m33 88x60sec

[TiffsAndAstro]

hapy with result for less than 90 minutes. quick stretch and unpurple stars.

m33 is not pretty though. not my fault  also i've clearly discovered a purple supernova and/or nebula right at the tip of its southern arm....

 

[TiffsAndAstro]

second version is nicer  i'm quite happy with this, not sure about purple blob. i see something vaguely similar in some others' images though.

 

 

[Clarkey]

Given the minimal integration times that is a good result. M33 has quite a low surface brightness.

Now the controversial bit....

Instead of taking an hour or so on each target, increase this to a few hours on one. You will see a huge difference in your results. Also, good data is easier to process. I appreciate as a relative beginner you want to image everything, but after a while you will look back and wonder why you didn't do this to start with. From my Bortle 6 garden, I rarely collect less than 8 hours on a subject - and often muck more.

Clearly you have the basics sorted - time now for the APOD👍

[TiffsAndAstro]

1 hour ago, Clarkey said:

Given the minimal integration times that is a good result. M33 has quite a low surface brightness.

Now the controversial bit....

Instead of taking an hour or so on each target, increase this to a few hours on one. You will see a huge difference in your results. Also, good data is easier to process. I appreciate as a relative beginner you want to image everything, but after a while you will look back and wonder why you didn't do this to start with. From my Bortle 6 garden, I rarely collect less than 8 hours on a subject - and often muck more.

Clearly you have the basics sorted - time now for the APOD👍

i agree with you and everyone else on here who's told me same thing. for the rest of july ill add to my ngc6888 or 10 hours whichever is sooner. 

[TiffsAndAstro]

What concerns me still, is the end results will improve as they should. My attempts at combing multiple sessions on other targets doesn't fill me with confidence, though I have improved my skills a bit.

It's difficult to describe exactly what I mean, but if you look at my pics in my gallery, they appear to have a similar feeling/colouring/palette (non of these are correct terminology) even with two different cameras and I have a feeling it will just end up of more/brighter version  if I was Rembrandt this would probably be a good thing. Not convinced it is for a wannabe astrophotographer they look too stylised? And I'm definitely not consciously trying to do that. I think I *want* a more sciency accurate look but not Hubble blue and yellow. Was gonna be funnier but some people like it so best not

I guess I need to get 10 hours on NGC 6888 and see how it looks. 

I love clouds.

and 5/8th 1/4 unc20 countersunk bolts. Not 3/8th ones though, they're clearly evil 

Edited July 18 by TiffsAndAstro

[TiffsAndAstro]

Also I think I need actual time imagining so I can f£#*k up and change things.

Lots of help from this forum  and maybe some good  luck seems to have made my progress relatively easy, but UK weather sucks.

[Clarkey]

6 hours ago, TiffsAndAstro said:

It's difficult to describe exactly what I mean, but if you look at my pics in my gallery, they appear to have a similar feeling/colouring/palette (non of these are correct terminology) even with two different cameras and I have a feeling it will just end up of more/brighter version  if I was Rembrandt this would probably be a good thing. Not convinced it is for a wannabe astrophotographer they look too stylised? And I'm definitely not consciously trying to do that. I think I *want* a more sciency accurate look but not Hubble blue and yellow. Was gonna be funnier but some people like it so best not

I think everyone has their own style, which will develop over time. Some people like bright colours, some more muted. Similarly, some like scientific representations, others are more artistic. Ultimately, there is no right and wrong (within reason). It is also tricky when you start to actually achieve the result you want. As I have said before, I am 4 years in to this hobby, and I only just feel competent now.

[chrisecurtis]

I think it can be helpful to ask yourself what you are trying to reveal or clarify by additional integration and processing. There's an Adam Block video (I think the one about GHS on YouTube) where he goes round a process of looking at the image and asking himself what's important in it (e.g. is it showing the faint outer nebulosity or retaining detail and contrast in the core) It's simple, but leads to clear decisions about how to process and render. It's also useful because any real image is a compromise: you can't show everything equally well, so it's good to decide what to prioritise. It also catches things like colour balance and sharpening so that you more consciously make choices about them and can even experiment.

Most of all, it helps you manage diminishing returns. Increasing integration time improves the signal to noise ratio, which can help contrast and detail, but that's all it does. You can only decide when "enough is enough" when you have decided what's most important in that image. Sometimes that might be the lowest possible level of visible noise and so push you to make many hours of sub-frames, but surprisingly often, you can see and enjoy the critically important things in much less time and adding many hours doesn't really do much to improve them.

I'd caution against the idea that you always have to do tens of hours on a target. Sometimes you do, but it's not magic: you can get what you want and need with much less, surprisingly often. 

