Setting correct exposure for RGB with 314L+

[AndyUK]

Having just attempted an LRGB image of NGC7023 @ f5.3 (MN190), despite capturing between 4hrs and 8hrs each RGB, I was disappointed to get so much noise, and one of the comments put forward was that the exposures weren't long enough (only 4 mins used)

When I first got the 314L+, I started using 10mins, and the result was totally saturated stars which meant they almost all came out white in post-processing. So I ramped it back to 4 mins to desaturate the stars... but this is obviously at the expense of capturing enough colour in the object itself (which I guess is resulting in the noise when I then stretch the RGB to try to bring it all out)

Is there a happy medium, or is it trial and error, or maybe someone can advise what exposures they use (as standard) when mono RGB imaging with their 314L+? Or am I barking up the wrong tree here and should I be accepting that if I want colour in both stars and object then I need two sets of exposures and use a star-layer to superimpose the lesser exposed stars on top of the longer exposed object? If so, as a guideline, what exposures should be used for a star layer...?

[Freff]

Hi Andy.

I think I read that using shorter exposure in luminance and maybe less, reduces the over saturation.

I could be totally wrong but it rings a bell.

[AndyUK]

Cheers Tony - That might explain it... The Luminance I took was 10 mins (to get the detail). Perhaps I'd better go back to the same exposure for all channels and then consider creating a separate star layer with lower exposure subs if necessary...

[swag72]

I'm sure that those who create the images that we both drool over, and the same people that are responsible for our lack of imaging confidence (!!!! - Oh yes they ARE!!), don't use a seperate star layer for the star colour. But I guess that there's more than one way to skin a cat. I've said before that I am sure that this is the way to go for me - Nice long exposures and forget the star colour then add in a few nice short RGB exposures, with previous exposure stars made smaller so it all fits together like a well fitting glove.

That's the theory anyway - Just need the skills to do it!!

[AndyUK]

Cheers Sara - I think that was our Plan A wasn't it... I must have forgotten it... I'm beginning to think that mono RGB might perhaps be too complicated for me after all !

[Black Knight]

I have a 314l+ arriving in a few days hopefully, so any guidance on exposure times is very handy - RGB and narrowband .

Sent from my mobile using Tapatalk, so please excuse the speeling and granma!

[AndyUK]

I'm obviously (and certainly) not the best person to ask about RGB(!), but for Narrowband I generally go for at least 10mins and sometimes 15 to 20 mins (depending on subject).

However, if you look about (even on SGL), you'll also see some people imaging with 30 mins exposures - If the guiding seems to be going well, I go long... but I don't yet have the courage to go up to 30 mins!

PS The 314L+ is a great little camera

[Black Knight]

Yep, I'm looking forward to mine immensely. I've wanted one for ages, and FLO's clearance offer was just the thing I needed to tip me over the edge!

Sent from my mobile using Tapatalk, so please excuse the speeling and granma!

[swag72]

I won't do anything less than 20 min subs for narrowband, aiming for longer if the guiding is going OK and I'm feeling VERY brave!

I generally go for 10min subs for RGB, and am considering a set of shorter exposures for star colour alone.

[AndyUK]

Sara's braver than me, but the message is the same - With narrowband go as long as you feel comfortable with...!

[Black Knight]

Right. 10-30 minutes NB coming up!

[ollypenrice]

I rarely do a separate star layer but I do routinely remove the Luminance from the stars. If they are on plain background its easy. Erase the lot.

Olly

[AndyUK]

Cheers Olly - Sorry - I'm probably being a bit obtuse (I'd like to blame lack of sleep caused by too many clear nights on the trot whilst still having to get up for work... but that wouldn't be totally true!)

Does this mean you'd process the full image with luminance, then select all stars and delete them (if feasible), and then overlay an image processed without luminance, select all stars and overlay them... Is that (generally) right...? Or is LAB mode involved here somewhere?

