27 hours on M27

[DaveS]

5 minutes ago, andrew s said:

At this gain you get 5 electrons per ADU.  This level of quantisation noise may be your issue.  If your nowhere near saturation your throwing bits away. I think you would be better off taking more shorter exposures, to avoid saturation,  at say 1 electron per adu (gain 139).

Regards Andrew 

OK. It looks like I may have a clear moonless slot on Saturday, so I'll see what gain 139 looks like, but I don't have any Dark frames for that.

[x6gas]

5 minutes ago, DaveS said:

OK, single uncalibrated Oxygen sub.

Dumbbell-Oxygen-0018.fits 31.27 MB · 0 downloads

Well when I open that file in AA5, View>Range>CCD range gives the full monty, as I'd expect.

Edited October 6, 2020 by x6gas

[x6gas]

1 minute ago, DaveS said:

OK. It looks like I may have a clear moonless slot on Saturday, so I'll see what gain 139 looks like, but I don't have any Dark frames for that.

 

Just now, x6gas said:

Well when open that file in AA5, View>Range>CCD range gives the full monty, as I'd expect.

So I don't think that this has anything to do with your capture (the individual sub and the stack itself are definitely not over saturated)... so in my view it's something to do with the CCD range that AstroArt is applying to the stack, for some reason.

[andrew s]

9 minutes ago, DaveS said:

OK, single uncalibrated Oxygen sub.

Dumbbell-Oxygen-0018.fits 31.27 MB · 3 downloads

Just looked at this in AstroimageJ it's defiantly 16bit and while noisy looks to have plenty of detail. The ASI software converts it to 16 bit even is the camera is not. Regards Andrew

[Laurin Dave]

The uncalibrated sub looks much better (apart from the amp glow)..   So maybe a calibration issue, also your calibrated sub has lots of zeros..  do you use a Pedastal to prevent black clipping ?

Here they are with PI screenstretch .. top one is calibrated

[Laurin Dave]

Dave ..  if you post the relevant dark and flat and I’ll run it through Pixinsight 

Dave

[andrew s]

@DaveS what offset have set on the camera? Regards Andrew 

[DaveS]

@andrew s I will check tomorrow when I can reboot the platform computer

[DaveS]

15 minutes ago, Laurin Dave said:

Dave ..  if you post the relevant dark and flat and I’ll run it through Pixinsight 

Dave

Again single dark and flat, or the stack used for calibration. Sorry for the questions.

[DaveS]

Now here's a think I should have clocked, my lights were taken at -20c, but the dark and flats were at -25c.

I know PI would have thrown a wobbly but AA is quite happy with mismatched temps. I can take some more -20c Darks, not sure about the flats and dark flats.

[DaveS]

This is the 12 sub master dark

Master Dark.fit

I've just found out all my Flat frames were taken with an exposure of 0 seconds, instead of the 2 sec that I thought. The Dark Flats are at 2 sec. Argh!

Will have to do a load of Flats as well and the weather is rubbish for doing T-shirt flats

[Laurin Dave]

25 minutes ago, DaveS said:

Again single dark and flat, or the stack used for calibration. Sorry for the questions.

Probably both..  ie stacked flat stacked dark and single flat dark and flat dark 

[DaveS]

OK, just doing a trial stack without Flats and Dark Flats to see what happens. 
Will post single Darks in a mo

[DaveS]

OK, single Dark frame. Not worth posting a Flat is it is effectively a Bias frame. Histo shows just a few spaced out bars.

ASI1600 10M-0001Dark.fit

[DaveS]

Oh, and the stack without flats looks no different to the one with.

[andrew s]

8 minutes ago, DaveS said:

OK, single Dark frame. Not worth posting a Flat is it is effectively a Bias frame. Histo shows just a few spaced out bars.

ASI1600 10M-0001Dark.fit 31.27 MB · 0 downloads

What? A flat should be far from a bias frame! It should be exposed into the region where the fixed pattern noise dominates read and shot noise. Am I missing something?

Regards Andrew 

[DaveS]

It was *supposed* to be set at 2 seconds, which I determined by trial exposures, but somewhere along the line I must have left the exposure in Maxim on zero seconds, without realising it.

[DaveS]

Now my eyes ache, so will sign off until the morning.

