UPDATE FROM MABULA
I have had a very rough 2 months unfortunately health wise. I was struck 3x in a row with bacterial infections. The second infection occurred after a routine hospital checkup and I became very sick. I had to rest and take a lot of antibiotics. Once I was recovering and restarting work 1 month ago, I again became sick. The infection was not yet gone, so even more antibiotics and rest was needed. Needless to say, it took a lot of my energy and I needed a lot of rest.
Finally, the infection is really gone and my energy is coming back step-by-step now and I have started work again. I am terribly sorry to have kept you waiting for support. I will address all outstanding questions and e-mails step-by step and will be on the forum daily from today.
APP 2.0.0-beta47 will come soon as well, a lot of work was already completed before I became very ill, so the release is also nearly ready for you. Beta47 will be much faster actually. Many workflows will be more than 2x faster, mosaics can even be 10x faster than before because registration really received a major boost... all compared to beta46. Before I release it, I will make sure that everything is working properly and then I will release it.
I thought this topic might be helpful for a number of people in my situation: learned some but need more understanding to get better results.
I’ve gotten to the point that when I look at my data I get to the normalize step and stop. Then I do an analytical graph for star shape, quality, and RMS. I throw out some data based on some arbitrary threshold. It definitely helps but not enough.
When evaluating the subs, what should I really be looking for? Like, what star shape is generally considered good? What is a bad RMS? What rules of thumb should I use to throw out data?
I know the obvious ones like when tracking is bad and other obvious anomalies. I’m wanting to understand what I’m looking at better, while not being a data scientist, so I can more readily throw out bad data that may be giving me undesired results.
Thanks!
It may be a good idea to load say 5 images with very different quality, then do a bit of processing and check the numbers. This may give you a good idea already, what is good or bad is a bit of a setup specific thing in astrophotography so there is no hard boundary there. A registration RMS closer to 0 is better, FWHM can differ for different setups. The quality score is based on the analysis you do at each step of the process and it accounts for things like star shape. Maybe focussing on that is good enough already?
In general, never throw away data with a satellite trail, clouds and stuff I would throw away manually.
In my case I have hundreds and hundreds of subs so looking through a handful won’t help too much. I do go through them all to throw out obvious bad ones. I was thinking more along the lines of what star roundness is considered acceptable by people, or other measures that people look for.
I do look at the FWHM but there’s no way to sort on that measure so it’s more challenging to throw out bad ones when you’re dealing with several or more hundred.
I have seen in other tools that people have come up with various formulas to help them automatically select the good frames. I don’t understand the formulas entirely (I’m rather new at this plus I’m not a mathematician) but I was curious what people who use APP look at to weed out bad frames that could scale.
It's an interesting topic - this is one area where I feel APP could do with some work - I've found the stats often bear little relationship to the actual sub (especially where it's cloudy, they often have better scores than clear skies!) - so I have found the only way to do it reliably is to go through each by hand - which can be slow and somewhat tedious.
It's not so bad that I would start using an alternative tool, so it cant be THAT bad! (but could be improved)
