APP vs DSS vs Pixinsight

I recently started trying to use APP, I'm not crazy about the interface sometimes, but it makes sense as you start working with it, and I think it does a better job stacking than DSS. I know there's more that it can do too, and I am still just scratching the surface. I've only really been at this for under a year, so I would probably still be in that beginner category too.

I know APP has a free 30 day trial, and I think PixInsight does too. You can try them before buying (or renting in my case with APP) and see how you feel about them.

For me it depends on what I am doing. I shoot with a color camera (Canon EOS Ra) and use filters from my light polluted skies. If I am wanting a straight process, I use Deep Sky Stacker and go straight into Photoshop. If I am using a narrowband filter (such as the L-eNhanced) I will also use APP to extract the individual channels to do a simulated narrowband palette, such as SHO.

I have never used PI - but I know from friends it is good. In the end, I think a lot of it is personal preference

DSS major feature is the stacking while APP has additional, a bit more advanced features, for stretching and post processing. New features as star reduction are going to come with the next release. Therefore, I would consider DSS not comparable to APP or PI.

From a learning perspective, I'd say you'd produce results quicker with APP. If you are more advanced, you might be interested in PI for the final notch. 

An if you are a nerd, you do everything in Affinity Photo (from stacking to post processing and publishing).

I can't comment on APP or PI, but since you mentioned a free stacking only program like DSS you might want to have a look at ASTAP. Apart from being a really nice software to use for plate solving in NINA and being free it is a pretty capable stacking program IMHO. So far, I stacked several of my data sets with DSS and ASTAP and compared the resulting stacks. The ASTAP stack was always superior in terms of noisiness and signal quality. With the ASTAP stack it has always been easier to make faint details visible without flooding the image with noise. If you don't want to spend money on software to replace DSS yet, I would recommend to try ASTAP.

Best regards
Stefan

Hi, I'd recommend SIRIL highly, as it provides an end to end workflow, including excellent stacking and registration capabilities, stretching, background extraction and photometric colour calibration etc.   I found that DSS struggles with faint images, and also seems to strip colour in one channel of another whichever settings you use.   I've played with PI quite a bit, also Startools, and I've done things the hard way in Photoshop, but I prefer SIRIL by quite some margin.  Its UI could perhaps be more intuitive, and it took me a while to work out how to filter based on quality (my approach is to run the one shot processing script, then directly load - i.e. double click on - the calibrated sequence file and restack with a quality filter, and other settings of my choice).  The whole end to end process takes minutes.  SIRIL also happens to be free…

I learned it right at the start of my adventure into this wonderful hobbit

You use Hobbits for your workflow, Brian? Maybe I should try this too. My pictures really need some improvement. ;-)

You are telling me into my second dive into PI. ;-)

There are just two additional problems: 

Books sometimes refer to an older version of the program and the look of the new version is completely different or some parts of the program do not exits any more at all. 

And a course would be nice. But as a non-native speaker it can become pretty hard to follow instructions into a new field in a foreign language. When I was still in research as a scientist I was exhausted after a day at an international symposium or meeting. 

But to get specific: Which are the best up-to-date ressources for somebody who wants to start into PI? Maybe you have some good suggestions for books, free videos, and paid courses - live or even better on video.

Well, I started long time ago (or what appears to be sooo long ago). I built my first CCD and then built my second one. At that time I learned to use IRIS and used it for well over 10 years. Then took a long hiatus fro the hobby, mostly because I hit the buffers and the hardware was letting me down. The last thing I did was buying PI. Last year, complicit the pandemic, I restarted the astrophotography jig and obviously looked at what new stuff was there. PI was hard to get the grips of mostly because of its exceedingly poor help system (polar opposite of IRIS and this one was and still is free!) and customer support not up to it. But eventually I got the handle of it, more or less (bugs and all). Not an easy program to learn. Before that I also looked at what else was out there so I gave DSS and APP a trial. I found DDS too basic and APP incomprehensible in terms of interface with even less in the way of self-learning than PI. So I let it drop and concentrated on PI. Then discovered (here, incidentally) SIRIL which I am kind of fond of and would be my preferred processing software didn't I invested so much in PI (time more than money). I'm waiting to see whether SIRIL will evolve into a kind of IRIS for the 21th century. If it does then probably I'll ditch PI and switch allegiance to free software once again. One thing I cannot stand is all this cottage industry of learning tools/books/courseware that seems to be sprouting over PAID programs. I mean I paid for it and expect to learn how to use by the example/help system embedded into the system not spend even more money in dubious quality books (which more often then not  are behind the curve with respect the actual program you have on your PC).

Out of the three, APP is I think the most straightforward. It can combine hundreds of pictures into a mosaic, and do so even with the standard registration routine if they all overlap (this is actually the default registration/stacking mode). Its gradient removal algorithm gives outstanding results and its local normalization and background neutralization algorithms are better than PixInsight. Its digital development processing will stretch your image in very good ways.

Like I said, you can probably do the same in PixInsight or better but it takes a lot of effort, including trial and error. The only thing that you cannot do in PixInsight is its advanced debayering that gives very nice colors and relatively fewer artifacts compared to whatever PixInsight has to offer.

What I don't like in APP is the inability to save your session. You cannot register your images, shut down the computer and continue the next day from where you left off. The moment you close APP, everything is forgotten and you need to start the next time from scratch. This is not much fun if you want to try out different parameters and the integration takes hours, you might well end up with your computer on for three weeks. 


Also, it has no concept of noise reduction. Realistically, unless your noise levels are extremely low, you must save the stretched image as 16 bit TIFF or something and apply at least noise reduction using a third image processing program (but a "normal" one like Gimp or Darktable). 

But it delivers very, very good results out of the box at those parts of the workflow that it can handle and for the most part it is fire and forget. Definitely worth the inability to save a session and the need for a third program to apply NR, in my opinion.

My comparison DSS, MaximDL, Pixinsighr 1.8
https://astropolis.pl/topic/68949-preprocessing-kalibracja-alignacja-stackowanie-dss-vs-maximdl-vs-pixinsight/?do=findComment&comment=784992

I used all three - I do not claim to be an expert in either software, but I do prefer APP. PixInsight is extremely powerful but one needs to know a lot to properly use it and draw the benefits from all the various features and settings. I must confess, that I am not experienced enough to  really like it. DSS is easy to use and for many targets, it gives acceptable results - as a starter. 
APP is my favourite - why? Well, it is for me at least, intuitive and logical. I can align and integrate subs, which with my knowledge of settings, do not work with PI. I was up to now never really disappointed by APP. What I must say is, that with PI the image quality enhancements are better accessible and one has many more options. As far as my knowledge reaches, I do calibration, aligment and integration with APP and image enhancement first with PI and then with Photoshop 20 or 21…

Both APP and PI are excellent.  I find one works better than the other depending on the object and what I am doing, e.g., narrowband or broadband.  Explore both!  Don’t limit yourself.