NGC1333 LRGB

[Rodd]

FSQ 106 with .6x reducer and ASI 1600.   1,348 1min subs: Red-353; Green-302; Blue-302, Lum-391.  luminance channel was a super luminance made with all 1,348 subs.  While it is definitely my best effort at this target, it is also woefully short of my expectations.  The dust still has a clayey aspect, and the palette has gone haywire.  I have pretty much reached the limit of the number of subs I can effectively handle without getting a new external hard drive.  And even this many has not improved the noise levels.  This image has over 20 hours of data and over 1,300 subs.....just how many are necessary?  At normal viewing its ok.  At full resolution it is painful to look at.

[wimvb]

Looks good, Rodd. Maybe you're pushing the dark nebula a bit too much, though.

This target is a royal pain to process, because of the mixture of bright and dark dust, and a background peeping through.

 

[Ronclarke]

That's a great image, I shouldn't beat yourself up over it! Most folk would be well satisfied with it!! 

Cheers

Ron

[vdb]

Well I think you pushed to hard on it, my first version was noisy as hell as well, 18H of data ...

Anyway it's a nice result and you will find that most samples on the net are noisy as hell, Olly and I did it once in the ODK 14 if I remember correctly ... not enough data but not that bad, maybe this one needs aperture.

V2 I was much more gentle, is it perfect, no I could probably add another 18H to get to a level that is acceptable.

How did you determine that 1 minute subs was enough?

/Yves

 

 

[Rodd]

3 hours ago, wimvb said:

Looks good, Rodd. Maybe you're pushing the dark nebula a bit too much, though.

This target is a royal pain to process, because of the mixture of bright and dark dust, and a background peeping through.

 

I have no choice.  If I don't drop the histogram the dynamic range is close to zero, and there I no background..  As it is, some of the background looks like dark nebula.  

Rodd

[Rodd]

3 hours ago, Ronclarke said:

That's a great image, I shouldn't beat yourself up over it! Most folk would be well satisfied with it!! 

Cheers

Ron

Thanks Ron.  

[Rodd]

2 hours ago, vdb said:

not enough data but not that bad, maybe this one needs aperture.

V2 I was much more gentle, is it perfect, no I could probably add another 18H to get to a level that is acceptable.

The whole point is FOV--aperture is the last thing I want.  Widefield is the goal.  I have seen amazing widefield shots of this region--with far, far less data.

The ASI 1600 is supposed to work very well with very short subs--some charts I have seen say 10 sec subs is sufficient.  I see very little difference between 2min and 1min, except the 2min have very high medians because my sky is not dark.  1min is certainly over the noise threshold. One is supposed to use these camera with  a lot of subs.  I think 1348 is enough.  

The idea that more data will help is just not true.  I see no difference between a stack with 250 compared with 1,200.  The grain does not disappear with increased data.

Rodd

[ollypenrice]

3 hours ago, vdb said:

Well I think you pushed to hard on it, my first version was noisy as hell as well, 18H of data ...

Anyway it's a nice result and you will find that most samples on the net are noisy as hell, Olly and I did it once in the ODK 14 if I remember correctly ... not enough data but not that bad, maybe this one needs aperture.

V2 I was much more gentle, is it perfect, no I could probably add another 18H to get to a level that is acceptable.

How did you determine that 1 minute subs was enough?

/Yves

 

 

We did do the colourful nebula close up in the 14 inch but not with enough data for a full size presentation. It was gloriously colourful when presented small, though, and that data finally went into my widefield from the Tak 106/Atik 11000. At some point in the processing of that data I got too much brown in the background but long subs in the CCD didn't suffer too much from noise. I think I had similar exp. time to Rodd. I'll see if I have the original widefield data but I suspect I'll have canned it.

Olly

[Rodd]

38 minutes ago, ollypenrice said:

It was gloriously colourful when presented small, though

So the colors are OK.  My first version was a mess, and I thought the rainbow-like colors were one of the things that was wrong with it.  I thought that that same problem carried over to this attempt.  I did not boost saturation at all.  I am currently working on other versions, but can't seem to do much better.  I am looking for a gasseous appearance (there is dust, but let's face it, these structures are 99% gas).  

The really frustrating thing for me with this image is at normal viewing it looks pretty decent.  It rings the bell (softly) at that scale.  But that is an illusion, for when you view it at full resolution, it is like diving into a bucket of pea soup.  And I weeded out many subs.  I think I collected a total of 1,500+.  I don't think more data is the answer.  Even if it is--I can't collect any more--my hard drive is chock-full, in fact I had to delete some of the subs after I made my stacks in order to make room to register another stack.  I am stuck with this data set,

Rodd

[carastro]

It's looking great, but I do find it depressing when you post up such good images better than most of us can achieve and you say things like this. 

