Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

DSLR vs CMOS - Is it THIS easy?!?!

  • Please log in to reply
15 replies to this topic

#1 wumpus

wumpus

    Lift Off

  • -----
  • topic starter
  • Posts: 7
  • Joined: 30 Aug 2018
  • Loc: Florida, USA

Posted 21 January 2021 - 11:26 PM

Although I shot some (very basic) astrophotos back in the 80s (anybody else remember when Kodak came out with Ektar 1000?!?), I only got back into the hobby 2 years ago.  Starting with a NexStar 6SE and a D3400, I was amazed at what stacking and stretching could do - even with short exposures on an alt-az mount.  2 years later, and I've just picked up a AVX mount and started doing some guided shots.  So far, I'm VERY happy with the mount, tracking, and guiding.  Though, now that I can shoot longer, I'm starting to see amp glow from the DSLR showing up.

With that out of the way, I'll get to the point of this post.  I *KNOW* that a CMOS camera is better than an unmodded DSLR.  But I've also seen plenty of folks using DSLRs here, on Facebook and Astrobin, so I was expecting there was room to improve my game.  I mean, there's even a recent article here on CN about shooting with an unmodded SLR.  But then I saw a post on a Facebook group, and it either beggars belief or makes me think all of the "you have to buy a CGEM" folks are getting kickbacks from the vendors.

Attached is an example image, supposedly taken with an SLT 130, on the stock alt-az mount, using a 294MC Pro, under Bortle 6 skies (and with patio lights galore).  But the part that I can't get over is the OP claims the mount wasn't tracking well so he had go with 1.8 seconds subs!  I can't get this much detail under Bortle 5 using 2 minute exposures!  When questioned by somebody else, who was also raising the "not buying it" flag, the OP said that because he's a professional photographer, he really understands Photoshop processing.  Maybe so, and I'll admit that getting a decent stretch is tricky, but I keep coming back to 1.8 seconds!!!!

With the AVX, I've been excited to be able to pull more detail out from the skyglow than with the 20 second subs I got with the SE mount.  And the 294MC is on my short list as the next bit of kit to add in a year or so (once I recover from the AVX purchase).  But if this shot (and others like it) is really that easy with a 294, I'm thinking I screwed the pooch in buying the AVX...

Attached Thumbnails

  • 140633372_10158565424738005_404116980360133107_o 2.jpg

Edited by wumpus, 21 January 2021 - 11:28 PM.


#2 dhaval

dhaval

    Vendor

  • *****
  • Vendors
  • Posts: 2,196
  • Joined: 21 Jul 2008
  • Loc: Round Rock, TX

Posted 21 January 2021 - 11:42 PM

A key component that is missing from your write up is - the total integration time. Using 1.8s subs is OK, as long as you get thousands of them and then stack them.

 

A lot of folks are doing "lucky imaging" - basically, taking thousands of subs that are really short - this not only removes the effects of earth's rotation, it also removes the effects of bad seeing. The downside to that is, you need a really good computer to be able to process that much data. 

 

But, it can be done.

 

CS!



#3 jonnybravo0311

jonnybravo0311

    Apollo

  • *****
  • Posts: 1,083
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 21 January 2021 - 11:43 PM

How many 1.8s subs were stacked together?

 

If he's saying that's a single 1.8s exposure, I would be seriously doubtful. See, I live under Bortle 6 skies. I targeted Andromeda as the very first DSO I imaged. I shot it completely untracked with just my G9, a 50-200 lens and a tripod. I took something like 300 * 2s subs. It didn't look like that. In fact, it looked pretty much like this:

 

first M31 - PI edit

 

Here's another one... this time it's 10 * 60s subs stacked and processed:

 

M31 10 minutes

 

The 294 isn't magic. It can't suddenly do in 1.8s what takes MANY times that in integration on some other camera.



#4 Gipht

Gipht

    Mercury-Atlas

  • *****
  • Posts: 2,624
  • Joined: 12 Nov 2016
  • Loc: Prescott Valley, AZ.

Posted 22 January 2021 - 12:13 AM

I don't think you made a bad decision in purchasing an equatorial mount.  People can produce good images with an alt/az mount, at least with bright objects like M31.   You can do much better than the picture you have shown.  Likely with the AVX, you will have to incorporate auto guiding to get your best performance from that mount.

