StarTools 1.9 Beta

[decay]

Addendum: Having a look at my lasted data set, there doesn't seem to be any overshooting in saturated stars:

2023-10-03 13_46_38-StarTools.jpg (19.32 KiB) Viewed 126756 times

BTW: Nice dataset, Steve

[alacant]

Hi and thanks for the tip. Lower shadow detail does indeed mitigate the effect.

I suppose that in the end, it's only a few pixels in a few stars to correct so NP in painting manually.
Cheers and clear skies averyone.
Steve

Oh, and yeah. I always shoot star clusters at and around full moon, so wipe has a fair bit of work to do on the gradient. It may well be connected to that too.

[decay]

wipe has a fair bit of work to do on the gradient. It may well be connected to that too.

Hmm I'm quite sure that the problem already lies in the dataset itself. The star profile showing the overshooting is from your original FITS data, without any processing (I forgot to mention that). I've no idea what could be causing this. Have you changed any settings in your stacker or changed your workflow? Pre-processing of individual subs or something?

Good luck, Dietmar.

[Mike in Rancho]

Hey Dietmar, just wondering what program you are using that shows a graph of the stellar profiles like that?

An another note (and pending Ivo's hard-to-find free time of course ), I think the SVD deringing may still need some tweaking before it's ready for prime time.

Most of my own data, not that I've had much, has actually been quite good coming out of SVD, except for ringing around scattered super tiny stars against nebulosity. But when working with other user dataset last night, I can still get some medium stars that just won't dering no matter what. My belief, which could be off base, is that a few of these items all relate to a deringing support mask, or whatever other identification of stars to dering (or non-star areas not to dering) ST is using.

[decay]

Nothing special Mike, that's just Fitswork. Latest version is from 2015 and I think that's why Ivo calls it legacy
https://www.fitswork.de/software/softw_en.php
I still often use it to have a first look at my stacks, to see if everything is OK. I guess since now I'm using ASTAP I could use that viewer as well.

I totally agree about SVD. Most times it works like a charm for me, but sometimes there are medium sized stars that show strong ringing I can't get rid of. Linearity (you know, UNKNOWN ) has to be reduced to single-digit values to work and the other two related sliders do not work at all with my data sets. I guess it all strongly depends on the characteristics of the respective dataset. And yeah, I'm sure Ivo is aware of this and yeah, we all have to respect that sometimes there may be other important things in life (and even if it's just the nuts job-that-pays-the-bill ).

There are too still some pending translations and clean up to be done, but we (Jochen & Carles keep track of that) decided to wait until Ivo finds some free time to go on. It's already great as it is, isn't it?

Best regards, Dietmar.

[Mike in Rancho]

Thanks Dietmar. I actually have Fitswork 4. Fiddled with it a couple years ago maybe, didn't know it had a stellar profile graph.

And yeah you are probably right, all this stuff will be jotted down or caught up to in due time. I didn't realize finishing the new multi-lingual was still an open item too.

[alacant]

All, thanks so much for your interest and tips.
I suppose the moral of the story is to just go easy with the HDR. In fact, just go easy!

p2.jpg (550.73 KiB) Viewed 126661 times

Cheers and clear skies,
Steve

[alacant]

... almost forgot. No thread is complete without the ai starless version:

p3.jpg (5.26 KiB) Viewed 126660 times

[decay]

the moral of the story is to just go easy with the HDR

I already wondered about your intention to use HDR on a star cluster. But it is possible that you will have problems processing nebulae with HDR if this really is a general problem with your data sets ...

[decay]

No thread is complete without the ai starless version

Well done, Steve! Excellent! Now you can try to stretch the starless version as much as possible and then layer in your stars in a most subtle way