And if it's not a star field, instead of deleting the stars on the LRGB image, would you select the stars on the LRGB image and then blend in the RGB stars across the top?

I'm getting myself completely wound up in circles about star colour (sorry!)

[martin_h]

Cheers Olly - Sorry - I'm probably being a bit obtuse (I'd like to blame lack of sleep caused by too many clear nights on the trot whilst still having to get up for work... but that wouldn't be totally true!)

Does this mean you'd process the full image with luminance, then select all stars and delete them (if feasible), and then overlay an image processed without luminance, select all stars and overlay them... Is that (generally) right...? Or is LAB mode involved here somewhere?

And if it's not a star field, instead of deleting the stars on the LRGB image, would you select the stars on the LRGB image and then blend in the RGB stars across the top?

I'm getting myself completely wound up in circles about star colour (sorry!)

I think what Olly is saying is to process the RGB and the lum then delete the stars from the lum then combine, that way you get the lum AND RGB for the main object but only the RGB for the stars.

[swag72]

I think what Olly is saying is to process the RGB and the lum then delete the stars from the lum then combine, that way you get the lum AND RGB for the main object but only the RGB for the stars.

Now that makes sense - but it leads me to another question (Which I'm sure is on the tip of Andy's tongue!! )

If you remove the stars from the luminance, so that only RGB is in the stars, you would still end up with over saturated white stars if you pushed the RGB exposures long enogh to get good colour in the target area.

I will go for 600s in RGB, but that does give me some over saturation, so should I be lowering that then? Where do you put the measure of best exposure time? To date I have been allowing some maximum saturation on the brightest stars in the field of view, but you can never claw that colour back unless you exposure seperately for a star layer.

Sorry, am I going off the point there? Or just too fixated on star colour, which I've never been able to achieve in any case

[derrickf]

<SNIP> When I first got the 314L+, I started using 10mins, and the result was totally saturated stars which meant they almost all came out white in post-processing. So I ramped it back to 4 mins to desaturate the stars... but this is obviously at the expense of capturing enough colour in the object itself (which I guess is resulting in the noise when I then stretch the RGB to try to bring it all out)

Is there a happy medium, or is it trial and error, or maybe someone can advise what exposures they use (as standard) when mono RGB imaging with their 314L+? ...

Andy,

when imaging an object like the Iris Nebula which has a low surface brightness and high dynamic range in the object itself and is surrounded by some pretty bright stars, it can often be advantageous to image in RGB and then create a synthetic luminance. The individual colour planes will contain far fewer saturated stars so you can reach the best compromise between DSO depth and star colour.

FWIW ISTR that when the LRGB technique was originally developed independently by Kunihiko Okano and Robert Dalby, Dalby's variant of the method advocated the use of synthetic luminance rather than using an L filter for the best match between L and RGB channels.

HTH

Derrick

[AndyUK]

Thanks Martin - That makes a lot of sense to me also... and thanks Sara - That was indeed my next question...

Derrick - It's funny you should mention using false luminance from unbinned RGB as that was where I first started before then trying LRGB!

So, in the example of the Iris, would you advocate going for 600s RGB's (even though that might saturate one of more of the channels?) or is it a case of trial and error to see which filter saturates first (probably blue for the Iris) and then set that as the max exposure? (Ignoring G2V calibration for a moment)

[derrickf]

Thanks Martin -

Derrick - It's funny you should mention using false luminance from unbinned RGB as that was where I first started before then trying LRGB!

So, in the example of the Iris, would you advocate going for 600s RGB's (even though that might saturate one of more of the channels?) or is it a case of trial and error to see which filter saturates first (probably blue for the Iris) and then set that as the max exposure? (Ignoring G2V calibration for a moment)

Providing you are not saturating the core of the DSO or too many stars I would go with 600s (unbinned). If there is still severe saturation (unlikely IMHO) you could back off as you suggest. You can always G2V correct later if you wish.

Derrick