Quick questions about multi channel/session stacking (I hope)

No, mono and RGB at the same time is not (yet) possible. We also want to change that in a future version. For now it's best to split the RGB data into mono R, G and B and then load those in again together with the Ha mono data. These can then be normalized and stacked together after which the integrations can be loaded into the RGBCombine module.

Thanks. More questions though!

• Is this how to split the RGB channels - https://www.astropixelprocessor.com/community/main-forum/splitting-osc-channels/ ?
• Where is the RGB Combine module?
• What algorithm would I be using to process the R,G,B and Ha frames?

Thanks, Brendan

Posted by: @brendanc

Is this how to split the RGB channels - https://www.astropixelprocessor.com/community/main-forum/splitting-osc-channels/ /a> ?

Yes. Basically you need to load the RGB integration result as a light and do not load any other files (not lights nor calibration frames). Then go to tab 2 and press "create Masters & assign to Lights". That will basically do nothing. Then enable "split channels" a bit further down in tab 2 and press "SAVE (calibrated) Light frames" and you're done.

Posted by: @brendanc

Where is the RGB Combine module?

In the Tools tab. It is called "combine RGB".

Posted by: @brendanc

What algorithm would I be using to process the R,G,B and Ha frames?

In "combine RGB" you can select the HaRGB1 or HaRGB2 formula (not algorithm). Note that when you switch between the two, you need to click the "new formula" button.

Thank you. I would actually be bringing the channels separately into StarTools, not processing in APP (I just use APP for stacking) so I don't know how this would affect your advice. I'll see how I get on.

OK, so I haven't got around to figuring this out yet, but I just realised that I still don't really understand why there's a multi session feature, in which I can load up RGB and H-alpha, when it seems APP won't actually stack RGB and H-alpha? Just while I'm still working on this, what's the answer ie which algorithm will it be using in the RAW/FITS tab if I do this?

Posted by: @brendan

why there's a multi session feature

Multi-session processing needs to be used if you want to combine similar data (e.g. RGB only or H-Alpha only) from multiple nights where different calibration files (usually flats) are used. Also if you want to combine similar data (e.g. RGB only or H-Alpha only) from multiple telescopes because those will require different calibration files (in this case flats, darks and dark flats).

Posted by: @brendan

which algorithm will it be using in the RAW/FITS tab if I do this?

APP will always use the selected algorithm.

Hi, thanks for this, but I'm still puzzled. Not confused, puzzled, and I'm not really closer to having my question answered!

I have similar RGB and Ha data, with separate flats and associated calibration files.

I want to stack them together using Multi session, which has RGB and Ha options.

From Vincent's first response it sounded like RGB and Ha couldn't be processed together, and that I'd need to split out the RGB.

However, I still don't get, therefore, why multi session has RGB and Ha options, which clearly implies that I can process them together, and I don't need to split out the RGB. Am I right?

Also, given that there are different algorithms for RGB (Airy disk) and Ha, among others, my question is still, which one is used when processing RGB and Ha together?

If it's the one selected, then that would not make sense, because it would be used for processing different types of data using the wrong algorithm, so for example if I select Airy Disk, but I'm stacking RGB and Ha, then that would be incorrectly used for processing the Ha data.

Or (which would ideally be the case), does APP select the correct algorithm for the data, so that the RGB session is processed using the Airy disk algorithm, and the Ha session stacked using the Ha algorithm?

Thanks, Brendan

Posted by: @brendan

From Vincent's first response it sounded like RGB and Ha couldn't be processed together, and that I'd need to split out the RGB.

However, I still don't get, therefore, why multi session has RGB and Ha options, which clearly implies that I can process them together, and I don't need to split out the RGB. Am I right?

Yes and no. APP allows for loading the data together but processing them together will lead to errors. Mabula is working on a fix though.

Posted by: @brendan

Also, given that there are different algorithms for RGB (Airy disk) and Ha, among others, my question is still, which one is used when processing RGB and Ha together?

Like I said, the one chosen in tab 0 will be used. If you choose "Ha-OIII extract Ha" for instance then that will be used on all data.

Posted by: @brendan

If it's the one selected, then that would not make sense, because it would be used for processing different types of data using the wrong algorithm, so for example if I select Airy Disk, but I'm stacking RGB and Ha, then that would be incorrectly used for processing the Ha data.

Correct. This is exactly what APP does which is another reason to not mix data.

Posted by: @brendan

Or (which would ideally be the case), does APP select the correct algorithm for the data, so that the RGB session is processed using the Airy disk algorithm, and the Ha session stacked using the Ha algorithm?

It doesn't. Not sure if this is the intended behavior once Mabula has fixed the processing issue mentioned above.

OK, thanks. So, APP has the option to mix data, but I shouldn't mix data - and this is being fixed.

I think I get it now.

Any ideas when it'll be fixed?

@brendan Mabula told me this will be fixed in 1.084, which currently is in the make. It will also allow for mixing "pure" RGB and dual band filter data such that the RGB data get processed into an RGB image and the dual band data into separate Ha and OIII mono images.

OK, sorry, but I still don't get this.

I have two sets of data. One is RGB, the other is Ha.

My objective is to get stacks that I can pull into StarTools, which I use for my post-processing. For this, they need to be exactly registered and the same size. The only way I've found to do this is to use the multi-session option, using RGB and Ha - but apparently I shouldn't do this!

So, I've split the R, G and B out as above, and now I have R, G, B and Ha.

Now, if I don't use Multi-channel, how do I do this?

If I do use Multi-channel, then I'm back to where I started! In which case, generally, what's the story with Multi-channel, in terms of which algorithm to pick? If you're saying we can't mix data, then why have multi-channel at all?

Multi-sessions are there to be able to load in e.g. flats that are session specific together with the data for that session.

If you have your split data (which should be calibrated now) you can load them in as mono data (just set it to the standard settings APP has), you add tags to them (R, G, B, Ha) and then go to the normalize or register tab and "save registered/normalized frames". Then all that data will be registered with each other.

I'm referring to multi-channel, not multi-session.

I understand multi-session, and I thought I understand multi-channel, but I'm not so sure any more!

Anyway, I think I understand what you mean by tagging the data - assigning the data to the channel - so hopefully I can move on with this now.

Let me know if it indeed works out, otherwise we'll help you for sure!