The 7D mark II is the best Canon DSLR for astrophotography.

[mbaddah]

According to the following:

http://www.clarkvision.com/reviews/evaluation-canon-7dii/

Dramatic improvement from the previous generation that's for sure.

[AstroJunk]

You might just have helped me make a decision :)

[zenith]

I just bought a Canon 6D, I decided to go with a full frame sensor. I Also considered the 7D mark II and also Nikon D750, as I really like the articulated screen and inbuilt flash.

Here are a couple of good websites showing noise in various situations.

http://www.imaging-resource.com/IMCOMP/COMPS01.HTM
http://www.brendandaveyphotography.com/?page_id=726

[gregbradley]

You also have to take into account that full frame has something like a 2 stop advantage over APS sized sensors merely from size of the sensor alone.

So images from both 6D and 7D2 side by side would be needed to make a more informed decision.

For example my Fuji XT1 (APSc) and Sony A7r show this difference. Both are very low noise. But 2 images at ISO6400 show the Sony as being many times brighter for the same lens/exposure. Its just real estate size of the sensor capturing more photons.

Greg.

[Camelopardalis]

I'm struggling to understand why this would be :confused2:

What happens when the object in question occupies roughly the same number of pixels on each chip?

[pluto]

Surely it's about pixel size and not sensor size. For example, a 8x8µm pixel will collect the same amount of light whether it's part of an APS-C or a FF sensor.
In general APS-C sensors tend to have smaller pixels than FF sensors but that doesn't have to always be the case.
EDIT: for example the fuji x100 with an APS-C sized sensor has slightly larger pixels than a Canon 5DmkII. I'm sure there are many other, better, examples but this is one I know of.

[LightningNZ]

Gotta agree with Dunk and Hugh - if it were down to sensor size most CCDs would be rubbish compared to dSLR sensors, but that clearly isn't so. The pixel size and structure is what matters.

[LightningNZ]

DXOmark.com (http://www.dxomark.com) strongly disagrees with the noise/sensitivity analysis presented above.

According to them it's very much middle-of-the-pack.

EDIT: removed direct links to DBs as they don't want to load.

[gregbradley]

Pixel size does come into it as well and can muddy up the comparisons for sure. But the basic here is the difference in surface size that is collecting light irregardless of pixel size. A full frame sensor is simply larger and therefore collects more light.

There is a Wikipedia article about this and modern sensors with more efficient circuitry have tended to minimise this difference in some instances but still there is usually a difference. Especially when pixel size is the same.

Full frame sensors often have larger pixels as well which also gives an advantage.

This article touches on it but the Wikipedia was very specific about sensor size giving a 2 stop advantage. If I can find it I will post it.

http://www.gizmag.com/camera-sensor-size-guide/26684/

I also agree his analysis may be flawed as I have not read any reports about the 7D2 having super low noise at high ISO rather what I have read is it is better than the 7D but not massively so in the noise department. Certainly it would be worth verifying from owners rather than racing out and buying one instead of a 6D and being disappointed later.

http://www.dpreview.com/forums/post/54846228

Greg.

[pluto]

I'm a bit confused.
Are you saying that, for example, a 5.4x5.4µm pixel on a FF sensor will receive more photons than the same sized pixel on an APS-C sized sensor?
Or is there some connection between physical sensor size and QE or read noise?

Of course a larger sensor receives more light than a smaller sensor but that light is divided amongst its pixels, or photosites.

EDIT: regarding that dpreview link, its hardly a surprise that the 7DmkII has better high I SO performance that the original - the original 7D had the worst noise of any Canon camera I've ever used!

[gregbradley]

Assuming both have the same sized pixels and same lens the full frame sees a wider field of view and thus more light. Not assuming any difference in QE, noise and certainly no extra light for each pixel of the same size between the 2 sensors. There may be a connection with less photon noise as the larger area can capture more. Not 100% sure of that though as generally it simply means a wider field of view when using a full frame versus an APS, hence more light.

