M13 - testing a bunch of new stuff, with a big thumbs up for Registar

[geoflewis]

I recently purchased 2 new laptops, one for observatory control and the other for image processing. I also purchased the more sensitive Lodestar X2 to replace my old original Lodestar guide camera and added Gary Jarrette's KISS guide camera focus device into the mix. Finally after many recommendations both from users here and elsewhere, I decided to buy Registar from Auriga Imaging to improve my channel alignment.

Here is the result of all that new gear and testing...

Registar has at last allowed me to get much better LRGB channel alignment producing much crisper and colourful stars without the rainbow of colours that I was always getting when viewed iunder zoom previously. This has enabled me to bring out the lovely golden/red stars that feature throughout the cluster, so Registar gets a big thumbs up from me ?. I'm glad to report that everything else worked well too, with the new obs laptop (a refurbished Dell i5 Win7) running all the astro gear perfectly. Gary's KISS focuser really made tuning the new guide camera to the main QSI583 sensor a piece of cake.

Cheers, Geof

 

Edited June 3, 2018 by geoflewis

[groberts]

Thanks Geoff, that's very interesting.  Like you I've seen the recommendations from others in favour of Registar but thought (a) what can be so wrong with DSS which I currently use and (b) the price.  I wonder if you have processed the image processed with another method whatever that is and what does it look like for comparison? Furthermore, what is your post processing - Photoshop?

Glad everything else worked well for you. I also bought the Lodestar X2 for guiding just over a year ago and am very impressed + very easy to use.

Graham

 

[geoflewis]

41 minutes ago, groberts said:

Thanks Geoff, that's very interesting.  Like you I've seen the recommendations from others in favour of Registar but thought (a) what can be so wrong with DSS which I currently use and (b) the price.  I wonder if you have processed the image processed with another method whatever that is and what does it look like for comparison? Furthermore, what is your post processing - Photoshop?

Glad everything else worked well for you. I also bought the Lodestar X2 for guiding just over a year ago and am very impressed + very easy to use.

Graham

 

Thanks Graham,

Yes I too found the price of Registar off putting, plus the trial version doesn't allow you save images and the provided tutorial was poor I thought. Registar does provide a merge RGB tool, but I couldn't see how to add L to that without saving the registered images, so in the end I shelled out with all my fingers and toes crossed....!!

I did previously process the M13 without Registar, so here it is...

Honestly that one didn't turn out too bad, but I spent a lot of time shifting the RGB channels around to get to that and even then when I tried any colour saturation the misalignment showed up.

My background comes from imaging with a DSLR so channel alignment was rarely a requirement, but I've been plagued with that since I switched to mono CCD LRGB about 18 months ago. The software that I use for image processing is ImagesPlus, which handles the initial stack of raw data very well, but I find that it doesn't align the 3 or 4 LRGB channels as accurately as I'd like. It does include colour channel transform and TSR (translate, scale align) tools, but I frequently can't get them to work to my satisfaction, sometimes spending hours fiddling around. Registar just did it first time in seconds.... ?. I will still use ImagesPlus to calibrate, align and combine my raw stacks for each channel and for most of my subsequent post processing, so Registar is a fairly expensive tool just to accurately align the 3 or 4 stacked channels, though it will also allow me to accurately combine data from different FOV, so I'm looking forward to doing that down the line.

Perhaps a better comparision is my M15 image from last year, which I have also reworked using Registar yesterday, so here is a side by side....

M15 Original version

M15 Registar version

The changes are subtle I know, but I think it was worth it. I had to crop the original a bit more as I couldn't fix the star misalignment at the edge FOV, but probably the most noticible difference is in the brightest stars where the original shows them somewhat distorted where the Luminence didn't fully align with the RGB, plus I could get more colour out of the Registar aligned version. I had to blur the RGB in my original M15 before layering the luminance as many of the RGB image stars were tricolour, an annoying 'feature' of lots of my images since I switched from DSLR (OSC) to mono LRGB...!!

Perhaps I could also push the luminence stretch a bit more in the latest M15 version, but so far I'm pleased with the much easier, faster and more accurate alignment provided by Registar (despite the cost ).

I hope that helps.

Geof

Edited June 3, 2018 by geoflewis

[groberts]

Thanks Geof, again that's very helpful. 

Sounds like we're on a similar journey as I moved to mono + filters at the beginning of last year with a ZWO1600MM-Cool camera.  I am in general quite adept with a DSLR camera and now it's processing requirements in astrophotography but have been surprised at the difference with the new - hitherto unkown - challenges with the mono camera, of course, ultimately with it's benefits too!          

Graham

[Hallingskies]

I have exactly the same problem with alignment of colour stacks and have also contemplated going down the Registar route.  I do wonder what the root causes are.  I manually refocus between filter changes (and the filter wheel is also manual) so I am motorising both to see if that eliminates image shift.  I also wonder if the slightly imperfect colour correction of my old Vixen refractor is having an effect as well.  I use the image registration feature in Astroart to align my stacked channels; sometimes this is quite effective, other times, not so.  If all else fails, it could well be Registar for me as well.  Your results look quite unequivocal.

