Canon 60Da or 'full frame' better for astrophotography?

SnappyChappie · Jul 24, 2015

Hi all,

I've often wondered why Canon chose to use it's crop-sensored EOS 60D to adapt to a version for astrophotography - the 60Da - instead of choosing a "full-frame" sensored DSLR, like for instance, the Canon EOS 6D which is meant to be good in low-light situations in terms of lower noise and greater sensistivity compared to the 60D's sensor. Has anybody had experience of shooting the night sky with both these cameras? Which one did you prefer? Is there any advantage of the crop-sensor over the full-frame? EF or EF-S lenses? Primes or zooms? Which body mounts better on a telescope set-up?

I'd like to hear your opinions and experiences. Thanks.

swimswithtrout · Jul 24, 2015

Given the choice of any Canon body, I wouldn't even bother with an "a" version.

The one, and ONLY ONE reason that Canon had their brief period of popularity, amongst AP shooters, was that they were the only company that had released their SDK's so 3rd parties could implement their own software, for a long while.

Canon sensors are hands done, no questions asked, the worst sensors on the planet, in regards to AP. Their noise levels are out the roof compared to Nikon or Sony or anyone else, but hey, "If You Build it, They Will Come"...

Sony has resisted releasing their SDK's, but Nikon jumped into the fray a couple of years ago, and since Nikon outsources their sensors, they can pick the cream of the crops. Several models use Sony sensors, and their Toshibas are not far behind the Sony in noise characteristics.

At this point, Canon is a dead fish if you want an AP capable camera. Sorry.

starman1969 · Jul 24, 2015

Get a Nikon D810a

Tristimulus · Jul 25, 2015

SnappyChappie said:
Hi all,

I've often wondered why Canon chose to use it's crop-sensored EOS 60D to adapt to a version for astrophotography - the 60Da - instead of choosing a "full-frame" sensored DSLR, like for instance, the Canon EOS 6D which is meant to be good in low-light situations in terms of lower noise and greater sensistivity compared to the 60D's sensor. Has anybody had experience of shooting the night sky with both these cameras? Which one did you prefer? Is there any advantage of the crop-sensor over the full-frame? EF or EF-S lenses? Primes or zooms? Which body mounts better on a telescope set-up?

I'd like to hear your opinions and experiences. Thanks.

I have the Canon 60Da (keeping one and selling the other). The camera was good when released but sensor technology have evolved - and the Canon is left behind. Clearly no hand picked sensors in these cameras (not in mine at least).

Would never never never buy a new 60Da now at full price - but a used one for less than half price would be ok. The cameras can deliver VERY GOOD images of astronomical objects with some effort so they are FAR FROM useless - but as said above: sensor technology has evolved.

The only reason for Canon tho modify the 60Da is price - this was a popular astronomy camera at this price point. Now users are willing to pay more but also demand a LOT more perfomance for the money paid.

The Nikon 810a may seem expensive but is not for what you get. The 60Da may seem reasonably priced but is not (overpriced and underperforming).

For astronomy with a telescope sensor size is not the main reason to choose one sensor over the other (have the Atik dedicated astro camera with a tiny 10x12.5 mm sensor but performing extremely well for astronomy).

Full frame is the hot spot for camera geeks right now and also selling well. No reason per se to chose a FF camera over an APS-C camera for astronomy as long as the camera performs well at pixel level. That is what is important for astronomy.

geo444 · Jul 25, 2015

Hi,

that depends on what you want to do :

- for Wide field photo a Full Frame camera is ok...
but Not All the Telescopes have a 43mm Corrected image Circle !!
the very good NiKon D600 on a Basic Petzval Refractor :
www.dpreview.com/galleries/7467909648/photos/3067558/
> its SNR at ISO 200 is so high that you can partly discard the Green Layer !