Quote

woefully short of my expectations

I know you set yourself a high standard though.  

You seem to me to be getting the best I have seen so far of images done with a CMOS camera, but all those subs!!!!  and as you say getting to the point that your computer cannot take any more.   To be honest I am yet to be convinced that the CMOS approach is the way forward.  Cheaper yes, short subs yes, but 20 hours of data and a struggling laptop???  Might as well do CCD imaging and take up less space.  One advantage of course is less risk of clouds and planes passing through the exposures.

Anyway, back to the image, I think it's great.

Carole 

[wimvb]

2 hours ago, Rodd said:

I have seen amazing widefield shots of this region--with far, far less data.

It's the combination of aperture (D) and pixelscale (r) that is important. The optical imaging efficiency of a rig (excluding quantum efficiency) is proportional to (rD) squared. With a widefield rig, each pixel collects data from a larger area of the sky, and with larger aperture more light from each sky patch is caught on those pixels. So, to get a fair comparison, you have to know both r and D.

The second, maybe even more portant, aspect is sky darkness. For every magnitude you lose in sky darkness, your total integration time needs to be 2.5 times as long, in order to reach a similar snr.

Both these aspects make it very difficult to compare results taken with different rigs at different locations.

From what I know about your imaging conditions, it seems to me that you did very well on this target.

[Rodd]

58 minutes ago, carastro said:

but I do find it depressing when you post up such good images better than most of us can achieve and you say things like this. 

I am sorry Carole....I should stop and consider.  But frustration does that to me.  I have seen images by you and others that I can only aspire to.  Regarding the CMOS thing--I went way overboard on this.  Normally I don't have 1,400 subs.  Normally I don't need them.  I tried this target with my STT-8300 (CCD) and the results pale in comparison to this (much less data though).   I find total integration times to be teh same between CCD and CMOS.  Its just more and shorter subs.  there are advantages and disadvantages.  A big disadvantage for me is I can't dither every few subs--with Maxim its all or nothing.  So between teh dither itself and the dither delay, I have to add about 30 sec to each sub, so the shorter the subs, the less imaging time I get.  When I shoot 5min Ha, it's not as bad.  With 1 min subs, I only get about 38 subs per hour instead of 60 (it should be 45 actually).  That's a lot of time wasted.

But I like the small pixels and big FOV.  No complaints with the ASI 1600.  It's my sky and my processing that are at issue!  BTW--I have a second attempt--any better? (see below
 

7 hours ago, wimvb said:

Looks good, Rodd. Maybe you're pushing the dark nebula a bit too much, though.

This target is a royal pain to process, because of the mixture of bright and dark dust, and a background peeping through.

 

any better Wim?

 

2 hours ago, ollypenrice said:

We did do the colourful nebula close up in the 14 inch but not with enough data for a full size presentation. It was gloriously colourful when presented small, though, and that data finally went into my widefield from the Tak 106/Atik 11000. At some point in the processing of that data I got too much brown in the background but long subs in the CCD didn't suffer too much from noise. I think I had similar exp. time to Rodd. I'll see if I have the original widefield data but I suspect I'll have canned it.

Olly

Any better Olly?

 

 

[Rodd]

or this

 

[ollypenrice]

A few thoughts: firstly you have some quite broad edge artefacts, I think. Scan round the edge of the frame to see if you agree.  When I get these it is from the need to stretch to a manic degree. I got them when struggling to find the 'Fabian Neyer' Ha background to the Double Cluster, for instance.

I guess I agree that there could be more signal for the time given but, as pointed out earlier in the thread, LP could be adding to your difficulty. Maybe this just is a dark sky target. I don't know because I've never imaged anywhere but where I now live, which is very dark.

I'm also not sure about the magenta zones. I didn't find them in my shoot but I see magenta patches creeping into some posted images. They are not in consistent locations, though, which suggests they are an incipient processing hazard in this target. I really don't know.

Olly

[Rodd]

10 minutes ago, ollypenrice said:

A few thoughts: firstly you have some quite broad edge artefacts, I think. Scan round the edge of the frame to see if you agree.  When I get these it is from the need to stretch to a manic degree. I got them when struggling to find the 'Fabian Neyer' Ha background to the Double Cluster, for instance.

I guess I agree that there could be more signal for the time given but, as pointed out earlier in the thread, LP could be adding to your difficulty. Maybe this just is a dark sky target. I don't know because I've never imaged anywhere but where I now live, which is very dark.

I'm also not sure about the magenta zones. I didn't find them in my shoot but I see magenta patches creeping into some posted images. They are not in consistent locations, though, which suggests they are an incipient processing hazard in this target. I really don't know.