 

One of the traps in AP is there is always better equipment and more money to be spent.  You can spend a ton of money in this hobby.

 

If you decide to buy an astronomy camera, I would take a look at the newer cameras like the ASI533MC.



#5 wumpus

wumpus

    Lift Off

  • -----
  • topic starter
  • Posts: 7
  • Joined: 30 Aug 2018
  • Loc: Florida, USA

Posted 22 January 2021 - 12:30 AM

I don't think you made a bad decision in purchasing an equatorial mount.  People can produce good images with an alt/az mount, at least with bright objects like M31.   You can do much better than the picture you have shown.  Likely with the AVX, you will have to incorporate auto guiding to get your best performance from that mount.

 

One of the traps in AP is there is always better equipment and more money to be spent.  You can spend a ton of money in this hobby.

 

If you decide to buy an astronomy camera, I would take a look at the newer cameras like the ASI533MC.

I know M31 is "easy," but he's also posted a couple shots of the Horsehead using 15 and 20 second exposures, and claims to have taken his first photo just 3 weeks ago.

 

Perhaps he's an Astrophoto Rainman?\

 

As to the 533, the sensor seems small for my scopes - 6" SCT w/6.3 reducer and a 6"/f5 Newt.  It may be great for the Dumbell or M13, but I'd have to do multi-shot mosaics for the Pleiades or M42.  That or drop another $1,000 on a short APO.  And that latter ain't gonna happen in this lifetime.


Edited by wumpus, 22 January 2021 - 12:47 AM.


#6 TelescopeGreg

TelescopeGreg

    Mercury-Atlas

  • -----
  • Posts: 2,680
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 22 January 2021 - 01:32 AM

Moving up in equipment from what you have now will take some money, but I don't think it's necessary.  Other than adding an autoguider, I recommend that you spend more time with what you have, working on (practicing) getting the most out of it.  You've still got a lot of headroom to grow into.

 

As others have noted, it's the total time that makes the big difference.  Longer exposures at lower ISO will bring down the noise in the individual frames, and 20 seconds isn't a long exposure.  I have that same AVX mount, and for several years a long exposure for me was 30 seconds.  Over time, I got better at finding the sweet spot for balancing my AVX (each is different), and got better at polar alignments so the Dec motor wouldn't have to run as much.  I got better at configuring PHD2's guiding parameters (guiding is required!).  I got better at configuring how my telescope was physically mounted on the rotisserie (hint: keep things centered and as compact as possible).  I did upgrade the imaging scope (impulse buy for Christmas), and spent some additional money on a lighter guide scope, but the main improvements in my imaging came from making lots of small adjustments and improvements in how things were set up.  I'm now taking 4 MINUTE subs with that same AVX that would only do 30 seconds before, and I didn't do any "hypertuning" of the mount itself.

 

Take a deep breath and keep the AVX and your DSLR.  Get an autoguiding solution; that's the only thing I would invest in right now.  Maybe a reducer to get a bit wider field of view and a faster optical path.  When you are truly limited by the equipment it will probably be that the DSLR's sensitivity to Ha will bug you enough that it will be time to upgrade to either a modified camera, or an astro model.  But until then, you've not hit the limits for what you can do with what you have.



#7 Stelios

Stelios

    Voyager 1

  • *****
  • Moderators
  • Posts: 10,857
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 22 January 2021 - 02:26 AM



I know M31 is "easy," but he's also posted a couple shots of the Horsehead using 15 and 20 second exposures, and claims to have taken his first photo just 3 weeks ago.

 

Perhaps he's an Astrophoto Rainman?\

 

As to the 533, the sensor seems small for my scopes - 6" SCT w/6.3 reducer and a 6"/f5 Newt.  It may be great for the Dumbell or M13, but I'd have to do multi-shot mosaics for the Pleiades or M42.  That or drop another $1,000 on a short APO.  And that latter ain't gonna happen in this lifetime.

Repeat after me: TOTAL INTEGRATION TIME. That's what (primarily) matters. 