- for Deep Sky Small Objects, take a look at Sensor Low-Light ISO rating per mm² :

CanonAPSc=5,33 - Eos 60d, 60d-Astro & 600d = ISO ~800 / 15.0 x 22,5 mm - 2,37 iso / mm²
Sony=APSc=4,40 - Sony Exmor APSc 16 MPix = ISO ~1200 / 15,6 x 23,5 mm - 3,27 iso / mm²
Sony=APSc=6,55 - Sony Exmor APSc 24 MP = ISO ~1365 / 15,6 x 23,5 mm - 3,72 iso / mm²
Sony=1"FSi=17,2 - Sony Rx100 1st FSI 1" = ISO ~390 / 8,8 x 13,2 mm - 3,36 iso per mm²
Sony=1"BSi=17,2 - Rx100² NX mini BSI 1"= ISO ~505 / 8,8 x 13,2 mm = 4,35 ISO / MM²
Sony=1/2.3=44,0 - Pentax Q & Q10 BSI = ISO ~186 / 4,6 x 6,2 mm = 6,52 ISO / MM2
Sony=1/1.7=29,0 - Nik P340 PQ7 BSI = ISO ~273 / 5,6 x 7,5 mm = 6,50 ISO / MM2
PanSony4:3=7,10 - PanaSony 4:3 = ISO ~765 / 13 x 17,3 mm - 3,40 iso / mm²

Pixel Density MPix/cm² --- ISO ~ DxOmark average Low-Light ISO rating

No surprise : the Best, most Sensitive Sensors are BSI CMos !
... Pentax Q are unusable due to 30 sec. Bulb Exposure

The Samsung Nx mini (~330€: MA4NXM adapter included) is really very sensitive
removing its internal IR-cut Filter costs ~75€ only, then you can use it either
- for High Speed IR photography = AF still working with the 9mm prime :
www.chassimages.com/forum/index.php/topic,236962.msg5462427.html#msg5462427
- for Astro-Photography + H-Alpha High sensitivity, here 13 Subs only :
www.chassimages.com/forum/index.php/topic,236962.msg5469256.html#msg5469256
+ the 2.7 crop factor is very interesting for Deep Sky Small Objects :
www.chassimages.com/forum/index.php/topic,236962.msg5471209.html#msg5471209

BSI (CCD) Sensors : a Must-Have for professional (research) Astro-Photographers...
... but they are so E X P E N S I V E... that... it's just a dream... until...
now... that you can have one for a ridiculous price !

--
http://en.astronomike.net/m/6809.html
http://en.astronomike.net/m/134251.html

just Tony · Jul 25, 2015

If I was to get an H-alpha capable camera today my choice would be to convert a D750.

What I like:

A superb sensor regarding sensitivity and noise
Flip screen (the camera is typically pointed up in astrophotography; go figure!)
The price. I could even consider using a gray market 750 because the conversion would probably terminate the manufacturer warranty.

What I could live with:

24 MP (my other cameras have 36). Well realistically, with very long exposures you will end up with some accumulation of atmospheric blurring anyway.

mermaidkiller · Jul 25, 2015

Well, the 6d already is two stops ahead in sensitivity / noise compared to the 60d while it is 2 steps behind in Halpha so this cancels out each other. The standard filter attenuates Halpha transmission to about 25% which equals 2 stops.

The D750 is at least as good as the 6d, costs about $2250, but, indeed the D810a is the ultimate choice when you can afford it ($3800).

Note, that Nikon bodies do not mount Canon lenses but the other way around it does with a cheap adapter for about $30. You'll lose AF and shutter-priority control, but that is no issue in AP.

You can always crop full frame images, e.g. when a telescope does not produce sharp images to the edges (only $4000+ scopes like Takahashi or Televue APOs do that), but you can never widen the fileld of an APS-C photo.

Consider other Red-Yellow (Canon-Nikon) differences:

* Nikon has better dynamic range, i.e. ISO 200 photos can be pushed to 6400 while retaining fine grain
* Nikon FF cameras do mount DX (crop) lenses and then act as a crop camera. Canon does not allow this.
* For Canon fullframe (and APS-C) clipin filters from Astronomik and Hutech (APS-C only) exist which eliminate the need for expensive large front screw-on filters on telephoto lenses and the inability at all to use filters on wideangle lenses. These filters can be used with ANY lens or telescope, even 14mm wideangles.
When you are only photographing via a telescope this is not an issue.
* The D750 / D810a have a tilt and flip screen.
* Whe you have an Android device, you can control the camera easily with DSLR controller ($7 Play store) which eliminates the need for a laptop to control the camera in the field. You can even focus on the smartphone / tablet screen. This is much more comfortable than the rear screen even when it is a flip screen. This unfortunately does not support Nikon.

