New Astrophotography Stacking

[Bencooper]

When one takes several (20?) identical frames of part of The Milky Way (for improving signal to noise ratio in" stacking"), one is actually capturing moving stars (not to be confused with stitching several adjacent and overlapping frames in a panorama)

Alignment in New Astrophotography Stacking workflow (shift/rotate to a base frame) supposedly removes this apparent movement. However,  lenses can incorporate a varying degree of optical distortion (barrel/pin-cushion),  particularly wide (and superwide) angle lenses, stretching dimensions at the edges of the frames. In effect, stacking results in star trails at the edges of the final image -see attached.

How can I use the Affinity built-in library of lens data in the New Astrophotography Stacking (RAW) workflow to remove the distortion provided by my excellent Tokina AT-X Pro 11mm-16mm DXII (shooting at 11mm) before alignment and stacking in New Astrophotography Stacking? Otherwise, should I be shooting in a longer focal length (like 14 or 16mm, losing some field width)?

AffinityRenderedReduced.tif

Edited July 9, 2021 by Bencooper
to clarify.

[Dan C]

Hi @Bencooper,

My sincerest apologies for the delayed response here!

I have spoken with our resident Astro-expert here who has recommended the following:

1. Pre-process the RAW files in the Affinity Develop persona, as this will apply the correct lens correction profile to your images and should reduce/remove the trails you're seeing when stacking.
2. Export each developed image as a 16-bit TIFF, then you can try stacking these exported images.
3. Another option to try is the File > New Stack method, rather than the specific Astro-stack as this has dedicated options to help force alignment for the stacked images.

I hope this helps

[Bencooper]

Thank you for your advice - i will try these options and report back ! 😀

[dizzwave]

I'm having similar issues using the new astro workflow, with my lens at 14mm (Sigma 14-24).  The sky looks great in the center, but I'm getting warping toward the edges.

Ben, were you able to solve it with any of the options that Dan mentioned?

I tried pre-processing the raw files with lens corrections, output to tiff (no luck); and also tried the non-astro stacking workflow, but I had even worse luck with that.  I will have to mess around with it more, but so far I've had better luck with Sequator.

Thanks!

[Bencooper]

Dizzwave, I haven't yet tried Dan's suggestions as I was hoping to find a batch process in Affinity to do the work (20 images). I did find that Sequator (in that respect, and so quick!) sorted the problem intelligently but were all those stars REALLY stars or flysh_t (noise)! It also removed some aircraft trails automatically !(see attached, sky pollution at bottom)

However, I did try selecting best images visually and found one to have camera shake (dual stars - dodgy tripod?). Barrel distortion effectively warps space and with stacking  speeds the affected stars, causing trailing. I think distortion correction is only approximate and even with the best optics the solution really is to use larger focal length (16 or 18mm), lose some field, and shorter exposure to reduce tracking.

I will try the workaround and report back (I am a newbie at Affinity but recognise practice makes perfect) but I believe Affinity should address these fundamentals ie incorporate distortion correction within New Astro Stacking and providing a batch process somewhere.

Sequator2Affinity rendered.tiff

[Old Bruce]

7 hours ago, Bencooper said:

camera shake (dual stars - dodgy tripod?).

You can have a great tripod but it will transfer ground vibrations and then there is vibration from the wind, a light breeze can set up some resonance. In my experience the best way to eliminate the vibration is to have the tripod as short as you can and maybe go so far as to add some mass to it, hang the camera bag (with the lenses and other stuff inside) on to it.

[IanSG]

1 hour ago, Old Bruce said:

You can have a great tripod but it will transfer ground vibrations and then there is vibration from the wind, a light breeze can set up some resonance.

Vibrations are going to give you blurry stars rather than dual ones.  If the tripod is the problem it's more likely that something slipped because it wasn't locked properly.

1 hour ago, Old Bruce said:

In my experience the best way to eliminate the vibration is to have the tripod as short as you can and maybe go so far as to add some mass to it, hang the camera bag (with the lenses and other stuff inside) on to it.

Good advice!  It's also worth using a remote shutter release (probably wireless these days) and maybe even going back to the hat trick.

[dizzwave]

9 hours ago, Bencooper said:

Barrel distortion effectively warps space and with stacking  speeds the affected stars, causing trailing. I think distortion correction is only approximate and even with the best optics the solution really is to use larger focal length (16 or 18mm), lose some field, and shorter exposure to reduce tracking.

Thanks Ben -- I believe it is this exactly, as (in both of our wide angle cases) the stars are aligned in the middle, but increasingly warped the farther from center you go.  And smoothly warped at that, i.e. not due to shake or some such operator error.

This leads me to believe that Affinity is simply aligning the entire layer, but favoring the stars in the middle for alignment, and letting the others slide. Whereas Sequator has a more intelligent algorithm that aligns all stars throughout the entire field.  My conclusion is that Affinity's astro stacking feature is probably useful for deep space stuff, but not the kind of landscape photos I want to do.  (Yet!  Fingers crossed for improvements!  I have faith in Affinity... It is my favorite piece of software overall right now.  :-))

-Dave

[Bencooper]

Of course, Dave, what we didn't mention (I think) in "star stacking" was that the natural imperative in night photography to open apertures wide (1.4/2.8 etc) to capture the light, leads to chromatic aberration and distortion in the corners leading to star trails and coma,  irrespective of the quality of glass (this might not be significant or noticeable in normal daytime landscapes). Similarly, cranking up ISO above 3200 might lead to increased noise but, as importantly, loss of dynamic range and colour !  In this regard, Star Tracking devices become more desirable !

[Bencooper]

On 7/31/2021 at 5:17 PM, Old Bruce said:

You can have a great tripod but it will transfer ground vibrations and then there is vibration from the wind, a light breeze can set up some resonance. In my experience the best way to eliminate the vibration is to have the tripod as short as you can and maybe go so far as to add some mass to it, hang the camera bag (with the lenses and other stuff inside) on to it.

thank you Old Bruce - I usually carry an old plastic shopping bag which I fill with kilos of local pebbles to hang from the tripod during shooting which seems to do the job. Also a home-made canvas windbreak on poles driven-in to protect the tripod (but not the camera !)

[dizzwave]

4 hours ago, Bencooper said:

Of course, Dave, what we didn't mention (I think) in "star stacking" was that the natural imperative in night photography to open apertures wide (1.4/2.8 etc) to capture the light, leads to chromatic aberration and distortion in the corners leading to star trails and coma,  irrespective of the quality of glass (this might not be significant or noticeable in normal daytime landscapes). Similarly, cranking up ISO above 3200 might lead to increased noise but, as importantly, loss of dynamic range and colour !  In this regard, Star Tracking devices become more desirable !

Ah, I didn't think about those effects of going wide open aperture; thanks for that.

I second that on star trackers--in fact I just ordered one a few days ago! I imagine I'll still shoot at 2.8, but.... we'll see. Lots of experimentation to do. 

[Bencooper]

Dave, IMHO once you have a star tracker, you can keep the exposure open as long as you like and there is no other reason to shoot at f/2.8 ?

In fact, you will get better depth of field if you stop down some,  using hyperfocal length to capture stars (infinity) and more foreground as well in the case of panoramic stitching (without star tracker, which would blur any fixed foreground). I'm not selling it, but Photopills App is a really great set of tools, and it has a useful H-F distance calculator !