[geoflewis]

14 minutes ago, groberts said:

Thanks Geof, again that's very helpful. 

Sounds like we're on a similar journey as I moved to mono + filters at the beginning of last year with a ZWO1600MM-Cool camera.  I am in general quite adept with a DSLR camera and now it's processing requirements in astrophotography but have been surprised at the difference with the new - hitherto unkown - challenges with the mono camera, of course, ultimately with it's benefits too!          

Graham

 

14 minutes ago, Hallingskies said:

I have exactly the same problem with alignment of colour stacks and have also contemplated going down the Registar route.  I do wonder what the root causes are.  I manually refocus between filter changes (and the filter wheel is also manual) so I am motorising both to see if that eliminates image shift.  I also wonder if the slightly imperfect colour correction of my old Vixen refractor is having an effect as well.  I use the image registration feature in Astroart to align my stacked channels; sometimes this is quite effective, other times, not so.  If all else fails, it could well be Registar for me as well.  Your results look quite unequivocal.

Hi both,

I've been aware of Registar for about 5 years, with some folks on the ImagesPlus user forum swearing by it even for DSLR stacking, but especially for combining data from mutiple imaging rigs, e.g. to add higher resoltion data for the main features in a widefied image such as M81 & M82. The annoyance of 'nearly but not quite' channel alignment with mono LRGB channels finally got the better of my astro pockets....!! ??. Given that I've got literally £'000s tied up in astro equipment and based on the 2 images I've reprocessed so far, I'm confident that the additional £140 will prove to be a wise investment.

Geof

[MikeODay]

Lovely images Geof.  Thanks for sharing.

[geoflewis]

1 hour ago, MikeODay said:

Lovely images Geof.  Thanks for sharing.

Thanks Mike

[jock1958]

Fantastic images Geof, looks like you've opened a box of sparkling jewels ?

[swag72]

Such a massive difference ......... Clearly a very beneficial addition to your imaging software  

[geoflewis]

2 hours ago, jock1958 said:

Fantastic images Geof, looks like you've opened a box of sparkling jewels ?

Many thanks, that's a very nice way to think of them. That said the amount I've spent on this hobby I could indeed have purchased a very large box of sparkling jewels - just don't let my wife read this....??

[geoflewis]

2 hours ago, swag72 said:

Such a massive difference ......... Clearly a very beneficial addition to your imaging software  

Thanks Sara, I thought so too, but I wasn't sure if I was just convincing myself into justifying the expenditure.... ?

[geoflewis]

Having seen some superb versions of M13 recently I was thinking that I need to have another try at it, but before doing that I decided to revisit the data that I captured back in 2018 to see if I could extract any more, so here is the result.

Most of the processing is the same as in 2018 other than I really pushed the core with a much stronger use of curves, holding back the background with some NR and levels. My previous version is at the top of this thread. No new data has been added, so it would seem that there was more in the original data than I thought, hence I'm not sure whether I need to shoot this target again.

What do you all think? Did I push the core a tad too much, I think the stars are still resolved, but maybe borderline?

[DaveS]

Gets a thumbs up from me.

Yes, the core is bright but, IMHO not burnet out and I'm seeing resolved stars. For me that looks to be the best of the lot so far.

[geoflewis]

Just now, DaveS said:

Gets a thumbs up from me.

Yes, the core is bright but, IMHO not burnet out and I'm seeing resolved stars. For me that looks to be the best of the lot so far.

Thanks Dave, I took a look it after seeing yours and John's @silentrunning recent versions, as I they got me wanting to shoot it again, or maybe add some more data to it. On looking again I thought it looked fairly dull, so I reprocessed it. I recall being pleased with it 2 years ago, but am now realising that I was maybe too cautious, or too gentle in my processing back then and am now more confident to push harder. I'm now not sure that I need to shoot it again - though it's always one of my favourite targets.

[silentrunning]

Wow, that does look awesome Geoff, wonderfully sharp. To my eye the core is spot on not over cooked at all. It depends on monitor settings. I left the core very bright on my recent image so as not to give the cluster a flat look but the core in my image does look a bit burnt out dependent on the angle my eye is to the monitor! If I look straight on it suits my eye and it's clearly not saturated with the very bright core being resolved.

I only ever used PI for all processing apart from final touch up in PS. No ones mentioned PI on here, is it a bit shunned upon!

 

 

[geoflewis]

3 minutes ago, silentrunning said:

Wow, that does look awesome Geoff, wonderfully sharp. To my eye the core is spot on not over cooked at all. It depends on monitor settings. I left the core very bright on my recent image so as not to give the cluster a flat look but the core in my image does look a bit burnt out dependent on the angle my eye is to the monitor! If I look straight on it suits my eye and it's clearly not saturated with the very bright core being resolved.

I only ever used PI for all processing apart from final touch up in PS. No ones mentioned PI on here, is it a bit shunned upon!