--
Ricoh KR-5 ... Pentax ME Super ... Canon T90 ... ... ... 40d ... 7d ... 6d

W5JCK · Jul 25, 2015

I had a Canon 60D (not the "a" model) until last summer. After getting a Sony a7 and Sony a6000 I realized just how lackluster the Canon 60D was at AP. I wouldn't waste my money on any Canon APS-C model for AP. I probably would go with a Sony or Nikon if you want to use a FF as opposed to the Canon 6D, and definitely a Sony or Nikon if you want an APS-C camera.

I have both FF and APS-C. Which I use depends on the circumstances. For wide field nightscapes I prefer the FF. For telephoto AP I prefer the APS-C.

As far as getting an "a" model goes, if you are going to mostly take wide field nightscape photos then don't waste your money on an "a" model. If you are going to photograph a lot of nebulae, then an "a" model might be worth the huge amount of extra cash, but I'm not convinced it is.

Marco Nero · Jul 25, 2015

SnappyChappie said:
Hi all,

I've often wondered why Canon chose to use it's crop-sensored EOS 60D to adapt to a version for astrophotography - the 60Da - instead of choosing a "full-frame" sensored DSLR, like for instance, the Canon EOS 6D which is meant to be good in low-light situations in terms of lower noise and greater sensistivity compared to the 60D's sensor. Has anybody had experience of shooting the night sky with both these cameras? Which one did you prefer? Is there any advantage of the crop-sensor over the full-frame? EF or EF-S lenses? Primes or zooms? Which body mounts better on a telescope set-up?

I'd like to hear your opinions and experiences. Thanks.

I can't answer that first question but I can venture a guess. In order to coax budding photographers towards the Canon brand over the CCDs used previously for telescope imaging, it made sense to go with a lower cost DSLR rather than a much higher cost Full Frame DSLR. One of the reasons for this was that people were stacking images taken with the CCDs so it made sense to use the 60Da to do exactly the same thing. Any noise could be averaged out and image stacking produced more detail with shorter multiple exposures. Producing a Full Frame sensor is still considerably (and comparatively) expensive today (see Canon's White Paper on Full Frame CMOS Sensor production costs for an explanation). The 60Da was a movie in the right direction at the time.
.
Full Frame is certainly the way to go today as they can gather up to 5x the light for the same exposure time as an APS-C CMOS sensor. Some of the suggestions posted already are a step in the right direction. One benefit of using a cropped sensor is that you're image circle crop will result in a "magnified" view of sorts.
.
There wasn't a lot of choice out there until fairly recently. But something your post doesn't specify is whether or not you're looking to mount your camera to a telescope (for specific Deep Space photography of galaxies, nebula or comets....) or if you want to produce tripod-mounted star-scapes of the Night Sky or the Milky Way - more like Landscape photography.
.
I see a lot of professional journalists using the 5D and 6D series cameras for this sort of thing. They can take a sweeping shot of the milky Way over a major landmark or landscape and capture everything in a single frame (or two). In fact I'd say that the main limitation these days for landscape-style astrophotography is suitable lenses. I've seen nice examples from some Nikon cameras as well although the main choice on offer now appears to be the Sony FF A7S Mirrorless. It's rated as the best option for people doing this sort of thing although I believe the 6D was/is highly recommended for this purpose. Slightly less noise than the 5D variants.
.
As others have noted, Nikon outperforms Canon pretty much all the way at this time. However, if you want your camera to cost under $2000, the EOS 6D is recommended. The best camera for low noise and lens flexibility is without a doubt the new Sony A7S camera variants. But it may be priced above numerous other options.
.
There's a couple of very good pages here that might be of interest to you: They contain quite a few opinions on Astrophotography equipment, lenses, cameras etc.
.
http://starcircleacademy.com/2012/07/milkyway-2/
http://www.lonelyspeck.com/lenses-for-milky-way-photography/http://www.lonelyspeck.com/astrophotography-101/
.
You'll need to use some creativity and search for the Sony A7S for Astrophotography review because it was listed as a banned domain in URLs when I attempted to post the direct link here.
Just Google "sony-a7s-astrophotography-review/" and you should find it.

W5JCK · Jul 25, 2015

Marco Nero said:
As others have noted, Nikon outperforms Canon pretty much all the way at this time. However, if you want your camera to cost under $2000, the EOS 6D is recommended. The best camera for low noise and lens flexibility is without a doubt the new Sony A7S camera variants. But it may be priced above numerous other options.

just Tony · Jul 25, 2015

mermaidkiller said:
The D750... costs about $2250,

Gray market $1499. The $750 that would buy you a soon to be nullified warranty is nearly double the cost of a spectral conversion.

mermaidkiller said:
but, indeed the D810a is the ultimate choice when you can afford it ($3800).

