M33 posted to help me learn.

[Rustang]

I'm posting this image here for my learning and progression. I will start with my set up, most of you may know this already: HEQ5 Pro mount, SW80 ED DS PRO scope, Modified Canon 600D.  5min guided subs ISO 800, calibration frames.

So I'm not happy with this image, its pretty pants and not my best processing but its enough for this post as I would like to clarify some reasons why I'm not happy with it! I would like to understand where I stand with my equipment and its limitations on certain targets. I feel this is a difficult target, is it? I fully understand integration time makes a difference, with this particular image its only an hours worth of 5min subs as I had to bin a load due to unnoticed star movement issues. But with my current gear is more integration time really going to make a significant difference? You can clearly see that the camera mostly is struggling, noise, hot pixels, banding etc etc, more noticeable since doing longer exposures. I really wouldn't go zooming in on the images as it looks terrible and really does show my issues. I did a test by stacking the binned subs from the same session to bring the integration time up to 3hours (the subs were good enough just to test this) and it didn't make any significant difference, as i say i know more integration time is important but what is that really going to do for me with this particular camera?. My nebula images seem to come out better with these problems but galaxies really show where the camera is struggling. Or is it just me, can I/ is it possible to get better images from this particular camera on targets such as M33? Is it my post processing? I've certainly got some improving to do but I do feel that with my current processing skills, they are enough to show that its possibly the camera that's letting things down along side some processing improvements. One thing I just cant get right is the strong blue colors that galaxies such as this and Andromeda seem to have in them, Ive tried many ways but just cant get that color, could it be the over powering red from the modified camera? the colors do seem balanced correctly though from the stacked image. The washed out light, orange/brown color is just really difficult to change, again is this because of the camera or the post processing!?

It would just be nice going forward to know that I'm not flogging a dead horse/ trying to polish a turd!! Stick with this camera or move on? You know me, I like to get the most out of my gear but I honestly feel the camera is now letting me down, please help me confirm this! 🙂

Edited October 12, 2020 by Rustang

[MarkAR]

From the close up there does seem to be a noise issue. Are you dithering between subs ?

[Rustang]

2 hours ago, MarkAR said:

From the close up there does seem to be a noise issue. Are you dithering between subs ?

Noise has always been a problem, one of those things with an un cooled DSLR and a DSLR at the cheaper end!. My thoughts about more integration time is that am I only going to be stacking more of the same issue!? I have starting dithering but I'm not 100% sure how effective its being, I've not fully got my head around what setting's I should have in APP to do so properly with my gear. How long should a standard dither last? From what I'm seeing my dithers go from approx 20secs start to finish to timing out sometimes, if they time out I'm guessing they haven't worked that time? It can run ok for a period then sometimes the dithers upset the tracking a bit to much so I turn it off. So all in all, yes I'm dithering but whether its set up properly or making a different I'm not sure but then when I think about it, I dont think I'm getting the "walking" noise as bad as Ive seen in previous images, just the noise you can see above!. Even with out the noise, the clarity and sharpness and colours are definitely not great, definitely more so with galaxies with this camera. I dont have the money to go for a dedicated camera so will have to try and soldier on but I am feeling I'm reaching the cameras limits but would really like some clarification on this.

I apologise for always being on here asking lots of questions, I promise you I'm not being lazy, I honestly struggle with the vast amount of technical intake of information needed to do this and putting it all together. Its one of the worst hobbies for someone like myself with my slow little brain and is very challenging with many, many things to have to think about. I have to battle with myself sometimes to keep going with it and not jack it all in as the more it seems to progress to more I feel I need to take on, but I'm going to keep going as the results can be worth it even if it does make me go insane! 😊

[MarkAR]

Not sure if I can help, dithering should only take a few seconds. Don't know anything about APP, might be some settings that need changing.

Maybe if you take shorter subs the read noise might reduce a bit.

[alacant]

2 hours ago, MarkAR said:

dithering

Hi

Nice image. I'm with @MarkAR regarding the dither. 

Lights, bias and flat frames but without darks. Stack with a clipping algorithm and take as many frames as you can. 

HTH

[Budgie1]

I'm about to go the same route and get a modded  EOS 1300D, currently using an unmodded 2000D. So I'm still experimenting with what can be done and what gets the best results.

I see for this image you've gone for 800 ISO and 5 minute exposures. Have you tried increasing the ISO but lowering the exposure time? 

I know that generally increasing the ISO will increase noise but if you're lowing the exposure time, meaning you can get more exposures in the same time period, then it may work to eliminate or reduce the noise.