[ollypenrice]

10 hours ago, steppenwolf said:

What is even more strange is that using Laurin Dave's O TIF, I am not getting this issue in Photoshop unless I am not trying hard enough?

It was a second stretch for me when it fell into combing, Steve but...

10 hours ago, x6gas said:

Hmmm.  This is an interesting one.  I should say that I am stuck with AstroArt 5 so things may be different with more modern releases, but:

If you click on View>Range>CCD range then you get the burnt out image that Dave posted originally.  This is strange as whenever I've done this with my data it gives the complete CCD range (as you'd expect from the name of the function) but that's not the case here - you get a very different result if you select View>Range>Min->Max... so saving that gives this (which behaves OK when I stretch it in PS):

9 Hours O.tif 31.26 MB · 0 downloads

so I wonder if AstroArt is somehow misreading or misinterpreting the bit depth of the camera?

Ian

this is working for me! Nice one Ciaran!!   EDIT. Sorry, spoke too soon. This is also breaking down under a hard stretch and also produces a massively posterized DBE model, which is new to me. The Ha didn't do this. See below.

Olly

PS My internet has suddenly started working at supernatural speed (for rural France.) Long may it continue.

Edited October 7, 2020 by ollypenrice
Stated.

[steppenwolf]

3 hours ago, ollypenrice said:

It was a second stretch for me when it fell into combing, Steve but...

I'm not trying hard enough then! 😁

[x6gas]

5 hours ago, ollypenrice said:

It was a second stretch for me when it fell into combing, Steve but...

this is working for me! Nice one Ciaran!!   EDIT. Sorry, spoke too soon. This is also breaking down under a hard stretch and also produces a massively posterized DBE model, which is new to me. The Ha didn't do this. See below.

...

Olly

PS My internet has suddenly started working at supernatural speed (for rural France.) Long may it continue.

OK could you indulge me here Olly...  by default I always get that posterized view of the background map (whether using ABE or DBE) but as I understand it that's just the Screen Transfer Function skimping on resources...  So could I ask: if, in PI, you go to IMAGE > Screen Transfer Function > Use 24-bit  STF LUTs (or indeed if you drop the STF into Histogram Transformation and apply the stretch for real) then do you still get the crude gradients?

I am curious whatever, because I assume that you don't usually see backgrounds looking like this... so something must be different with this file.

Ian

[DaveS]

Now taking a set of 10 min Darks at -20c.

Also checked the flat settings an I did set 2 sec exposures, so why they read 0.000 sec is a mystery.

[ollypenrice]

1 hour ago, x6gas said:

OK could you indulge me here Olly...  by default I always get that posterized view of the background map (whether using ABE or DBE) but as I understand it that's just the Screen Transfer Function skimping on resources...  So could I ask: if, in PI, you go to IMAGE > Screen Transfer Function > Use 24-bit  STF LUTs (or indeed if you drop the STF into Histogram Transformation and apply the stretch for real) then do you still get the crude gradients?

I am curious whatever, because I assume that you don't usually see backgrounds looking like this... so something must be different with this file.

Ian

I just used Ctrl A to get the screen stretch, Ian, and yes, when I subtracted the posterized background model I got an equivalent posterization in the image screen stretched in PI and  stretched as normal in Photoshop. This is the same workflow as I used on the Ha which worked well. Indeed it's what I always do.

For 9 hours the OIII breaks down fast under stretching. My own CCD data goes deeper in less time so I do think something is wrong with the file.

Olly

Edited October 7, 2020 by ollypenrice

[x6gas]

20 minutes ago, ollypenrice said:

I just used Ctrl A to get the screen stretch, Ian, and yes, when I subtracted the posterized background model I got an equivalent posterization in the image screen stretched in PI and  stretched as normal in Photoshop. This is the same workflow as I used on the Ha which worked well. Indeed it's what I always do.

For 9 hours the OIII breaks down fast under stretching. My own CCD data goes deeper in less time so I do think something is wrong with the file.

Olly

Interesting, thanks Olly.  I don't get the same thing when I do an ABE... something odd going on!  It does seem clear that the black point of the stack has been clipped somewhere along the line...

[DaveS]

OK, just checked the capture setting, and it was set to 16 bit, so that source of error is cleared up.