A limited kind of ultimate, but yes. The same money could buy a cooled dedicated astro camera that will go far beyond anything that's built on a DSLR foundation. The tradeoff is that a PC would be needed in the field.

A converted gray 7540 costs about half of that but would be much more than half as good.

mermaidkiller said:
Consider other Red-Yellow (Canon-Nikon) differences:

* Nikon has better dynamic range, i.e. ISO 200 photos can be pushed to 6400 while retaining fine grain

We could call that the fundamental theme of this sub-thread hijack. ;-)

mermaidkiller said:
* Nikon FF cameras do mount DX (crop) lenses and then act as a crop camera. Canon does not allow this.

I wouldn't go that way. If you have it (big sensor), embrace it. More total photons collected = a less noisy image for the same final presentation size.

mermaidkiller said:
* For Canon fullframe (and APS-C) clipin filters from Astronomik and Hutech (APS-C only) exist which eliminate the need for expensive large front screw-on filters on telephoto lenses and the inability at all to use filters on wideangle lenses. These filters can be used with ANY lens or telescope, even 14mm wideangles.

Someday they'll figure out what they are missing by ignoring the more suitable sensors. Did they decide which way to go based on the children's soccer game market share?

mermaidkiller said:
When you are only photographing via a telescope this is not an issue.
* The D750 / D810a have a tilt and flip screen.

Not the D810A. However it does have the live view mode that is optimized for previewing long exposures, probably via longer integration times. The built in long exposure timing is nice but you can level that field for $40 with any camera that can accept an external trigger.

mermaidkiller said:
* Whe you have an Android device, you can control the camera easily with DSLR controller ($7 Play store) which eliminates the need for a laptop to control the camera in the field.

Laptop to control a DSLR in the field?

mermaidkiller said:
You can even focus on the smartphone / tablet screen. This is much more comfortable than the rear screen even when it is a flip screen. This unfortunately does not support Nikon.

This does: http://dslrdashboard.info (it's another field leveler)

just Tony · Jul 25, 2015

W5JCK said:
Another thing to consider though is lenses. If you have a bunch of good Canon lenses, then you probably want to be able to use them with any new camera you buy. Canon lenses are easily adaptable to Sony mirrorless cameras. I don't think Canon lenses can be adapted to Nikon DSLRs though. Best to check on that if it is an issue.

An adapter for Canon lenses for Nikon bodies would not preserve infinity focus. Kind of a buzz kill for astrophotography. A Sony body would be interesting but it's ironic that the better applications of Sony sensors are in Nikon bodies.

SnappyChappie · Jul 25, 2015

Thanks for all the information guys, definitely food for thought!

rnclark · Jul 26, 2015

Some important information:

First, the sensitivities (system throughput) of most DSLRs of recent generations are all pretty close, within 30 to 50 %. For example, BSI sensors can reach a QE of 90%, and recent developments in front side illuminated now have channels funneling light around the electronics in the silicon, plus the microlenses, so QE can be around 60 to 70%. The difference in BSI versus FSI sensors is narrowing. And the FSI sensors have a technical advantage in better on-sensor dark current suppression at a given temperature.

But QE of 60 versus 90 % does not have a big effect on long exposure astrophotography if dark current varies more. DXOs low light scores are also irrelevant to astro. For example, look of the Nikon D800(e) and the Canon 7D mark 1. I'm sure they are worlds apart in the dxo score (note the dxo scores are normalized to 8 megapixels). But then why in Figure 10 here: http://www.clarkvision.com/reviews/how-to-interpret-reviews/ does the 7D record about the same level of faint stars as a D800e? Simple, it is because their QE, read noise and noise from dark current are close, and not worlds apart as in the DXO scores.

Once one has a decent QE in a sensor, the key to astrophotography is low noise from dark current. Dark current noise determines the ultimate limit to faint object detection if you have dark skies. The latest Canon sensors excel at superb low dark current, with the 7D Mark II leading the pack with some 10x lower dark current:

Canon 7DII Review Clarkvision.com: Camera Review and Sensor Analysis Series, Canon 7D Mark II

Dark current has not been mentioned in this thread so far from what I've read, and that is a major oversight, invalidating the rants here (as if they had any validity to start).