[Clarkey]

13 minutes ago, chrisecurtis said:

I'd caution against the idea that you always have to do tens of hours on a target. Sometimes you do, but it's not magic: you can get what you want and need with much less, surprisingly often. 

I don't think anyone is suggesting the need for 10's of hours. However, you will see a huge difference between an hour and four hours. After that it is diminishing returns, and the user has to assess what they feel is enough time. The image below is one of my earliest and is 4 hours on the same object. Using and ED80 and 1600MM pro - so probably similar in terms of signal gathering. The processing is poor as I was a newbie at the time - but there is still a big difference in the result.

[Elp]

Having done a number of different targets it's a general consensus to do around 5-10h, but a lot depends on the target being imaged, your optics and what you're looking to achieve.

I can technically do a finished image within 2h with my F2s but I don't, trying to do the same with an F6/7 will not give the same result, so needs more hours. Trying to capture very faint signal (gasses rarely seen in the majority of images, galaxy tidal tails for example) need more time, if you're not concerned about such things then project time can be dialled back.

You still have to image for time to reduce the noise level in signal and background sky, more so if imaging from a LP area, and by doing so you can sharpen your detail better too.

Edited July 19 by Elp

[TiffsAndAstro]

2 hours ago, chrisecurtis said:

I think it can be helpful to ask yourself what you are trying to reveal or clarify by additional integration and processing. There's an Adam Block video (I think the one about GHS on YouTube) where he goes round a process of looking at the image and asking himself what's important in it (e.g. is it showing the faint outer nebulosity or retaining detail and contrast in the core) It's simple, but leads to clear decisions about how to process and render. It's also useful because any real image is a compromise: you can't show everything equally well, so it's good to decide what to prioritise. It also catches things like colour balance and sharpening so that you more consciously make choices about them and can even experiment.

Most of all, it helps you manage diminishing returns. Increasing integration time improves the signal to noise ratio, which can help contrast and detail, but that's all it does. You can only decide when "enough is enough" when you have decided what's most important in that image. Sometimes that might be the lowest possible level of visible noise and so push you to make many hours of sub-frames, but surprisingly often, you can see and enjoy the critically important things in much less time and adding many hours doesn't really do much to improve them.

I'd caution against the idea that you always have to do tens of hours on a target. Sometimes you do, but it's not magic: you can get what you want and need with much less, surprisingly often. 

I'll have a look for his video Ty. I've watched one or two of his other videos but wasn't keen. Will try again.

At the moment I'm not asking those questions, I'm more at the level of "how do I make this look 'nice' "

 

Edited July 19 by TiffsAndAstro

[Elp]

But the thing is, getting an image to look "nice" requires good data, which also needs time. You can stretch, deconvolve/sharpen, denoise a short time image but you'll also be doing the operations to the noise and a short time project will have a lot of noise, just look at your histogram after it's been stretched, a lot of peaks and troughs is a noisy image.

I think youve done well with the short time data you've been doing, but start doing longer to get to the next level.

[chrisecurtis]

50 minutes ago, TiffsAndAstro said:

I'm more at the level of "how do I make this look 'nice'

That's a valid goal for making an astrophoto, but you might achieve it more quickly if you break down, even just a little, what you mean by 'nice' for this specific target. The images you posted look nice enough to me: is it that you are seeking to improve detail, have smoother, less noisy background, more contrast and fine detail in the central area, tighter stars, more saturated colour, a tighter crop on the galaxy etc. etc. It's tempting to say "all the above" but working on some of them will tend to make others worse (e.g. reducing noise and increasing the finest detail is tricky) so you often have to make choices. More integration (e.g. doubling from 90 minutes total to 3 hours) will help a lot, and more might help a bit more, but when you are experimenting, remember to experiment with a goal in mind. Otherwise, you might go down too many rabbit holes and forget what you were trying to achieve in the first place (been there, done that)

[TiffsAndAstro]

1 hour ago, chrisecurtis said:

That's a valid goal for making an astrophoto, but you might achieve it more quickly if you break down, even just a little, what you mean by 'nice' for this specific target. The images you posted look nice enough to me: is it that you are seeking to improve detail, have smoother, less noisy background, more contrast and fine detail in the central area, tighter stars, more saturated colour, a tighter crop on the galaxy etc. etc. It's tempting to say "all the above" but working on some of them will tend to make others worse (e.g. reducing noise and increasing the finest detail is tricky) so you often have to make choices. More integration (e.g. doubling from 90 minutes total to 3 hours) will help a lot, and more might help a bit more, but when you are experimenting, remember to experiment with a goal in mind. Otherwise, you might go down too many rabbit holes and forget what you were trying to achieve in the first place (been there, done that)

Lol I was going to reply "all of those things"

Is there any articles or videos that go into the compromises in more detail?

and as you and Elp and many others have told me, I need more hours on a target.

ngc6888 for rest of July or 10 hours total is my current goal.

