This is why I know I need more practice :(

[iapa]

Managed a few shots on Friday night of M51.

20 x 20s at ISOs 800, 1600,3200 to get a feel for which setting is best... 20s as images where trailing at 30s.

Here's the stack of the 20 x ISO 1600.

Tried again Monday night - ISO 800 only - nada due to thin cloud.

Also noted that the stars where trailing slightly too, not sure about why that tho.

[kendg]

Practice makes perfect.

All in all looking pretty good. Nice image.

[D4N]

That is really nice for less than 7 minutes of integration time.

[PatrickGilliland]

Second that - practice with some patience and dedication will fix a lot of things  

What's more there is a lot more lurking in there than you realise - hope you dont mind had a quick dabble!

A few histogram adjustments and some tweaking of light quickly releases data - while not much data you can see that its ready to be extracted.  Maybe another 20-30 subs at this length will give you something nice you can work with.

Just make the most of the data once you have.  Give me a shout if you need some pointers, happy to help

Paddy

 

[Stub Mandrel]

I got a similar result by clicking auto-adjust in photopaint.

Twenty minutes playing with levels and colour balance could give you a very satisfying result and you will learn a lot.

For me stretching a digital image is as exciting as developing a film, you never know what will appear...

[iapa]

I posted at lunch time, after I got home this eve, I started again, then saw the complimentary comments.

now I have the attached.

I thing that I need to work out the capturing before I really get into post processing.

and if I could figure out exactly how to do flats....people keep taking about -e, nose, ADUs bot no-one explains how the release, or what they are. over a year trawling and I still do not understand

 

[iapa]

5 hours ago, PatrickGilliland said:

Second that - practice with some patience and dedication will fix a lot of things  

What's more there is a lot more lurking in there than you realise - hope you dont mind had a quick dabble!

A few histogram adjustments and some tweaking of light quickly releases data - while not much data you can see that its ready to be extracted.  Maybe another 20-30 subs at this length will give you something nice you can work with.

Just make the most of the data once you have.  Give me a shout if you need some pointers, happy to help

Paddy

 

I'm struggling to get rid of the noise

Agreed more lights - I tried that Monday, but thin cloud just ruined everything

[D4N]

33 minutes ago, iapa said:

I posted at lunch time, after I got home this eve, I started again, then saw the complimentary comments.

now I have the attached.

I thing that I need to work out the capturing before I really get into post processing.

and if I could figure out exactly how to do flats....people keep taking about -e, nose, ADUs bot no-one explains how the release, or what they are. over a year trawling and I still do not understand

Flats are simple enough, point the scope at a plain white light source (a well lit white wall will work but I have a small scope so just put my iPad on top of it) and takes exposures that put the histogram peak between 1/3 and 2/3 of the way along.  AV mode on a DSLR should do this for you.  Everything else should be exactly the same as if you were taking a light frame, same filters, focus position, gain(ISO) and camera orientation.  This records the differing light transmission through your scope onto your sensor.

You also need bias which is easiest if you remove the camera and cap it then take loads of minimum length exposures at the same ISO as you are imaging at.  This records the read noise which you need to remove from the flats before using them.

Together they will calibrate your image.

[wimvb]

For a practice image, this is a very good one, that holds more data than your first rendering shows.

The best way to beat noise is to gather more data. If you can't do longer exposures, try more of them. Even 90 * 20 seconds is only half an hour in total, and will make quite a difference.

To get more signal, you will need longer exposures, so you might want to check your setup: balance, backlash, polar alignment, total load on the mount, etc.

Once you have the data, processing will be easier as well.

BTW, have you considered combining the different ISO subs? You may find that the total can be better than the sum of the parts, even here.

[iapa]

10 hours ago, wimvb said:

For a practice image, this is a very good one, that holds more data than your first rendering shows.

The best way to beat noise is to gather more data. If you can't do longer exposures, try more of them. Even 90 * 20 seconds is only half an hour in total, and will make quite a difference.

To get more signal, you will need longer exposures, so you might want to check your setup: balance, backlash, polar alignment, total load on the mount, etc.

Once you have the data, processing will be easier as well.

BTW, have you considered combining the different ISO subs? You may find that the total can be better than the sum of the parts, even here.

I though about combining but wasn't sure - do each set individually, match histogram (Nebulosity) and then stack again?

Though I saw something with DSS using groups - I did group 1 with ISO 800, 2 with 1600 and 3 with 3200. DSS then registered and report only one image would be stacked (I'd set to use best 95%).

 

JUst occurred, I also need to check if LENR is turned on as then I did a stack of the ISO 800s with some darks there was no detail at all to pull out

[wimvb]

My first thought was that you could stack them all as groups, as you already tried. But the signal difference between ISO 800 and ISO 3200 may be too large. You could try just the ISO 1600 and ISO 3200, that contain most of the signal.

Also you may want to sort out the LENR issue first.

[Stub Mandrel]

If you use DSS you can create load the ISO800 files, then create two additional 'tabs' for ISO 1600 and 3200 files at the bottom of the screen. It will then work out how to balance the three sets to give a single stack.