Which software are you using for the post processing? I've just got a trail licence for Pixinsight and that's made a big improvement on the DSS autosave files I previously processed with Photoshop.

Also, you say you're using APP for the dither, do you mean APT?

[Rustang]

24 minutes ago, Budgie1 said:

I'm about to go the same route and get a modded  EOS 1300D, currently using an unmodded 2000D. So I'm still experimenting with what can be done and what gets the best results.

I see for this image you've gone for 800 ISO and 5 minute exposures. Have you tried increasing the ISO but lowering the exposure time? 

I know that generally increasing the ISO will increase noise but if you're lowing the exposure time, meaning you can get more exposures in the same time period, then it may work to eliminate or reduce the noise.

Which software are you using for the post processing? I've just got a trail licence for Pixinsight and that's made a big improvement on the DSS autosave files I previously processed with Photoshop.

Also, you say you're using APP for the dither, do you mean APT?

Sorry, yes APT not APP, ! Its something I could look at in regards to higher ISO, shorter subs so may give that ago next session, I'm concerned it may add noise but hopefully the shorter exposure will counter act that so will just have to test, Ive got nothing to lose! Im post processing in PS, stacking in DSS until I can buy APP for stacking. What differences have you noticed with using Pixinsight?

[Budgie1]

Although Pixinsight takes more time for the processes to complete (on my PC, maybe faster on newer ones ) the built-in process make quite a difference even with the default settings.

There is so much I have to explore but I think I'll be buying it once the trial has finished.

Here are a couple of images of NGC7000. The top one was my original process using Photoshop CS3 and the second is the same autosave.tif file put through Pixinsight and following a tutorial I found on here:

Now the same with M31, top is Photoshop, bottom in Pixinsight.

I'm still learning with this software and there is more to play with. Like you, I need to get better at removing the noise form the background, so more playtime required.

[rnobleeddy]

I tried this last week with similar equipment but only got about an hours worth of data, so I'd take your image! 

 

I am far from an expert but from what I've read/experienced

- choose the best ISO for your camera (e.g. http://dslr-astrophotography.com/iso-values-canon-cameras/)

- dithering helps rather than darks, but make it a big dither (~50 pixels). I'm not entirely sure if my setup has an issue (or it's driven by DEC backlash) but it takes my mount/PHD2/Indi/Ekos about 50s to dither. It's still a time saving on dark frames though! Dithering actually stopped working, for some reason, for a set of subs I took a few nights ago and the difference was clearly visible

- if you don't dither, take lots of darks (e.g. 32), and make sure they're at the same temperature

 

Someone else will probably have a better understanding, but longer exposures generally reduce noise, not increase it. AFAIK, shorter exposures would only help if you're dithering and you were limited by the types of camera noise that dithering works to eliminate, because you'd be averaging that noise out more.

[ollypenrice]

Firstly I think that many aspects of the processing are quite simply excellent. The background sky is beautifully flattened both in brightness and consistency of colour (with the colour good) and it is at a respectably light level. So many backgrounds are too dark, too blue, too red, too green etc. Not this one. Its great. Stars are small, tight and round. Not enough blue? Vlaiv, on here, has made the point that amateur imagers push for too much blue in spiral arms and I think he's right. I've certainly changed my own approach in the light of his arguments. Look at the Hubble team's colour: https://www.spacetelescope.org/news/heic1901/

As you've said, where the image starts to fail is in its small scale details. Would more data help? Enormously. Build in a large scale dither of about 12 pixels, take a lot more subs and use a sigma clip routine when stacking. I don't use a DSLR but think Alacant's calibration suggestions will work best. No darks, just bias. (DSLR darks are not adequately temperature matched.)

Olly

[Rustang]

Thanks guys, it apperas there may still be somethings to try before I decided to, and can afford to upgrade the camera 👍

[Rustang]

19 minutes ago, Budgie1 said:

Although Pixinsight takes more time for the processes to complete (on my PC, maybe faster on newer ones ) the built-in process make quite a difference even with the default settings.

There is so much I have to explore but I think I'll be buying it once the trial has finished.

Here are a couple of images of NGC7000. The top one was my original process using Photoshop CS3 and the second is the same autosave.tif file put through Pixinsight and following a tutorial I found on here:

Now the same with M31, top is Photoshop, bottom in Pixinsight.

I'm still learning with this software and there is more to play with. Like you, I need to get better at removing the noise form the background, so more playtime required.

