RGB combine: multiply normalization problem with SII (OSC)

Did you, by any chance, use different calibration data compared to the previous sessions which did work? Any other differences?

@vincent-mod

What do you mean with "previous sessions"?

For the Ha and O3 done with l-extreme filter I used just its own flat.

For SII filter I obviously took its own different flat.

All other calibration files, that is darks, BPM and dark-flats are the same for both filters...

You mention "Then, as I normally do with great success..", which I interpreted as you having done this before in previous sessions. If that would be the case, I wonder about any possible differences in between the sessions. So the flats are those that are, I think I'd need to see some of the data to get a better idea of what may be going on.

@vincent-mod

Yes, Vincent, with this phrase I intended that I have been using APP for more than 1 year with great success doing exactly what I did with these SII and l-extreme filter: took flats in the same way, calibrating and stacking in the same way, etc....This same way gave me good results as usual...So, let me know please if you, Wouter and maybe Mabula too can have a look at my files and which files you would prefer. Thanks a lot

We certainly are going to look at it yes, just asking some background information. 😉

Please upload some of the relevant subs to our server;

Go to https://upload.astropixelprocessor.com and use upload1 as username and upload1 as password.

Create a directory named “ippiu-SIIProblem” and upload in there. Thank you!

@vincent-mod

https://upload.astropixelprocessor.com/index.cgi?launchApp=SYNO.SDS.App.FileStation3.Instance&launchParam=openfile%3D%252FAstroPixelProcessorUpload%252Fippiu-SII%2520Problem%252F

Thanks so much

No worries. Please allow for a day of processing as I have more data to go through.

Ok, starting to have a look. I'm not sure the flats for SII are correct, but I'm not sure why not just yet. I did do a new integration of the 3 SII subs you provided, without using the flats. I re-normalized them with the other already processed Ha and O3 files and did a combine-RGB, multiply-scale:

So I think the flats are the issue. Could you test this at your end as well? If that works, we can have a look at what exactly is going on. I might have to discuss that with Mabula as well.

@vincent-mod

I reproccessed from the beginning SII files with no flat calibration (just dark and BPM), re-normalize everything (Ha, O3 and SII) and saved everything and nothing has changed (used beta2 version)...

Let me know please if you need some more files...

Thanks for your patience

Well that is very strange then. And if you do it exactly like I did in my explanation above? So using the already created Ha and OIII? Maybe there is some data in another light frame of the SII dataset that is causing this? Neutralize background also doesn't help I guess then (on the right menu)?

@vincent-mod

I tried also normalize SII new (no flat) with exactly the already calibrated Ha and O3 I had uploaded. So I used SII_noflat-reg-norm  O3-reg-norm-reg-norm  Ha-reg-norm-reg-norm  right??? And nothing changed.... 🤔 

Yes, I also normalized them to each other again. Nothing more really, using beta 2. So I'm a bit stumped as to why we have different results here then. I guess you selected SHO1 and pressed "new formula" before choosing to recalculate right? Did you try it with just the 3 subs you send to us?

Posted by: @vincent-mod

Maybe there is some data in another light frame of the SII dataset that is causing this? Neutralize background also doesn't help I guess then (on the right menu)?

I tried using exactly your 3 SII lights with darks and BPM no flat...re-normalize everything...and...boom...everything works as you...

So you are right: there must be some weird SII light frame around my 34 total frames...

I tried using exactly your 3 SII lights with flats, darks and BPM...re-normalize everything from beginning icluded Ha and O3...etc...And everything works normally...

So master flat (and flats) isn't the problem. Also in this case, the problem must be related with some single light that causes this weird behaviour...Difficult to find out...

Yes it is, I've never seen an issue like this I have to say. So if you do find it, I'd find it very interesting to check that sub out as well. To be able to find what is going on, I would advise to process this in groups of maybe 10 subs each time and then hone in on the data that is causing it.

@vincent-mod

I divided 34 files into 3 row, 11 lights each...

Every single row shows the problem, so I can't imagine what and where the problem could be...

It's almost impossibile to isolate and find out the problem among 34 lights...Maybe could it be related with flats? With 3 lights randomly? How? With what? 

I think I leave this mystery to god...I can't spend hours to try to find out a so difficult problem...

So I'm going to use add or add-scale. Between add and multiply, as I read from the help file, the difference is marginal, I think...

Thanks for your help

Very strange indeed, sorry for the trouble and not being able to identify it at the moment. If it happens again, I will ask for your entire data set to dig into this.