 

Yes, you *can* get images like the one shown (nothing particularly special) with an Alt-Az mount and a LOT of patience. Or you can have dinner/read/play/take-a-nap while the camera is imaging on your equatorial mount. If you guide with your AVX, you can easily get MORE detail than the image you posted by taking 20 or 30 2-minute exposures. As for the amp glow, darks take care of it. Here's my first Astrobin image, taken with a CG-5 (not as good as an AVX) and a DSLR. get.jpg?insecure. Doublet objective, cheap DSLR, 2003 vintage mount, clueless photographer :)

 

It's a pretty terrible image, I was a beginner. This is a recent image of the same object.  Worth noting among the differences, other than the color (present in the 2nd image and lacking in the first) is the size, quality and quantity of stars. 



#8 wumpus

wumpus

    Lift Off

  • -----
  • topic starter
  • Posts: 7
  • Joined: 30 Aug 2018
  • Loc: Florida, USA

Posted 22 January 2021 - 12:12 PM

Repeat after me: TOTAL INTEGRATION TIME. That's what (primarily) matters. 

 

Yes, you *can* get images like the one shown (nothing particularly special) with an Alt-Az mount and a LOT of patience.

OK, so lets look at total time.  First image is his first try at the Horsehead, 14 minutes total at 20 seconds each.  Second image is my best attempt to date, using 1 and 2 minute auto-guided subs for a total of 80 minutes.  Nearly 6 times the integration on a rig that costs 3 times as much and takes 5 times as long to set up.  And yes, I know that my unmodded camera is poor at Ha, but these aren't even in the same ballpark.

As to the "a LOT of patience" comment, I fully agree and I stumbled, failed, and succeeded my way along using a 6SE for 2 years, before finally moving up to CGEM and auto-guiding.  The guy here has been shooting for just 20 DAYS.

Attached Thumbnails

  • 136638062_10158536978513005_1876421514803548180_o small.jpg
  • Horsehead (Better Processing) Small.jpg

Edited by wumpus, 22 January 2021 - 12:13 PM.


#9 imtl

imtl

    Gemini

  • *****
  • Moderators
  • Posts: 3,041
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 22 January 2021 - 12:51 PM

What he claims might be perpendicular to reality. Don't just swallow what any other person claims or says online.


  • Stelios, pedxing and jonnybravo0311 like this

#10 dx_ron

dx_ron

    Vostok 1

  • *****
  • Posts: 144
  • Joined: 10 Sep 2020
  • Loc: SW Ohio

Posted 22 January 2021 - 02:42 PM

Astrophotography is not a competitive sport smile.gif

 

But, there really is that much difference between a full-spectrum camera vs unmodified dslr *on H-alpha targets*.

 

Last night I was testing various guiding and balance issues, and along the way shot 48x1' of the same region. With an unmodified Canon 700D (rest of the setup in my .sig). The Flame Nebula is lovely(ish) while the Horsehead is barely visible at all yet (and the Horsehead is one of the brighter emission nebulae out there). I was really hoping to have M78 fully in the frame, too - oh well.

 

Attached Thumbnails

  • Orion_1_21_21_bkg_small.jpg

  • wumpus likes this

#11 wumpus

wumpus

    Lift Off

  • -----
  • topic starter
  • Posts: 7
  • Joined: 30 Aug 2018
  • Loc: Florida, USA

Posted 22 January 2021 - 03:10 PM

Astrophotography is not a competitive sport smile.gif

True (at least for some), but it's hard not to at least wonder.

 

I mean I was pretty happy with what I was getting with the stock SE mount.  But I could tell that - being stuck with short exposures - I'd be limited given my Bortle 5 skies.  No way to get past the sky noise at 20 seconds.  Now, with the AVX, I'm auto-guiding to about 1" and getting 60, 120, and even 240 second subs.  But, I've been disapointed that I've yet to have a "Wow! Look at the extra data!" moment.  Yes, there is "more," but not stunningly so.  And what there is, requires me to stretch so far, I end up getting noise or overcoming the darks so the amp glow shows up.



#12 imtl

imtl

    Gemini

  • *****
  • Moderators
  • Posts: 3,041
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 22 January 2021 - 03:18 PM

True (at least for some), but it's hard not to at least wonder.

 

I mean I was pretty happy with what I was getting with the stock SE mount.  But I could tell that - being stuck with short exposures - I'd be limited given my Bortle 5 skies.  No way to get past the sky noise at 20 seconds.  Now, with the AVX, I'm auto-guiding to about 1" and getting 60, 120, and even 240 second subs.  But, I've been disapointed that I've yet to have a "Wow! Look at the extra data!" moment.  Yes, there is "more," but not stunningly so.  And what there is, requires me to stretch so far, I end up getting noise or overcoming the darks so the amp glow shows up.