There has been a lot of discussion of this aspect on DPReview in the past. Especially in the Fujifilm forum as its an APSc based system and it promotes the marketing claim that its system matches full frame performance.

In practice the basic gain of a full frame sensor may be diminished by higher QE, lower noise of well designed APSc sensors. Although in Fuji's case as they use Sony sensors with a different colour filter array the performance difference is questionable.

For nightscapes I see a big difference in sensitivity of my Fuji APSc low noise images and full frame either from a Nikon D800e (since sold) or Sony A7r. Also complicated by the fact that because Fuji has its own filter array normal ISO calculations seem to be twisted and ISO3200 on a Fuji X camera is not ISO3200 in brightness of a Canon, Nikon or Sony. Sony in particular seems very bright for the same ISO of other cameras I have used.

In practical terms you can see this at work in the nightscape section here. The full frame images are generally much more pleasing,brighter, wider field of view and lower noise than the APS images. So at the end of the day generally speaking full frame dominates in the Nightscapes department. Probably also with DSLR images with a telescope.

Greg.

[zenith]

Canon 7D Mark II pixel pitch 4.1 µm.
Canon 6D pixel pitch 6.5 µm, or 58% larger.
http://www.astropix.com/HTML/I_ASTROP/COMPARE.HTM

I am only a beginner to astro-photogoraphy, but it seems logical that for the same sized image projected onto the two sensors, the 6D will have lower resolution but 58% more photons captured per pixel. As for the noise, that depends on sensor quality, reading the signal from the sensor, and random noise introduced by the electronics. The 7D Mark II is newer and includes the Dual DIGIC 6 processor which I would be expected to have superior noise suppression. That said, 58% is a significant difference in raw signal strength per pixel.

It would be really nice to see a side-by-side comparison of several astro images...and then wait for the 6D Mk II :)

[pluto]

You're exactly right Tim.

Greg - I understand what you're saying, and I have plenty of experience shooting nightscapes and through a scope with both ff and aps cameras. You're comparing different sensors which have quite different technologies, like fuji xtrans vs standard bayer, and some are better than others. Evidently the performance isn't necessarily linked to the pixel size, I guess this is to be expected with such different technology used, but I really can't work out where the link between sensor size and extra light is coming from.

To be more clear let's consider a hypothetical test. It has to be hypothetical as I don't know of any cameras that have identical sensors but one in ff and one aps.
So we have two cameras sitting next to each other and each has an identical lens attached. One has a ff and one aps sensor, both sensors have the same size pixels and QE.
Let's say they're looking at a field with a tree in the centre. In both images the tree takes up exactly the same number of pixels but the ff camera sees more of the field either side of the tree.

If we crop the ff image to the same fov as the aps image then they should be identical.
So the extra light you say the ff sensor receives is really just extra fov?
The only inherent benefit I can see with the ff in terms of nightscape photography is that you can get a wider image for any given lens.

Now I know that in practice most of the ff cameras I've used have given better low light results than the aps cameras I've used but I'm pretty sure that was because of: a. The ff cameras were more expensive than the aps cameras and had more advanced sensors and processors etc. And b. They had larger pixels.

[Merlin66]

http://www.clarkvision.com/galleries/gallery.astrophoto-1/web/orion.nebula.m42_61,10,4,2sec_c11.2 1.2014.0J6A1631-1657-SigAv.h-b5x5s.html
:eyepop::eyepop::eyepop:

[Camelopardalis]

Wow. Love that it picked up all that red nebulosity...can anyone say Ha?

[gregbradley]

Hi Hugh,

Yes I think that was basically my point. Although independent of pixel size which as you point out is a big difference there is total amount of light the sensor receives simply by its size. Full frame is 2.25 times the surface area of APSc. Yes FF usually have larger pixels but some don't like Sony Nex 5 (16mp) and Sony A7r (36mp). The A7r noise levels are way lower.

