A question about color

[Heno]

For the first time I have tried to integrate LRGB+Ha+SII+OII into one image. I have spent numerous hours trying to get a descent looking result, and I know there is no right or wrong doing this, it's a matter of taste, but I still would like to hear how others would be doing this. Each color have one hour of data. This is the process I ended up with (the 7th time I tried). (And yes, I have read the manual).
I restacked Ha and SII into one dataset and used this as my NBaccent file. I does not look like ST is able to discern between Ha and SII color. Is that correct? This also gave me lower noise as I had twice the imaging time into one image.
I processed this as any other LRGB set, then added the NBaccent data at the end. Here we can choose between Ha/SII Bright red and Ha/SII Balmer series. I chose the Bright red and adjusted the brightness and color until I liked the result. Removed noise and saved it.
Then I wanted to add the OIII data. I could not find a way to use the NBaccent this time so I went with the previous way of doing it, processing the image as mono chrome, the take it into the compose module to add color to it. I found the I could make it dark blue (RGB/RGB) or I could choose L + syndetic L from RGB (Bi-color) which gave me a bright blue/green color which probably was not that far from OII. Still, I went with the dark blue and added this to the existing image using the Layer module. (I will try again with the other option).
I know there is many ways to skin a cat, and I would tike to hear which way others would do it, including Ivo if you can take the time. This is my final result.

[hixx]

Hi Heno,
Your result looks pretty nice!
I'd say try to load O-III as G and combined stack with Ha/S2 as R and create an artificial Duoband NBA stack this way.Save linear and reload as NBA Compose/NBA will split this into red &green channels automatically . in NBA there are settings for Duoband filters which you could use to render bright red &cyan, or pink & green for example. To Your question: no, NBA will treat both Ha an S2 as red inputs, hence does not distinguish between those, neither between H-beta and O3. (Both rendered as G or G&B=cyan)
Hope this helps,
Jochen

[Heno]

Jochen
Thank you so much.
I never thought about doing it the way you describe, but I will certainly try it out at first opportunity.

I'd say try to load O-III as G and combined stack with Ha/S2 as R and create an artificial Duoband NBA stack this way.Save linear

So I load OIII and G in compose module, keep and then just save? Is that what you mean when you say "save linear"? Likewise for Ha/SII + R.

Helge