StarTools

Hi all,

First of all best wishes to everyone for 2023! Hopefully it will be a joyfull astro year...

IC 434, this time i used the NBaccent module, together with the Ha in the Luminance channel. I putted also the Ha in the NBaccent module. Don't know if it supposed to be like that but it came out well. In the compose module, i used the setting L+Synthetic L From RGB,RGB and not L,RGB.

No pink troubles. I used GIMP afterwards , just a bit not much, to adjust a bit the saturation and some other minor tweaks.

About just over 3 hours of Ha and 3 hours of color. Not that much integration time, seems that the horsehead is very bright.

.

Happy New Year to everyone. All the best for 2023.


That looks great, Freddy And even Alnitak is well under control. (BTW: Alnitak shows that inner ring we discussed here some weeks ago. But it does not annoy at all.) There are some artefacts visible around the brighter stars, but I think that's due to JPEG compression. And I like the overall colouring

fmeireso wrote: ↑Sun Jan 01, 2023 9:18 am I putted also the Ha in the NBaccent module. Don't know if it supposed to be like that

I'm not sure, but I would say 'no'. But as you said, it came out (very) well, and I guess that's the most important point for you. I'm still wondering, if your recurring colour balance/composition issues are due to weak red sensitivity of your unmodded DSLR.

And again, I cannot believe that there have been clear skies in Belgium lately . I haven't seen the stars for weeks

Best regards, Dietmar.

Hi Freddy,

Happy New Year to you and everyone also!

I think that's a fine looking horsey. I say we name him Traveler. Fight on!

Dietmar has a good point on the unmodded DSLR for RGB, you will be limited in your deep red capture, and thus correspondingly R's contribution to the luminance will be weak. I am unsure if our new Compose options (some are for DSLR, normal and modded) tweak anything in the background to account for that? TBD as we learn more about 1.9 I suppose.

However, by putting your Ha into L, I think some of that problem is ameliorated, by using L + Synth L. Hmm that's the same as I was doing on your Heart, Freddy, and perhaps that's why my images of your data were coming out different. Did you use L, RGB on that one?

But here I think it makes good sense, as the L + Synth L will, absent other changes to the exposure sliders, create the working luminance as 50/50 between your Ha and RGB. Now, we could probably look at QE and filter graphs and come up with a more mathematically "correct" blend, but I say probably good enough for now!

Now, once you get to NB Accent, having it be the same file that was already used for (part of) luminance, things might be a little different than one would typically "narrowband accent," but could probably be addressed. At least in 1.8 NBA. I'm still trying to wrap my head around the new NBA controls in 1.9. But 1.8 has fully alterable lum and color contributions, so I think you should be able to dial things in quite nicely for the added Ha oomph to the red. Maybe a bit weaker on L contribution, since it's already combined, and strong on the color for red boost. Pure red most likely too!

Good job Freddy!

And if Alnitak is too artifacted (did you use 1.8 or 1.9?), we could say it's the first Airy disk ring , or I suppose that could be a valid candidate for selective repair of clear artifacting. Personally I'm a lot less hard-line in that area, if something can be reasonably concluded to be artifacting, especially when it comes to oversaturated star cores.

Mike in Rancho wrote: ↑Sun Jan 01, 2023 7:18 pm

However, by putting your Ha into L, I think some of that problem is ameliorated, by using L + Synth L. Hmm that's the same as I was doing on your Heart, Freddy, and perhaps that's why my images of your data were coming out different. Did you use L, RGB on that one?

But here I think it makes good sense, as the L + Synth L will, absent other changes to the exposure sliders, create the working luminance as 50/50 between your Ha and RGB. Now, we could probably look at QE and filter graphs and come up with a more mathematically "correct" blend, but I say probably good enough for now!

Yes , on the heart i used L,RGB but not on the horsehead. Well i tried it but i got that pink again, so i abandon it eventually...

decay wrote: ↑Sun Jan 01, 2023 6:41 pm .



And again, I cannot believe that there have been clear skies in Belgium lately . I haven't seen the stars for weeks

Best regards, Dietmar.

Well, the clear night i took color was unexpected. I allready had settled down before the television set with a beer or 2, then to realise it was becoming clear. I hesitated but eventually putted the gear out in the dark about 09:10 PM. When it was set up some clouds passed by and then clear till i broke up about 02:30 AM...

You never can trust the weatherforecast especially about clouds concerned. Sat 24 is the best way to go, but that you can only evaluate a couple of hours on before hand....

fmeireso wrote: ↑Mon Jan 02, 2023 7:48 am I allready had settled down before the television set with a beer or 2, then to realise it was becoming clear. I hesitated but eventually putted the gear out in the dark about 09:10 PM.