Olly

I cropped the edges--see above--But I did not stretch hard.  these were there from the very beginning.  Maybe DBE is to blame.  DBE fails me at least 50% of the time--especially when there is very little clear background.  All in all--give me narrow band any day.  No matter how many subs I tackle, and no matter how long the exposures are--I am plagued with speckles/grain.  I would start to think teh equipment is faulty--but I have been down that road and therein lie no answers.

[ollypenrice]

1 minute ago, Rodd said:

I cropped the edges--see above--But I did not stretch hard.  these were there from the very beginning.  Maybe DBE is to blame.  DBE fails me at least 50% of the time--especially when there is very little clear background.  All in all--give me narrow band any day.  No matter how many subs I tackle, and no matter how long the exposures are--I am plagued with speckles/grain.  I would start to think teh equipment is faulty--but I have been down that road and therein lie no answers.

Could be DBE if the markers were right on the edge.

I've found my original data and will try a reprocess since this is an interesting problem bit of sky and my existing process is whacky. I've been obsessed for some time by a research project from my other life in English Lit., so it will be good to work on a picture.

Olly

[carastro]

There is a processing BLUE artifact (a blue line) at around 5.30 position in relation to the Nebula (the one marked (or this).  Approx the same distance again away from the nebula as the length of the Nebula itself.

Don't think it is on the other versions.

Carole 

[wimvb]

5 minutes ago, Rodd said:

All in all--give me narrow band any day.  No matter how many subs I tackle, and no matter how long the exposures are--I am plagued with speckles/grain

To me, this indicates that light pollution is to blame, rather than equipment. As @ollypenrice wrote the other day in another thread: it's not what filters let through, but what they block, which is important. And if nb filters effectively block light pollution (as 3 nm Astrodons very well should), you'll get better results with nb.

[carastro]

Correction the blue artifact seems to be on all of them.  

Carole 

[Rodd]

6 minutes ago, ollypenrice said:

Could be DBE if the markers were right on the edge.

I've found my original data and will try a reprocess since this is an interesting problem bit of sky and my existing process is whacky. I've been obsessed for some time by a research project from my other life in English Lit., so it will be good to work on a picture.

Olly

That's the trouble with DBE and images like this--at the time you place the markers it is VERY difficult to determine what is background and what is nebula--almost impossible..  I toil at this for hours.  

Rodd

[wimvb]

2 minutes ago, carastro said:

There is a processing BLUE artifact (a blue line) at around 5.30 position in relation to the Nebula (the one marked (or this).  Approx the same distance again away from the nebula as the length of the Nebula itself.

Don't think it is on the other versions.

Carole 

There's also a red artefact higher up to the left. Could it be caused by a minor planet in the fov, as was discussed in another thread?

 

[Rodd]

7 minutes ago, wimvb said:

To me, this indicates that light pollution is to blame, rather than equipment. As @ollypenrice wrote the other day in another thread: it's not what filters let through, but what they block, which is important. And if nb filters effectively block light pollution (as 3 nm Astrodons very well should), you'll get better results with nb.

But M8 and M20 came out hjust fine, as did the cacoon--even the tadpoles LRGB.  And galaxies (struggle with background)--but M31 was fine.   Anybody want to try there hand at 1,400 subs?  Might not get another chance!  I'll post the fits if you have the hankering

Rodd

[Rodd]

7 minutes ago, carastro said:

Correction the blue artifact seems to be on all of them.  

Carole 

Yup

[vdb]

4 hours ago, Rodd said:

The whole point is FOV--aperture is the last thing I want.  Widefield is the goal.  I have seen amazing widefield shots of this region--with far, far less data.

The ASI 1600 is supposed to work very well with very short subs--some charts I have seen say 10 sec subs is sufficient.  I see very little difference between 2min and 1min, except the 2min have very high medians because my sky is not dark.  1min is certainly over the noise threshold. One is supposed to use these camera with  a lot of subs.  I think 1348 is enough.  

The idea that more data will help is just not true.  I see no difference between a stack with 250 compared with 1,200.  The grain does not disappear with increased data.

Rodd

Yes I know perfectly well that CMOS is supposed to be used in that way, I have both the 1600 and the 183 camera, however 1 min subs is probably to short to get the faint dust noise free, whatever amount of subs thrown at it. I have the luxury of  being at Olly's dark place, but even at home I cannot get that deep with short subs, I once tested the 1600 camera on the Iris nebula with short subs and I was kind of in between happy and disappointed ...

/Yves

 

 

[Rodd]

6 minutes ago, wimvb said:

There's also a red artefact higher up to the left. Could it be caused by a minor planet in the fov, as was discussed in another thread?

No way--the palette is a smorgasbord of hues--as is always the case for me with this target.  I am done with it.  Never again