Some noticeable differences so something to consider 👍

[ollypenrice]

5 minutes ago, Rustang said:

Thanks guys, it apperas there may still be somethings to try before I decided to, and can afford to upgrade the camera 👍

Not being able to afford it never stopped anybody on SGL from buying new kit...  

😁lly

Ps Pixinsight versus Photoshop? They are very different and the key thing is to pick the one which suits you and suits the way your mind works. I do far more in Ps than PI because I'm more comfortable in the Ps environment.

 

Edited October 12, 2020 by ollypenrice

[Rustang]

2 minutes ago, ollypenrice said:

Not being able to afford it never stopped anybody on SGL from buying new kit...  

😁lly

Your not helping! 😊 😉

[Rustang]

21 minutes ago, ollypenrice said:

Firstly I think that many aspects of the processing are quite simply excellent. The background sky is beautifully flattened both in brightness and consistency of colour (with the colour good) and it is at a respectably light level. So many backgrounds are too dark, too blue, too red, too green etc. Not this one. Its great. Stars are small, tight and round. Not enough blue? Vlaiv, on here, has made the point that amateur imagers push for too much blue in spiral arms and I think he's right. I've certainly changed my own approach in the light of his arguments. Look at the Hubble team's colour: https://www.spacetelescope.org/news/heic1901/

As you've said, where the image starts to fail is in its small scale details. Would more data help? Enormously. Build in a large scale dither of about 12 pixels, take a lot more subs and use a sigma clip routine when stacking. I don't use a DSLR but think Alacant's calibration suggestions will work best. No darks, just bias. (DSLR darks are not adequately temperature matched.)

Olly

That Hubble picture really helps settle the mind a bit ! 👍

[ollypenrice]

8 minutes ago, Rustang said:

Your not helping! 😊 😉

Course I am! I'm helping you buy that camera!!!  I may be calling on you to help me buy a a new CMOS camera soon...

🤣lly

 

[Rustang]

I'm finishing shooting weddings next year which I've been doing part time for a few years, once thats over I'm selling that gear which will then be time for upgrades but must be patient till then! Must be patient, must be patient......

[Budgie1]

9 minutes ago, ollypenrice said:

Ps Pixinsight versus Photoshop? They are very different and the key thing is to pick the one which suits you and suits the way your mind works. I do far more in Ps than PI because I'm more comfortable in the Ps environment.

I've used Photoshop for years with various things and I do like it but I also like having software that's on my own and PS now seems to be on a subscription service. Having only PS CS3 at the moment, things have moved on since this was released so I was looking for something else and I saw Pixinsight has a 45 days trail licence and thought I would get it a go.

For me, it seems to work quite well. Although it's still easy to over-do the processing and make something that looks too "processed". I do find it easier to create masks and find the effects more subtle than in PS, but there are things I know I can do easily in PS that I haven't found in PI yet, so it's swings & roundabouts.

[Budgie1]

3 minutes ago, Rustang said:

I'm finishing shooting weddings next year which I've been doing part time for a few years, once thats over I'm selling that gear which will then be time for upgrades but must be patient till then! Must be patient, must be patient......

I've got to get some flying practice in, I've been asked to do a wedding in August next year with my drone. They want aerial shots and footage! 😬

[wimvb]

43 minutes ago, ollypenrice said:

I may be calling on you to help me buy a a new CMOS camera soon...

🤣lly

 

Osc of course. 😋

@Rustang, & others:

Regarding dithering. If you plan to dither 15 pixels, and you have a guide rate of 0.5 sidereal, then dithering should take twice your imaging scale ("/pixel) in seconds. Ie, if you image at 1"/p and your guide rate = 0.5, a 2 seconds dither will move your mount 15 pixels. The 12 pixel dither recommendation comes from a talk by Tony Hallas on astrophotography with a dslr.

Dithering will never by itself lower the noise in an image. Only more data will do that. And then only when the exposure time is such that the read noise is burried in light pollution (photon) noise or thermal noise. For a non-cooled dslr at high ISO, you may run into dynamic range problems. Keeping the ISO down can be a good thing. But dithering will spread any fixed pattern noise, and break up so called walking noise.

Edited October 12, 2020 by wimvb

[vlaiv]

9 minutes ago, wimvb said:

Dithering will never by itself lower the noise in an image.

Not true.

Dithering is very beneficial for lowering noise in the image, although most people don't know this. This is especially true when small number of calibration subs is taken.