Thanks John, I usually shoot 2m subs for globular clusters, as I find the data easier to process without blowing out the core, but then I'm also shooting bin 2x2 due to oversampling with the C14 and QSI camera. This was 9x2m each RGB and 13x2m L, so less than 1.5 hours total integration. I believe that the cluster reaches out even further than shown here, so that would be a good reason to shoot more data, or maybe up the sub exposures to 3m (or a combination of both).

I thought that a lot folks on here use PI and I've thought about heading that way too, but actually I like the software I'm using. Most of images are processed with ImagesPlus (IP not PI) with some final tweaks in PS, e.g. GradX, HLVG and maybe some of Noel Carboni's tools.

[Xiga]

On 16/05/2020 at 21:10, geoflewis said:

Having seen some superb versions of M13 recently I was thinking that I need to have another try at it, but before doing that I decided to revisit the data that I captured back in 2018 to see if I could extract any more, so here is the result.

Most of the processing is the same as in 2018 other than I really pushed the core with a much stronger use of curves, holding back the background with some NR and levels. My previous version is at the top of this thread. No new data has been added, so it would seem that there was more in the original data than I thought, hence I'm not sure whether I need to shoot this target again.

What do you all think? Did I push the core a tad too much, I think the stars are still resolved, but maybe borderline?

That's amazing Geoff. Very well processed i must say, and i don't think it's been pushed too far at all. 

You've really made the most of the C14's aperture here. The cluster extends really far due to all the fainter outer stars, and crucially, without having to over-stretch the data to show them (as most of us have to resort to) so bravo! 

[Guest]

Given my first ever imaged DSO was M13 last night I have now have a yard stick of where I want to be in the future. Stunning image sir!

[geoflewis]

8 minutes ago, Xiga said:

That's amazing Geoff. Very well processed i must say, and i don't think it's been pushed too far at all. 

You've really made the most of the C14's aperture here. The cluster extends really far due to all the fainter outer stars, and crucially, without having to over-stretch the data to show them (as most of us have to resort to) so bravo! 

Thanks Ciaran, I really appreciate the feedback. Cheers, Geof

[geoflewis]

1 minute ago, Chris Wright said:

Given my first ever imaged DSO was M13 last night I have now have a yard stick of where I want to be in the future. Stunning image sir!

Thanks Chris, this hobby is quite the ride, so I hope that you enjoy it as much as me. As you may have noted, 2 years on from when I originally captured the data, I was able to process it far better this time around. Cheers, Geof

[Guest]

38 minutes ago, geoflewis said:

Thanks Chris, this hobby is quite the ride, so I hope that you enjoy it as much as me. As you may have noted, 2 years on from when I originally captured the data, I was able to process it far better this time around. Cheers, Geof

I am well and truly addicted! Having had a keen interest in astronomy and astrophysics for the best part of 30 years, it is the current situation we are in that has been the catalyst for looking at the practical side of a theoretical interest. Only three weeks ago I purchased a Celestron 130 SLT and since then I have built up a modest astrophotography setup. I can say with all honesty I am loving it and cannot wait for the next opportunity to get into the garden with clear skies. My very brief journey has been aided with such a wonderful, welcoming community.

[x6gas]

Lovely captures and processing.

I must confess that I bit the bullet and bought RegiStar ages ago because at that point it was the only software I could find that allowed me to register images of different scales captured on different kit.  It is eyewateringly expensive for a one trick pony but as has been pointed out time and again - and proven in this thread - it does that trick really well.  Now all that said I do tend only to use it in special cases but your results will spur me to do some comparisons of alignment using RegiStar, AstroArt and PixInsight to see what the difference is.

Interesting thread, thanks.

[geoflewis]

6 minutes ago, x6gas said:

Lovely captures and processing.

I must confess that I bit the bullet and bought RegiStar ages ago because at that point it was the only software I could find that allowed me to register images of different scales captured on different kit.  It is eyewateringly expensive for a one trick pony but as has been pointed out time and again - and proven in this thread - it does that trick really well.  Now all that said I do tend only to use it in special cases but your results will spur me to do some comparisons of alignment using RegiStar, AstroArt and PixInsight to see what the difference is.

Interesting thread, thanks.

Just to clarify, I only use Registar to align the already calibrated, graded, aligned and combined individual channel stacks. ImagesPlus does a perfectly adequate job of that, but for some reason it didn't accurately align the 3, 4 or sometimes 5 HaLRGB stacked channels when building the final colour image, resulting in rainbow stars. Registar completely solved that for me.

[x6gas]

8 minutes ago, geoflewis said:

Just to clarify, I only use Registar to align the already calibrated, graded, aligned and combined individual channel stacks. ImagesPlus does a perfectly adequate job of that, but for some reason it didn't accurately align the 3, 4 or sometimes 5 HaLRGB stacked channels when building the final colour image, resulting in rainbow stars. Registar completely solved that for me.

Ah OK - that's interesting.  I wonder if there would be a further improvement if you used RegiStar for all of the subs alignment?  Be interesting to see if there was a perceptible difference...