This will depend on what you are imaging, what you're trying to pull out of your data and your skills.

The wow factor that a lot of people keep mentioning is directly proportional to the parameters I'm mentioned. Especially image processing skills. If you (not you specifically) don't know what you are doing then no amount of data will help your final image. If you know what you are doing, you can make even mediocre data look stunning.


  • wumpus and RogerM like this

#13 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 160
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 22 January 2021 - 04:33 PM

And yes, I know that my unmodded camera is poor at Ha


Yes, it's worse by a factor of between 5x and 20x depending on what cameras you're comparing. It's a BIG DEAL to get Ha sensitivity.

#14 RogerM

RogerM

    Ranger 4

  • *****
  • Posts: 303
  • Joined: 13 Aug 2020
  • Loc: California

Posted 22 January 2021 - 05:22 PM

I hope this post will properly illustrate what Total integration time will do for imaging DSO.  Both of these are LRGB using the very same gear and location under Bortle 6 skies and not ideal seeing conditions being here on the coast...a few here on CN know what my location is like (I've posted a few shots of my primary viewing site before.  

 

Both sets of subs use identical 20 sec Lums and 60 sec RGB.  The difference is the first image has a total integration time of just 52 minutes while the second one I managed just over 4.6 hours (to date) spread over two nights of imaging.  Basic processing (you can tell because I've not even addressed halos nor star bloat amongst a few other things but you can see a big difference in what details begin to show up.

 

Wumpus, your image is a great beginning at only 80 mins!  Keep adding to your data set and you'll see improvements! smile.gif

Attached Thumbnails

  • M42_11-14-20_52mins.jpg
  • M42_01-17-21_277mins.jpg

Edited by RogerM, 22 January 2021 - 06:24 PM.

  • bobzeq25 and wumpus like this

#15 wumpus

wumpus

    Lift Off

  • -----
  • topic starter
  • Posts: 7
  • Joined: 30 Aug 2018
  • Loc: Florida, USA

Posted 23 January 2021 - 11:32 AM

Wumpus, your image is a great beginning at only 80 mins!  Keep adding to your data set and you'll see improvements! smile.gif

 Thanks, and I was fairly happy with it.  Especially for a non-modded camera.  And I don't plan to give up (especially since my scopes and clothes will be on the front lawn if I try to drop ANOTHER grand on a camera right after getting the CGEM)!

Still, I know that my DSLR sensor is fairly good and per the specs, it's better than some of the older, well trusted, Canon sensors.  So I'm struggling to believe that a 294MC is 100 times more sensitive.  1.8 seconds vs 180 seconds, with the same total integration (15 minutes), and I'm left at the starting gate.



#16 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 160
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 23 January 2021 - 01:10 PM

Subexposure time matters very little, it's all about total photons measured.

It's probably not JUST the camera. If you want to compare your image capture speed to someone else's you need to know the sky brightness at both imaging locations, the relative sensitivity of each camera, and the speed of the optics for both systems. All of these things multiply.

Lets do a really simplified bit of math using these factors: Lets put the camera response at a conservative 3x for the 294, your c6N is f/7 and the 130SLT is f/5 (approx 1 stop faster, so 2x), and lets say you're 1.1 sky magnitudes brighter(which is roughly the upper/lower limits for a b6 zone).

15 min * 3 * 2.52^(19.1-18) * 2 = 248 minutes for you to capture an equivalent image.

Things this is missing is the contribution of read noise, pixel size, and the reduction in FPN from cooling (and probably others, too), all of these advantages go towards the 294 based setup.

Anecdotally, I tried capturing the whole orion constellation last year from bortle 6 with my d5100 and 35mm lens at f/4, I got ~3 hours of data. The horsehead was a vague reddish smudge buried in the noise. This year I tried with my (new to me) Fuji XT-1, which has ~double the Ha response or so, and a 135mm lens at f/4 from bortle 4. In 30min the whole horsehead is apparent including the fainter area surrounding sigma orionis. This was with 30sec exposures.

Orion_135mm-RGB-session_1-1-mod-csc-lpc-cbg-St-1_1280.jpg

Edited by copper280z, 23 January 2021 - 01:11 PM.

  • jonnybravo0311 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics