Please please please help with Running Man Nebula!!

[Costas Soler]

Hi!

I've been into astrophotography for about 6 months now, and I really love it. I've had a chance to get some pretty decent pictures so far. However, I've been pretty displeased by some of my recent targets. I'm including one of my better images, the lagoon nebula (three hours 49 minutes of data, 3 minute subframes at ISO 400) to contrast with my not-so-good images of M1 and the Running Man Nebula. My M1 picture is from about 90 minutes of data, made with 4 minute subframes at ISO 400 as well. This one was a bit grainier than I'd normally accept. However, since it's such a tiny object to image with my f/3.9 scope, and no barlow, I was actually a little happy with this one. The image of the Running Man Nebula, is a different story. I included a jpeg of the unmodified autosaved picture that Deep Sky Stacker spits out (labelled "bobby.jpg" so I could find it easily on my desktop). No matter what I do with it, I really can't bring out much more data than you see here! In fact, when I do play with it in DSS or in photoshop, I can tease out a bit of the nebula's structure, but with it comes some horrible red noise. I don't know if this means anything, but when I try to process running man nebula, the orion nebula at the bottom of the image actually looks ok. Perhaps this is because some of the frames that cover the running man, but not M42 just have really, really bad data? Or am I making some incredibly newbie mistake and I've just had insanely good blind luck so far? Let me know how I can improve!!! Always willing to take another pass at this guy when the clouds leave!

I am taking all these images with a Nikon D3200, an 8" Orion Newtonian Astrograph, an Orion Atlas mount, and using an Orion SSAG.

[Atreta]

Hi,  nice captures. 

How are you processing it after you stack the subs? I'm not an expert,  but looks to me that you are black clipping your data.

  I prefer to line up the rgb levels in dss and move them against the curve to the right,  then go to saturation tab and move it to about 23% and preview it,  then i save it with applied settings(i find easier this way).

Looks like most people here prefer to save with settings embedded instead of apllied. 

  Thinking only about integration i guess there is a lot more to show from your lagoon nebula ( can't say much about others).  I'll post a screenshot of how i do it in dss,  and one lagoon nebula i did about 2 years ago,  nowhere near the same integration time as yours. 

 

 

[Marky1973]

Have you tried upping the ISO to 800? Often a sweet spot for imagers - come do push to 1600 at times to try and get more, but you will, obviously, increase the noise potential. You can tweak things a bit in DSS to see what sort of image you might have lurking, but most people would not recommend using it for any sort of processing. I have a very absic DSS tutorial on my blog in my signature if you want to have a look. There is also an Orion image in the gallery which was taken with an unmodified Canon 1100D using 90second 800ISO subs.

[Astrosurf]

Nice images! As said, up the ISO to 800. I use 1600. Did you take darks? These are essential to reduce noise.

Have you tried The GIMP to process? It's free and good enough for basics, I've heard. I've not used it myself. 

Alexxx

[Costas Soler]

Hi! Thanks very much for replying! I think I'm gonna try doing another run at ISO 800. Before I do that, I was curious: why do you think that will help improve the image quality even though noise will be increased? I'm still getting a feel for photography, and I really want to understand it:). 

Also Marky1973 I'd love to see that tutorial!

[Marky1973]

Hi Costas - click on the weblink in my signature and it will take you there! Or you can cheat with this link! :-)

https://markwalkerscreenwriting.wordpress.com/deep-sky-stacker-tutorial/

It is very basic though and you have to remember there is a wealth of possible ways to process the image once you have stacked it.

I'm not an expert to be honest, but even if you increase noise, you will also be increasing the signal you collect. Its all about the signal-noise ratio and finding the best combination of exposure length and ISO to get the best of both worlds. Many APers will suggest looking at the histogram of your exposure and aim for a combination of time and exposure that puts the peak of the histogram about 1/3 of the way across from the left!

You can deal with noise through processing and use your calibration subs as well. I think noise is largely inevitable, especially with DSLRs. But if you up the iso, you may be able to drop the exposure time and, potential, the noise in the image.  

You can also help prevent hot pixels building up in your stacked subs if you dither your exposures, which just means the camera is moved very slightly between shots. Search for it on this site and there will be loads of stuff to help you with that.

You could also try leaving longer between exposures to give you camera chance to cool down a bit!

And remember, the running man is just  that bit fainter than M42, which is a pretty bright target.

There is a lot to think about! :-)

[MARS1960]

10 hours ago, Astrosurf said:

Nice images! As said, up the ISO to 800. I use 1600. Did you take darks? These are essential to reduce noise.

Have you tried The GIMP to process? It's free and good enough for basics, I've heard. I've not used it myself. 

Alexxx

I was always under the impression that taking darks with a DSLR adds noise and it's  only flats and bias that are needed? Modern dslrs have something called on sensor dark current suppression built in I believe, negating the need to take dark frames.

[Thalestris24]

13 minutes ago, MARS1960 said:

I was always under the impression that taking darks with a DSLR adds noise and it's  only flats and bias that are needed? Modern dslrs have something called on sensor dark current suppression built in I believe, negating the need to take dark frames.

Some cameras take a dark after a normal exposure. But... if you take a 3 min light then the camera takes a 3min dark which loses valuable imaging time. Darks don't add noise provided you take enough of them - ~30-40 is usually sufficient. For a given iso and exposure you can store (master) darks in a library. I find it best to keep to one iso (800) and limit exposure lengths to a small selection such as 120s, 180s, 360s and maybe 600s.

Louise

[newbie alert]

Id say great images..you have the know how in your processing...only thing i could add is like everyone else is saying and up the iso to at least 800..and yes do take darks....add the new data to the old and combine it in together and re stack it..

[wimvb]

A lot of useful help already provided. What nobody pointed to is to start processing an image with cropping. Remove any edges due to stacking. If you don't, you won't be able to use the information given by histograms in your processing software.

The running man neb is a lot weaker than the orion neb. You will need much longer exposures. Try the longest single sub exposures that still give acceptable (= not bloated) stars.

Upping iso will result in more noise. But you can reduce this by also increasing the number of exposures. Calibration frames can never remove random noise, only fixed pattern noise. Using more subs to build master calibration images, will result in less noise being added to the light frames.

Unfortunately, any light pollution will add its own noise, increasing with exposure time. How much you can tolerate, is up to you. As always in this game: experiment.

Good luck

[dannybgoode]

7 hours ago, Thalestris24 said:

Some cameras take a dark after a normal exposure. But... if you take a 3 min light then the camera takes a 3min dark which loses valuable imaging time. Darks don't add noise provided you take enough of them - ~30-40 is usually sufficient. For a given iso and exposure you can store (master) darks in a library. I find it best to keep to one iso (800) and limit exposure lengths to a small selection such as 120s, 180s, 360s and maybe 600s.

Louise

Because noise is related to sensor temperature and dslr's are not generally cooled it is best to take custom darks at the same time as the light frames otherwise the noise profile will differ. This will eat in to imaging time but you could automate the task and have the dslr shooting darks whilst you're setting everything up and then a few more when packing it all up again. 

You could I suppose build up a library of darks shot at different ambient temperatures and exposures but a master dark from a dslr captured in the winter say and then applied tool lights captured in the summer or vice versa may do more harm than good.

[Peco4321]

16 hours ago, Marky1973 said:

Or you can cheat with this link! :-)

Thanks for that. Used some of your suggestions and I'm getting some better results of my Orion Nebula. I suppose all this cloudy weather gives us time to process and play with old data ?

[Marky1973]

Yep, been doing the same here Peter - although we have a forecast clear night tomorrow - first for two weeks.....and I'm not going to be around to take advantage! Typical!

[Peco4321]

Just now, Marky1973 said:

Yep, been doing the same here Peter - although we have a forecast clear night tomorrow - first for two weeks.....and I'm not going to be around to take advantage! Typical!

Does this sound familiar-"if your not outside your sat at the bloody computer all night"!!!!!

[Marky1973]

Yep....and as my other hobby is writing.... 

[Thalestris24]

52 minutes ago, dannybgoode said:

Because noise is related to sensor temperature and dslr's are not generally cooled it is best to take custom darks at the same time as the light frames otherwise the noise profile will differ. This will eat in to imaging time but you could automate the task and have the dslr shooting darks whilst you're setting everything up and then a few more when packing it all up again. 

You could I suppose build up a library of darks shot at different ambient temperatures and exposures but a master dark from a dslr captured in the winter say and then applied tool lights captured in the summer or vice versa may do more harm than good.

Yes, they do have to be temperature matched.

[MARS1960]

10 hours ago, Thalestris24 said:

Some cameras take a dark after a normal exposure. But... if you take a 3 min light then the camera takes a 3min dark which loses valuable imaging time. Darks don't add noise provided you take enough of them - ~30-40 is usually sufficient. For a given iso and exposure you can store (master) darks in a library. I find it best to keep to one iso (800) and limit exposure lengths to a small selection such as 120s, 180s, 360s and maybe 600s.

Louise

Absolutely Louise, i always recommend that everyone using a DSLR turn off "in camera dark subtraction/in camera noise reduction".

As for taking dark frames this has been a bone of contention for quite some time, some very knowledgeable astrophotographers say yes, some say no, my personal opinion is no BUT definately dither and use sigma clipping and then forget the hassle of dark frames, you shouldn't need them.