M101... Signs of improvement!

[StargeezerTim]

This is based on 38 x 7.5 minute exposures at iso 800. Taken over two nights last week. Plus darks, bias and flats that don't seem to do much!

Better than my previous attempts at this one but even more data would be nice... 

Suggestions for improvement please....

[Leveye]

Use levels and curves to bring out even more detail. Nice work!

[glowingturnip]

very nice work mate !  I think you might be white-clipping your stars a tad in processing ?

[StargeezerTim]

very nice work mate !  I think you might be white-clipping your stars a tad in processing ?

Hi Stuart,

Thanks...i'm not sure what 'white clipping' stars is but I'm sure I am doing it!

To be honest, my processing skills are really naff... seat of the pants stuff... I've tried processing this image loads of times and can't really produce anything I am happy with. Processing is definately something I need to improve on. 

I'm really pleased I have learn't how to use the scope and camera to get good images... but it has highlighted to me my basic skills in post processing. What a learning curve this is!

Tim.

[glowingturnip]

hi mate,

Easy enough to explain.  Imagine if you graphed your pic in 3 dimensions, with the z dimension being the brightness at any given point, then the stars should all look like nice little bell curves, like this:

However, if you move your white point down too much on your histogram, or stretch it too hard, then you end up slicing the tops off of these nice bell curves, a little like that mountain in Close Encounters of the Third Kind.  if you do that, then the part of the curve that's been sliced off presents as a simple round white circle with no definition and no colour in it, which is what you've going on there.

Easy enough to control.  I almost never move the white point in my histogram nowadays, and only adjust the black and grey points, and I try to make sure if possible that none of the final image is reading a brightness of pure white (1.000 or 256 or whatever scale you're looking at)

Definitely an image to be proud of though, and again, it beats my attempt at the same target hands down !  I need to go a revisit a lot of my old things I think !

[StargeezerTim]

hi mate,

Easy enough to explain.  Imagine if you graphed your pic in 3 dimensions, with the z dimension being the brightness at any given point, then the stars should all look like nice little bell curves, like this:

However, if you move your white point down too much on your histogram, or stretch it too hard, then you end up slicing the tops off of these nice bell curves, a little like that mountain in Close Encounters of the Third Kind.  if you do that, then the part of the curve that's been sliced off presents as a simple round white circle with no definition and no colour in it, which is what you've going on there.

Easy enough to control.  I almost never move the white point in my histogram nowadays, and only adjust the black and grey points, and I try to make sure if possible that none of the final image is reading a brightness of pure white (1.000 or 256 or whatever scale you're looking at)

Definitely an image to be proud of though, and again, it beats my attempt at the same target hands down !  I need to go a revisit a lot of my old things I think !

Thanks Stuart,

You are right, the first thing I do when processing is move the black point towards the start (on bottom of graph) of the curve and the white point towards the end of it (on top of graph) in Curves. I'm sort of flailing about a bit (or experimenting!) but enjoying it none the less.

Thanks, Tim, 

[Marky1973]

Tim...two things....

Did you plate solve over the different nights, or just target the same area of sky and let the stacking software do its thing?

And, processing, there is a lot to learn and I have mentioned it before but, if you haven't  - check out Doug German's tutorials on YouTube - it is the basics, but just his info on curves has helped me actually figure out what they are an understand the principles behind them - still a long way to go, but they do make a different once you figure them out and apply them gently! They are based around PhotoShop, but the principles are the same wherever your curves are I guess!

Great image!

[StargeezerTim]

Tim...two things....

Did you plate solve over the different nights, or just target the same area of sky and let the stacking software do its thing?

And, processing, there is a lot to learn and I have mentioned it before but, if you haven't  - check out Doug German's tutorials on YouTube - it is the basics, but just his info on curves has helped me actually figure out what they are an understand the principles behind them - still a long way to go, but they do make a different once you figure them out and apply them gently! They are based around PhotoShop, but the principles are the same wherever your curves are I guess!

Great image!

Thanks Marky,

I did see his one on Curves... but will find time to watch them all. I don't know what plate solving is! I just used synscan to go to the same object, after a very careful two star alignment. To be fair, it was very accurate, I used DSS to align and the overlap between the two nights was about 96% ish.

I have done another edit here...hopefully the stars are a little more realistic now. I know I am a beginner, but it seems its about give and take and balance, curves can bring a lot of 'hidden' stuff out, but at the expense of noise. I am seriously considering getting a light pad for about 30 quid as my flats don't seem to be having any impact.

Cheers, Tim

Just thought, too much sharpening may be affecting the stars....

[glowingturnip]

I preferred the colours in the first one to be honest, and yes, don't over-sharpen the stars - if you look closely at the latest one, you've got little black rings round the stars now, which is a sharpening artefact.   Maybe look into building a star mask, and then sharpening with the stars masked out, there should be videos on that.  

If you want to, dropbox me a copy of the initial raw file, and I can get rid of that gradient for you, colour balance it and give it a first stretch all in Pixinsight.  You've got a cracking photo in there trying to come out.

[Notty]

Hi that's very nice, I've been playing with my 1st M101 this week too.

[StargeezerTim]

I preferred the colours in the first one to be honest, and yes, don't over-sharpen the stars - if you look closely at the latest one, you've got little black rings round the stars now, which is a sharpening artefact.   Maybe look into building a star mask, and then sharpening with the stars masked out, there should be videos on that.  

If you want to, dropbox me a copy of the initial raw file, and I can get rid of that gradient for you, colour balance it and give it a first stretch all in Pixinsight.  You've got a cracking photo in there trying to come out.

Thanks,

I'll take you up on your offer Stuart. I'll message you. I also prefer the first one but can't seem to get there again!

Anyway, I'm imaging M109 tonight... lovely clear night here so far....

Tim.