To best explain this we need to look at what is happening when dithering. Let's observe single pixel and part of target on that pixel. If we have perfect tracing, same piece of target always lands on same pixel - with dithering it is always different pixel.

This means that with perfect tracking stack of pixels for our part of target come from single pixel in each sub and consequently get calibrated always with same bias value - bias is in this case constant and does not average out beyond it already being stacked and averaged.

With dithering - target covers always different pixel and bias sub calibrating always different pixel will have always different value as residual after stacking is random - this makes bias noise that we inject back in the image much smaller than in above example.

I know this is poor explanation, but point is - with perfect tracking, bias that we use to calibrate is always the same with respect to target (not image) and can be mathematically "pulled in front of average" and noise from master bias pollutes final image after stacking. With dithering - it pollutes each sub "differently" (same with respect to image but differently with respect to target) and does so prior to stacking and when stacked - same thing happens as with every other noise source - it gets reduced with respect to signal (stacking improves SNR).

In fact - you can check this yourself - take any set of calibrated and dithered subs and do two stacks:

1. plain average stack after aligning subs

2. plain average stack before aligning subs

Now take empty patch of sky (try to avoid stars) and measure standard deviation. You will find that standard deviation is smaller in first case - noise in background is smaller than in case without alignment.

No alignment would be necessary when one has perfect tracking / guiding.

 

[Rustang]

49 minutes ago, Budgie1 said:

I've got to get some flying practice in, I've been asked to do a wedding in August next year with my drone. They want aerial shots and footage! 😬

I've worked with a couple of drone companies at weddings, sounded like I had a giant bee over my head alot of the time but they produced some great stuff 😊 

[Rustang]

44 minutes ago, wimvb said:

Osc of course. 😋

@Rustang, & others:

Regarding dithering. If you plan to dither 15 pixels, and you have a guide rate of 0.5 sidereal, then dithering should take twice your imaging scale ("/pixel) in seconds. Ie, if you image at 1"/p and your guide rate = 0.5, a 2 seconds dither will move your mount 15 pixels. The 12 pixel dither recommendation comes from a talk by Tony Hallas on astrophotography with a dslr.

Dithering will never by itself lower the noise in an image. Only more data will do that. And then only when the exposure time is such that the read noise is burried in light pollution (photon) noise or thermal noise. For a non-cooled dslr at high ISO, you may run into dynamic range problems. Keeping the ISO down can be a good thing. But dithering will spread any fixed pattern noise, and break up so called walking noise.

Cheers, I never fully got my head around the calculations for dithering and with that, the settings needed in APT, I will try and go over it again 👍

[Rustang]

25 minutes ago, vlaiv said:

Not true.

Dithering is very beneficial for lowering noise in the image, although most people don't know this. This is especially true when small number of calibration subs is taken.

To best explain this we need to look at what is happening when dithering. Let's observe single pixel and part of target on that pixel. If we have perfect tracing, same piece of target always lands on same pixel - with dithering it is always different pixel.

This means that with perfect tracking stack of pixels for our part of target come from single pixel in each sub and consequently get calibrated always with same bias value - bias is in this case constant and does not average out beyond it already being stacked and averaged.

With dithering - target covers always different pixel and bias sub calibrating always different pixel will have always different value as residual after stacking is random - this makes bias noise that we inject back in the image much smaller than in above example.

I know this is poor explanation, but point is - with perfect tracking, bias that we use to calibrate is always the same with respect to target (not image) and can be mathematically "pulled in front of average" and noise from master bias pollutes final image after stacking. With dithering - it pollutes each sub "differently" (same with respect to image but differently with respect to target) and does so prior to stacking and when stacked - same thing happens as with every other noise source - it gets reduced with respect to signal (stacking improves SNR).

In fact - you can check this yourself - take any set of calibrated and dithered subs and do two stacks:

1. plain average stack after aligning subs

2. plain average stack before aligning subs

Now take empty patch of sky (try to avoid stars) and measure standard deviation. You will find that standard deviation is smaller in first case - noise in background is smaller than in case without alignment.

No alignment would be necessary when one has perfect tracking / guiding.

 

Your a clever man but most of the time your knowledge is wasted on me as it goes right over my head! 😁

[Aramcheck]

On @alacant's suggestion of stacking without Darks, you can of course try this & compare the noise levels on a background area. I tried it once & found that with Darks included in the process, my background noise with my astromodified 600d appeared to be about half that compared with the same set of Lights stacked with only Master Bias & Master Flats.

BTW I also have the problem with Dithering timing out on occasions.

Cheers
Ivor