[TiffsAndAstro]

1 hour ago, chrisecurtis said:

That's a valid goal for making an astrophoto, but you might achieve it more quickly if you break down, even just a little, what you mean by 'nice' for this specific target. The images you posted look nice enough to me: is it that you are seeking to improve detail, have smoother, less noisy background, more contrast and fine detail in the central area, tighter stars, more saturated colour, a tighter crop on the galaxy etc. etc. It's tempting to say "all the above" but working on some of them will tend to make others worse (e.g. reducing noise and increasing the finest detail is tricky) so you often have to make choices. More integration (e.g. doubling from 90 minutes total to 3 hours) will help a lot, and more might help a bit more, but when you are experimenting, remember to experiment with a goal in mind. Otherwise, you might go down too many rabbit holes and forget what you were trying to achieve in the first place (been there, done that)

Ty for all this btw

[chrisecurtis]

Charles Bracken's "Deep Sky Imaging Primer" is excellent - not cheap but I've used my copy (2nd edition) a lot for years and still go back into it regularly to clarify things in my mind. The detailed techniques and "how-tos" go out of date fairly quickly but he always explains why things are as they are (right down to the quantum level in places!) which stays valid.

[TiffsAndAstro]

28 minutes ago, chrisecurtis said:

Charles Bracken's "Deep Sky Imaging Primer" is excellent - not cheap but I've used my copy (2nd edition) a lot for years and still go back into it regularly to clarify things in my mind. The detailed techniques and "how-tos" go out of date fairly quickly but he always explains why things are as they are (right down to the quantum level in places!) which stays valid.

Ty I'll have a look for it. 

Let's see what I think if I manage to get 10 hours ish on NGC 6888

Btw I used to hate my stars, then I hated them after I fixed them with full resynthesus.

Now all I do on star mask is  a unpurple in gimp, +20 asinh from linear in siril and a slight GHS when combing back with starless. They seem ok and have some colour.

Red stars seem very rare so I'm a tad dubious, but fine for me I think

Edited July 19 by TiffsAndAstro

[TiffsAndAstro]

3 hours ago, Elp said:

Having done a number of different targets it's a general consensus to do around 5-10h, but a lot depends on the target being imaged, your optics and what you're looking to achieve.

I can technically do a finished image within 2h with my F2s but I don't, trying to do the same with an F6/7 will not give the same result, so needs more hours. Trying to capture very faint signal (gasses rarely seen in the majority of images, galaxy tidal tails for example) need more time, if you're not concerned about such things then project time can be dialled back.

You still have to image for time to reduce the noise level in signal and background sky, more so if imaging from a LP area, and by doing so you can sharpen your detail better too.

Yeah I think my f5.8 is slower than I realised. Doesn't make it bad (I have no other experience of scopes just my crap vintage camera lenses) but I quite like it weird disappearing tube slop aside. Just means need more time on target. Could be 3.5 hours non astro dark tonight except clouds.

If I had a Rasa, it would still be clouds being the bottleneck.

[Elp]

You need to learn to temper your "eagerness" and listen to Gold Five when he says to Luke Skywalker:

"Stay on target!"

 

[Elp]

I was like you at first, trying to image everything and anything. Now ive got more than half a dozen, probably more which needs data which I add to per year, stack and see the difference.

My excuse though is I have a lot of options, so I'm usually testing more than getting data, but for the longer projects is not unusual for me to do 20-30 hours per target (usually narrowband so aim for 6-10 hours per band, then add in some RGB). The short projects are normally the low targets where there's minimal imaging time, but even these I add data every year if they're interesting.

Autumn/Winter is best when you can setup in the evening and pack up around midnight and still get good rest, or image until dawn.

Edited July 19 by Elp

[TiffsAndAstro]

2 hours ago, Elp said:

You need to learn to temper your "eagerness" and listen to Gold Five when he says to Luke Skywalker:

"Stay on target!"

 

I was looking at M16 on telescopious. Very low maybe viewable between houses, but I'm gonna try stick with ngc6888

Too cloudy tonight

Did take more darks for my library last night.

[TiffsAndAstro]

1 hour ago, Elp said:

I was like you at first, trying to image everything and anything. Now ive got more than half a dozen, probably more which needs data which I add to per year, stack and see the difference.

My excuse though is I have a lot of options, so I'm usually testing more than getting data, but for the longer projects is not unusual for me to do 20-30 hours per target (usually narrowband so aim for 6-10 hours per band, then add in some RGB). The short projects are normally the low targets where there's minimal imaging time, but even these I add data every year if they're interesting.

Autumn/Winter is best when you can setup in the evening and pack up around midnight and still get good rest, or image until dawn.

In winter I'm hoping to sleep in-between pressing play and doing flats at 6am