So, if you want superb low light astrophotography performance, look for a recent model camera with the lowest dark current. (And note, all modern cameras have on sensor dark current suppression so one can't directly measure an increase in dark current with time--only the increase in noise. So some of the methods online saying how to measure dark current no longer work and many reported values using those old methods on new cameras are in error.)

I'll soon be adding more info to my 7D2 review. I ran some tests of 162 one-minute exposures stacked and the noise is impressively low: about 0.5 electron RMS per pixel! And that includes the effects of read noise and noise from dark current. The results produce an extremely uniform response across the whole sensor.

The other factor in astrophotpgraphy is post processing. I see a lot of poor post processing these days, in fact more poor that great. This has resulted in the common belief that one must modify a camera to see much H-alpha. Try my post processing challenge:

Astrophotography Image Processing 2, Clarkvision.com

The raw data are there and you can use your standard/favorite methods. Can you bring out the H-alpha with the methods you have been using for your astrophotos? You can follow the links to reddit and see what others have done, and all have failed to bring out much H-alpha. (Don't post your results here; I'll start a new thread and you can show us your results there.)

I'm sure I'll get flamed now....

swimswithtrout · Jul 26, 2015

rnclark said:
I'm sure I'll get flamed now....

No flame, but your reviews are always strictly Canon centric. Where's a review for the Nikon D5100, D5300, Sony A7s, etc. ??

rnclark · Jul 26, 2015

I am working on D800 and D810a reviews. A problem with evaluating such cameras is the processing of the raw data by the camera--that skews the results (makes them look better by sensor scores than they really are). Sony is even harder with their lossy compression (which can skew the other way too).

swimswithtrout · Jul 26, 2015

rnclark said:
I am working on D800

How long in the tooth is that camera and it's not even the prime choice for AP due to it's small sensor pixels.

Where's the reviews for the more commonly used Nikon D5100, with it's superior Sony sensor ?

rnclark · Jul 26, 2015

Small pixels are actually better. See that same figure 10 I referenced and compare the Canon 6D to the 7D and D800. Large pixels see more sky so have lower contrast with faint stars. Smaller pixels have smaller contribution from the sky and higher contrast on faint stars. Smaller pixels also have lower dark current. This also goes for fine detail, like a dark lane in a galaxy.

Given 2 sensors of the same physical size and same technology, but one with more pixels, the one with more pixels will be the better astro camera.

It is not intuitive until one considers all the factors in real-world imaging.

rnclark · Jul 26, 2015

I should add, that what I'm seeing in cameras from Nikon to Canon to even Sony with its lossy raw files, is all recent cameras are capable of great astrophotos. Some pull ahead mainly on two factors: smaller pixels (better) and low dark current.

Also, back to the DXO low light metrics: note the high score of the Canon 1DX. But the 1DX is so massive that the heat builds up and the dark current is very high. So camera mass can play a role too.

Forget read noise, QE and all the other commonly cited factors in DSLR astro cameras and concentrate on low mass for cooling, low dark current and small pixels and recent models.

Tristimulus · Jul 27, 2015

rnclark said:
I am working on D800 and D810a reviews. A problem with evaluating such cameras is the processing of the raw data by the camera--that skews the results (makes them look better by sensor scores than they really are). Sony is even harder with their lossy compression (which can skew the other way too).

Just a note: Sony cameras do aggressive spatial filtering in bulb mode effectively diffusing fine detail and burying the very faintest stars. Test exposures in ordinary mode (with the usual 30 sec exposure limit) and similar exposures in bulb mode tell the whole story.

Did a test with 8 sec exposures (having a bright sky in the summertime as beeing a northener living at 60 deg N) with ordinary settings and bulb setting with an intervallometer.

Made several exposures to elliminate spurious results.

Bulb mode and intervallometer 8 sec - Arcturus (note loss of faint stars)

Camera Setting 8 sec - Arcturus

Canon 60Da or 'full frame' better for astrophotography?

Forum Enthusiast

Senior Member

Senior Member

Veteran Member

Leading Member

Senior Member

Senior Member

Senior Member

Veteran Member

Senior Member

Senior Member

Senior Member

Forum Enthusiast

Senior Member

Senior Member

Senior Member

Senior Member

Senior Member

Senior Member

Veteran Member

Similar threads

Keyboard shortcuts