Very impressive, Freddy . There's no way to get me up if I already settled down and had two beers. (And for sure it was too much of one or two beers and TV the last two weeks. )

fmeireso wrote: ↑Mon Jan 02, 2023 7:48 am Sat 24 is the best way to go

Thanks for pointing out, I'll give that a try. This company is from the Netherlands, so probably they will have a good forecast for Western Germany as well.

Best regards, Dietmar.

decay wrote: ↑Sun Jan 01, 2023 6:41 pm .

if your recurring colour balance/composition issues are due to weak red sensitivity of your unmodded DSLR.


Best regards, Dietmar.

Actually i don't really think that that is the culprit. Taking L, ha with a mono cam either DSLR or in most cases a mono astrocam and combined with an OSC is a known technique. The Elf uses it, from which i took over the 'know how'. He seems to have not much issues with color allthough he uses PI.

But the question remains, how much time does the OSC needs.Even the Elf could not really answer that. One sees that his OSC shots are always less in time the his Ha shots. So are mine but maybe one should perhaps take more time or at least as much time as the Ha shots...after all the more integration time the more nebulosity and probably color. I did not have too much of a problem with color for the Horsehead, the object is bright and well my OSC time is about as large as my time i took for the Ha shots..

Dunno, it is just a thought.

Next target is the Pacman, i allready have the Ha, which came out nice, just waiting now for a clear moonless night for the color shot.

Freddy, HaRGB with Ha as L may be a known technique, but also has known limitations which have been discussed quite a few times. And in most cases even then it's done with discrete RGB full astrocam, OSC astrocam, or modded DSLR. That stock filter in front of your Canon sensor is very real, ramps down the red response quite strongly (usually down to about 20-25% by the time you hit the Ha band), and so it will never have the same response relative to the other channels or even lower nanometer red. More integration will just clean up the SNR, and thus perhaps you can stretch more (but you will be stretching everything else too), but Ha for coloring purposes will remain weak in comparison.

Whether you find that "appropriate" as a match to human eye response gets into your theory of AP and whether you feel it is correct. But, by using Ha as L you have kind of strayed off anyway...

Remember too that the Elf uses PI, with a workflow that could (I'm guessing) probably be considered be a bit artistically manipulative, and (from the big practice thread) Elf even admitted that he is doing AP more from an artistic standpoint than he is trying to hold to a more true-to-the-data workflow. And that's totally fine, his images look great.

I think you are just going to have to experiment and practice, and the needed workflow may differ per target. You've already tried HaRGB with various utilizations of NB Accent. You could try some actual L from the mono cam, for LRGBHa? Chrominance from the umodded camera will still be what it is, but I wonder if that could help boost the visual spectrum stuff somewhat. There's also the possibility of blending some percentage of the Ha into your red channel beforehand and recompositing, but again what to do/use for L and/or NB Accent, you'd just have to try those out. And at what point too many complicated combinations edges too far into the artistic realm, I don't know.

Mike in Rancho wrote: ↑Tue Jan 03, 2023 5:41 pm And in most cases even then it's done with discrete RGB full astrocam, OSC astrocam, or modded DSLR. That stock filter in front of your Canon sensor is very real, ramps down the red response quite strongly (usually down to about 20-25% by the time you hit the Ha band), and so it will never have the same response relative to the other channels or even lower nanometer red. More integration will just clean up the SNR, and thus perhaps you can stretch more (but you will be stretching everything else too), but Ha for coloring purposes will remain weak in comparison.

Freddy, Mike - that's almost exactly what I wanted to say. I should have written one or two more sentences and I was not aware of the (likely) misunderstanding / missing piece of information for Freddy until I read Freddy's last post. But Mike was faster clearing this up. Thank you.

Best regards, Dietmar.

Ok guys, thanks for your insights.

So in any case, if i understood well, it should be better to take color with an modded cam?

I might consider that. But as Mike suggested, i will experiment further. One of the reasons i like the LRGB approach is the fact that you only need 2 sessions to produce an image of emission nebulae. I also think that mono cams seems the way to go but if you use only a mono cam and color filters you need at least 4 sessions, L R,G,B. In a country like Belgium that gives troubles, due to the fact that clear nights are rare...j

Maybe some day i'll switch to dedicated astrocams but for now i feel a bit unsecure about it. I believe the filesize of the subs they produce are way bigger then a DSLR sub. Which might have it consequences in capturing then and processing them..dunno, guess i got some reading to do about that if i ever want to jump to that direction...