I kind of like Siril for stacking now despite my initial annoyance. Can PS actually register and stack Astro images?

Nope!

Wait wait wait wait -- sure it can! I mean, I know the NGC 7000 image I posted a few days ago ain't exactly APOD-worthy, but it was done entirely in Bridge and Photoshop -- select the DNGs in Bridge, Tools | Photoshop | Open as Photoshop layers, then in PS Edit | Auto-align layers, select all layers, right-click, Convert to Smart Object, Layer | Smart Object | Stacking Mode | Median. Then start with the stretching and the fiddling and the occasional naughty language.

To be fair, I have had PS fail to register my images from time to time. But these were taken on two different nights and I didn't align the two sets very well, and PS did just fine registering them. And yeah, Photoshop median stack is pretty primitive compared to what DSS or Siril or PI can do for stacking. Still.

Huuuh - sorry


By "registration" I understand that the positions of all the stars detected in an image are measured and put into a "registry" to be used for the alignment process that follows. The stacking program produces a dataset or file for each light that lists all the positions. PS can't do this.


But PS can surely align images that have little distortion and sufficient concordant structures which means stars in our context. And if PS manages the alignment the layers can be stacked with basic statistic functions. So if there are good data and maybe structures other than stars in the lights PS has a chance to align the layers properly and calculate a median or averaged result. I think you guys use the expression "this is another ballpark".

I gave up on Siril. I was frustrated by the lack of consistancy is it’s operations. Sometimes applying camera WB, sometimes not.

I testing pixinsight now which is nice but I’ll probably go back to Nebulosity because of the cost of PI.

Neb is manual registration but after spending 4 hours with Siril to guide it away from mistakes, I realized that manual registration although a bit tedious is more dependable... And easy if your PA and guiding is good.

I tried DSS on a Mac with a thing called Wine and it worked beautifully but took at least 6-8 hours. (39 images of veil neb - lots of stars!)

I’m wondering what others here are using?

DSS, Lightroom, photoshop. I also have the new acdsee program and I like it for general photography but I haven't tried it for astro.

Allen Dyer is prompting an all PS solution. It’s auto panorama stitching algorithm does the stacking automatically. This could be all one needs with a DSLR

I too have had problems with color in Siril, have not experimented to see if it's any different if original frames are FITS or DNG. I must admit I'm intrigued by "automatic stacking", though as a software engineer who likes command lines and a pilot who enjoyed "steam gauge" cockpits, I'm somewhat possessive about lots of knobs and switches. Still, it's the image that matters. I see that he does workshops on this stuff, do you have a tutorial link?

Yes! You just made my day. I'm probably the only one who called them out about their colour rendering. The treated me like a pain in the butt. From my perspective they were not consistent with their colour management and making it frustrating for the user. But they just wouldn't get my point. They just kept telling me its normal and to do a colour balance. (The settings are not carried through the process - for example if you do pre-processing the camera WB is not used even though its selected to be used.)

The stacking didnt work well with the veil nebula. Not like PI, DSS or even Photoshop. I just did a complete PS pre-process and stack and I have to admit that it was better than anything else so far - Although I'll probably have to take a closer look at DSS again.

The video tutorials are helpful. If you sign up you can get them here: Apprenez à utiliser Siril: The basics

I was referring to Dyer, not Siril, though I see I did not make that at all clear!

Here's the Dyer PS tutorial: 10 Steps to Processing: Deep-Sky

Brilliant, thanks mon!