Roger’s comments on my last April 2026 Jupiter image got me to thinking that I should re-evaluate some aspects of how I post-process planetary images going forward. I have been imaging Jupiter for many years, but over the past year my equipment configuration has radically morphed into a much more automated setup.
I used to do the bulk of my Jupiter imaging with my homemade Dobsonian and a DSLR camera. These were, clearly, highly manual sessions, a far cry from the motorized Celestron SCT I now use with a dedicated ZWO planetary camera, among several other automation components I am still learning to use. I should have realized much earlier that, at the least, a re-assessment of my end-to-end techniques was well past due.
Roger mentioned that he usually only stacks 10-20% of frames. Though he does mostly deep sky photography, this also makes sense on the planetary scale. If I am capturing over 11,000 frames in 90 seconds, ~20% should be more than sufficient and potentially produce better results than my prior method of going for 50-85% of frames, a range I settled on from Dobsonian/DSLR experience visually comparing a range of samples.
So I went back to AutoStakkert and pre-processed my April 26/27 raw stacked image for only the best 10% and 20% of frames. Further, I compared a small set of alignment points and a large set, for a total of four new takes on Jupiter.
In quick summary, of the four, I found 20% with 170 alignment points to be the best image. Compared to my original 50% frame selection, I feel this newer result is a little better. For example, you can see the festoons in better contrast. My March 9th Jupiter was, in my assessment, a better result, and there I used 75% of frames. As time allows, I plan to go back to that session and try the same 10-20% sampling there as well.
You have to be open to the possibility that continual experimentation may eventually produce superior results. I was too locked into a technique that had become outdated, that I had grown out of, and I thank Roger for his perspective that nudged me to a different point of view.
Equipment Used (mostly same as prior post, with new Autostakkert tweaks):
- Celestron NexStar Evolution 9.25 Schmidt-Cassegrain
- ZWO ASI676MC planetary camera
- Software for telescope control: CPWI
- Software for capturing: ASICAP
- – Capture Area Size = 1280 * 720
- – Capture Limit = 90 s
- – Colour Format = RAW8
- – Debayer Preview = ON
- – Debayer Type = RGGB
- – Exposure = 7.758ms
- – Frame Count = 11509
- – Gain = 114
- – Temperature = 19.5 C
- – TimeZone = -5
- No Barlow, no filter
- Created from a 90s video, best 20% of frames via Autostakkert
- 170 aliignment points in Autostakkert
- Focusing via Celestron’s motorized focuser in CPWI
- Software for post-processing:
- – PIPP
- – Autostakkert 4
- – WaveSharp 3
- PaintShop Pro for minor touch-ups
Discover more from Computer Looking Up
Subscribe to get the latest posts sent to your email.
Thanks Paul,
Great image.
As you say, I haven’t done a lot of planetary imaging lately.
So I’m no expert. When I first imaged planets I simply stacked every frame, thinking that the more frames, the better the image. Wrong. I noticed what other astro imagers chose to do and realised that only the best frames should be stacked. The Autostakkert frame quality graph provided a good guide to help choose which percentage.
Since then I’ve been interested in the settings which others use: total frames, total exposure time (to limit rotation) and percentage of frames finally stacked.
That’s because one day I’ll get back to it.