East Veil - 1st try at adding OSC stars to duoband

[dx_ron]

I think it is OK, though I am sure I will look at it later and see only weird artifacts.

AT65EDQ on a CEM40, guiding with 50mm guide scope + ASI120mm-mini
Risingcam IMX571C at 0°C
KStars/Ekos; stacked in Siril
duoband: Antlia AlpT filter. 110x300s (9.2 hours) home (high Bortle 6)
OSC: uv/ir cut, 63x60s (you know, an hour) at club dark site (high Bortle 4)

E_Veil_NB+OSC-KY_stars_500k.jpg (499.87 KiB) Viewed 1647 times

Maybe not the wisest choice to try this first with such a dense star field

[Mike in Rancho]

Looks good!

The RGB star blend-in has been tough for me, thus I haven't tried it much. Considering the density of the star field here, I think you pulled it off. I can maybe see a couple purple stars within the East Veil itself?

Background in on the bright side for me, but that's just personal taste. You were able to show the leading edge of Ha, which is cool.

I too have just stuck with 0C for the 2600, doubt I'll bother changing. What gain have you been using? I am not sure if it follows the same numbers that ZWO uses, though as a 571 it must still have the same two settings that would tend to be best. I've actually stuck with gain 0, whether broadband or narrowband, in order to have the full well. AT65 might be fine, but I think with a 6" at f/3.8 I might burn out stars too quickly at gain 100 (the point where that shift kicks in).

[dx_ron]

Thanks, Mike

The Touptek driver works a bit differently. Gain goes from 100 - 40000 (maybe higher, but no one is going to bother with the high end). Then there is a separate LCG/HCG switch. Gain 100 (LCG) is the same as ZWO gain 0. With the Risingcam, though, you can also set gain 100 (HCG). The numbers look ever so slightly better than ZWO gain 100, but the difference is probably not meaningful.

For duoband I've been using 5 minutes at gain 100 (HCG), and 1 minute at gain 100 (LCG) for OSC.

The time is approaching when I will have to start using -10° or maybe lower. You wouldn't understand, but it has to do with winter...

[Stefan B]

Maybe not the wisest choice to try this first with such a dense star field

Looks pretty cool! Have tried this kind of combination myself several times but it's not so easy...especially when dealing with data from a Newtonian. I always failed at including the star spikes in the mask So I omitted the bright stars with spikes and only blended the colors for non spiked stars. But yours seems to have worked on bright and dim stars. Did you use layers, a star mask and the "color of fg" blend mode?

The Touptek driver works a bit differently. Gain goes from 100 - 40000 (maybe higher, but no one is going to bother with the high end). Then there is a separate LCG/HCG switch. Gain 100 (LCG) is the same as ZWO gain 0. With the Risingcam, though, you can also set gain 100 (HCG). The numbers look ever so slightly better than ZWO gain 100, but the difference is probably not meaningful.

Do you have a reference for this? I reasoned it should be this way but couldn't find anything. I have a Omegon veTEC571C which is a rebranded Touptek clone... Most of the time I forget to switch from the default HCG to LCG for my broadband imaging. On the other hand gain 100 HCG seemed to work well for NB and broadband. Are there any figures on how much of the full well I am losing this way?

Regards
Stefan

[Mike in Rancho]

The time is approaching when I will have to start using -10° or maybe lower. You wouldn't understand, but it has to do with winter...

I've been in the mountains, which is where cold and snow belongs. I can even see it way up there from my backyard.

Now, I did spend my first 13 years in the Chicago burbs - but that's too long ago to remember.

[dx_ron]

Did you use layers, a star mask and the "color of fg" blend mode?

Hi Stefan. Yes, I followed the simple suggestions Ivo posted recently. I actually made the star mask from the stretched OSC image. That worked better than I expected it to - just had to do one 'shrink' and it looked OK.

