Concise Supportive
Brian Puhl avatar

Kevin Morefield · Sep 8, 2025, 11:29 PM

Virtually everyone with a scope beyond 1500mm is oversampled to some degree these days.

If only I could make this echo throughout the halls of astrophotography…. I would even go as far as to say 80% of people with a focal length over 1000mm. It isn’t the same as 5 years ago with 9um pixels. Alot of the information spread around the internet have gone stale since modern CMOS and 3.7um pixels are pretty standard now. Aperture is alot more important than focal length.

…but to the point… You can’t battle seeing. You are always at it’s mercy. Build your scope/setup around your sky quality. BlurX is the equivalent of polishing a turd. It’s powerful but you can’t magically make your data something it isn’t. Some folks experiment by imaging in the IR range, but it hasn’t been of interest to me so far.

Robert (Bob) Hamers avatar

Modest oversampling is perfectly fine. It’s always a tradeoff between getting enough photons per pixel (favoring undersampling) and resolution (favoring critical sampling or oversampling). Diffraction “limits” are not hard walls — the ability to resolve two objects does not suddenly vanish as one hits any particular pixel size or focal length, so slight oversampling (say, up to ~2x) would seem to make sense. As long as you’re not photon-starved, there’s no need to worry too much.

Well Written Helpful Insightful Concise
Tony Gondola avatar

I have experimented with over sampling and the net result seems to be a smoother result and maybe slightly more detailed result. You can measure improvements in FWHM but it’s not huge. Of course it effects SNR so it really depends on the subject and what you’re trying to do with it. I also tend to shoot short subs and do a lot of culling so it’s hard to untangle where the positive effects are mostly coming from. My basic setup is 2.9 micron pixels so I’m normally running at 0.66” per pixel with a 6” aperture. On good nights I feel that finer sampling would be warranted. I have also tried .33” arc sec per pixel with that same system as well as cameras with pixels as small as 1.49 microns with results as stated above. Overall I would say that if seeing decent and you are willing to shoot short subs and aggressively cull that yes, you can see a benefit in terms of sharpness but certainly not in going faint where SNR is everything. As Bob said, most of the time, it’s a balance.

Helpful Insightful