StarTools

One of my next projects is using my Mono rig and capturing an SHO data set , capturing RGB stars on a bright nebula like M42 and trying to process it as a visual spectrum or broadband result.
I’m imaging under Bortle 8 skies so my broadband RGB subs are limited to 30 to 60 sec subs, luminance is worse only 15 to 20 sec subs

I’ve seen a procedure using Startools a while ago but can’t seem to find it.
Any advice , procedures would be most helpful
Thanks
Martin

Not 100% sure about your request, Martin.
Maybe this discussion?

viewtopic.php?t=2245

Hope it helps.

Greetings from cloudy and cold Germany. Maybe 10 rainy weeks, as Freddy wrote and now temperature has dropped below 0 and it has snowed. But still cloudy

Dietmar.

Hi Dietmar,
Thanks for the link
I’ll try to follow Ivo and Guys procedures once I capture some good data
Stay warm , keep safe , clears skies will eventually come to Germany I’m sure.

Martin

Hi Martin,

I may be a little confused by the issue. Is it just "normal" RGB stars, captured separately, added to a NB image? Dietmar's link seems to instead be more about fully recoloring SHO to resemble RGB. Which seems to consist of just shoving Sii into the red to combine with Ha (as they are both red).



Otherwise, there's this older thread that some ruffian started a couple months ago: viewtopic.php?f=10&p=14941



And of course several more that can be found just by googling startools and then "rgb stars" in quotes. Freddy and Ron also do that technique with some frequency.

I tend to avoid it as I rarely get a blend-in that I like, especially across multiple star-size scales and if there is, as you say, a lot of bright nebulosity.

If capturing broadband only for star blend-in though, I would not bother with the L filter, and you might even consider dropping exposure times even more for the RGB filters - meaning very short, as saturating cores can impair the whole RGB star color project.

Hi Mike,
Thanks for your reply
Maybe mentioning stars had you confused
Dietmar’s link was what is I was looking for
I’m trying to process an SHO data set to mimic closely an LRGB or RGB image result.
I know it’s impossible to achieve that exact look, but Dietmar’s link led me to Guys User Notes in Compose and he has a procedure.
Imaging true broadband under Bortle 8 skies limits me to only 2 weeks a month ( New Moon period) and during our Summer months that generally limits you to 1 or 2 clear nights at best when the Moons not around. Narrowband I can image under a full moon no problem.
I will give M42 a go in true broadband using say 15 to 20 sec Lum subs and 30 sec RGB subs but that requires a lot of subs and memory (50MB file size each frame ).
This idea of NB data to process to look like RGB is just an experiment . It may be a waisted exercise ??

Cheers
Martin

Oops, sorry I got that wrong, Martin.

Ivo's answer is still sort of just a bicolor HOO though, though maybe more of a SHOO. But will be somewhat wavelength correct, in the same way a standard HOO is close to mapping emissions to their appropriate hues.

If we ever get full control of color mapping (ST 1.10 beta? ), one could conceivably make Ha a slightly lighter red, and SII a bit deeper, though query if the eye could tell the difference.

Now, if you mean going more full-artsy and just wanting to create something RGB colorful out of SHO data, there's the well-trodden PI/PS method of faking things up: http://www.starrywonders.com/bicolortechniquenew.html

Possibly could be done by coming up with some workflow using ST layer module in various steps, or taking some stuff into Gimp too.

I wouldn't necessarily shy away from the moon completely though, just when it's too close to target and can't be blocked from entering across the top of the OTA maybe. Especially for something as bright as M42.

The gigabytes do add up for sure, and will also require processing computing power. That said, for a similar SRN/time on target (considering how much of each color is let through to the sensor by the Bayer matrix), individual OSC files will be triple the size, and each has to go through debayering interpolation as well.