Do you have a reference for this? I reasoned it should be this way but couldn't find anything. I have a Omegon veTEC571C which is a rebranded Touptek clone... Most of the time I forget to switch from the default HCG to LCG for my broadband imaging. On the other hand gain 100 HCG seemed to work well for NB and broadband. Are there any figures on how much of the full well I am losing this way?

There's a sensor analysis on CN: https://www.cloudynights.com/topic/7472 ... p=10755632
While it looks like you lose a lot of FWC (52000 vs only 17000 for HCG), the lower read noise means you only lose about 1/2 stop worth of dynamic range.

In the INDI driver you could set it up to autoswitch at, say gain 101 if you wanted to. So gain 100 would be LCG and anything else would be HCG.

[Stefan B]

Yes, I followed the simple suggestions Ivo posted recently. I actually made the star mask from the stretched OSC image. That worked better than I expected it to - just had to do one 'shrink' and it looked OK.

Sounds great! I've always made the star mask from the NB image since the stars tend to be tighter and I didn't want to color the stars' rather black surroundings with the color of the broadband stars. But I am sure it also works the other way round. Your image is proof

There's a sensor analysis on CN: https://www.cloudynights.com/topic/7472 ... p=10755632
While it looks like you lose a lot of FWC (52000 vs only 17000 for HCG), the lower read noise means you only lose about 1/2 stop worth of dynamic range.

Thanks so much! Exactly the information I was looking for Great to know I don't lose a lot of DR although it's a lot of FWC.

In the INDI driver you could set it up to autoswitch at, say gain 101 if you wanted to. So gain 100 would be LCG and anything else would be HCG.

Probably doesn't work with the native NINA driver. I'll have to look...

Regards
Stefan

[dx_ron]

If you do see a gain threshold setting, you should also look for 'HCG:LCG ratio' - though I don't remember what the default setting was (you can only override the auto-switching by setting the ratio to 1:1 and I don't recal what the default setting was).

[dx_ron]

E_Veil_NB+OSC-KY_stars_v2_500k.jpg (495.64 KiB) Viewed 1557 times

I went back and re-processed to try to make a darker background for Mike . But also to implement a couple of things from current discussions.
I knew the crop I wanted, having had the foresight to write down the crop settings to make the duo and OSC files match. Then I binned all the way down to what was to be the final dimensions (1600x1051). Wipe, then autodev. There's not much to be done to push the background darker in autodev - there's just so darn many stars that there is hardly any real background. Making it darker would have to wait for superstructure. Contrast -> HDR, no sharpening. Then for decon I just used only synthetic. As Ivo suggested, with the wide field synthetic was quite effective (how does he know all this stuff??).
Color, then SS isolate, with strength bumped up to 125%. This time I made a star mask from the duoband image, using Alt-Stars + one Shrink. NR (not much visible effect). Finally Layer for the color-morphing. Either way of making the star mask seemed effective.

It doesn't stand up to much zooming in. Partly, I think, due to the substantial bin factor (things get blocky) and partly due to the soft mottling introduced by SS.

I might collect some more duoband data this week, though my pesky Bubble-Claw project needs more data more.

[Mike in Rancho]

Nice! It's a wee bit darker. I actually think it reveals those outer wisps around the veil even better. Trying to stretch those out and also not bring along the background which could mottle is maybe not so easy.

Contrast can squeeze shadow dynamic range if AutoDev isn't up to it.

But the veil looks good in both, especially clicked to make it bigger where you can see the tendrils better.

Did you make the AltStars mask during tracking and it carried over to Layer? I think I tried saving an AltStar mask once, figuring I could then use it outside tracking, and it didn't work. I forgot about hitting a shrink though when I was just playing with AltStar earlier.

It seems like maybe two stars got a bit of shape change between the two versions though - one kind of oval towards the top center-left, and another like a plus symbol towards the left edge? Both seem pretty white.

Well, I've been getting nothing this month, not even a peek at M31 for the monthly. But at least that Kay finally spawned a hellacious thunderstorm for a good rain dump on Sunday. Might have clear skies by the weekend?