It may be clearer this way - full frame image compared to tiny sensor in an Iphone image. The iphone has small pixels but also a tiny sensor. Noise levels are nowhere near the full frame image.

Micro 4/3rds cameras have excellent IQ until light levels drop and then the biggest complaint I read is noise even at low ISO levels. Yes small pixels but there is the factor of size of the sensor collecting the total amount of light.

There is a very clear Wikipedia article about this that explains it way better than I am. I'll try to find it again.

I agree the above gets lost because of the different sized pixels, different efficiencies of sensors and the general rule that large pixels have less noise and deeper wells than small pixels and thus lower noise.

Another way of expressing it would be APS sensors will have to work much harder and be more efficient to keep up with a full frame sensor.

By the way the Xtrans and Sony A7r are both Sony Exmor sensors so the differences there are not as great as you would think. So in a way they are a good example. A better example is this one:
http://www.dpreview.com/reviews/image-comparison?attr18=daylight&attr13_0=sony_nex3n&attr13_1=sony_a7r&attr13_2=sony_nex3n&attr13_3=sony_a7r&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=raw&attr16_0=6400&attr16_1=6400&attr16_2=12800&attr16_3=12800&normalization=print&widget=1&x=-0.892823179443937&y=0.18262535884370765

The Xtrans colour filter array does complicate things but basically its designed to allow no moire with no antialiasing filter and to reduce luminance noise both of which it does and hence the marketing claim that their APS sensor can match full frame (not entirely true but it does do a good job).

Greg.

[pluto]

I think we're basically saying the same thing :)

But when you say that a "full frame has something like a 2 stop advantage over APS sized sensors merely from size of the sensor alone" perhaps it would be more accurate to say that it has an advantage for any given FOV. For any given object, in the context of shooting through a scope, the advantage will always go to larger pixels (all other things being equal) and has nothing to do with sensor size.

[gregbradley]

Yes I think that is correct. Although with a telescope a small sensor has the same effect as increasing focal length. Use a small sensor on even a small aperture/short focal length scope and you get quite a magnified image.

For example the Sony 694 sensor on my CDK17 gives a highly magnified view compared to my 16803 camera on the same object. Much like a full frame camera that has a crop mode - you get an instant 1.6X zoom.

The interesting development to me in camera sensors is likely about to be released by Sony. An active pixel sensor that does not have a Bayer filter array, shoots in colour and somehow gets red, green and blue info of each pixel, has 24 stops of dynamic range at ISO5120 and low noise and capable of 16,000 frames per second. If all true it would be an unbelievable jump in sensor performance. I believe its not far away.

It potentially could be a huge advance for astrophotography. Low noise at high ISO being the unkown factor but likely to be good as the pixels can be very large due to not needing 4 pixels to make a colour pixel like in Bayer sensors.

Greg.

[AstroJunk]

I have to admit to having wandered by the local camera shop and played with the 7D mk II.

My impression with the Canon (and Nikon) cameras at the prosumer end of the market is that they are lumpy old dinosaurs. Is that being unkind? But it is nearly 2015...

I was drawn like a moth the flame of the Sony alpha series and discovered a camera that was beginning to feel and act like something I would expect in an internet connected world, and was especially interested to discover that you can buy apps for it including star trails and time lapse software.

My dilemma though would be the choice between the A7R and A7S which are so different, and I just feel that the monster pixels of the A7S are a big price to pay for the increased sensitivity, on the other hand...

[zenith]

I am not ready for Sony. My expensive Sony television died long before its time, and my mate's play station did not work out of the box. It took him hours on the phone, and 3 months in total to have it replaced (with him footing all shipping costs). I have a Sony DVD player and it has every feature know to mankind, and all of them are executed poorly, including the only ones I really need: stop, play, pause, top-menu.
They will need to lift their quality and customer service before I shop with them again.

[AstroJunk]

Indeed, I own some rubbish Sony products as far a execution is concerned, hence Jobbsie getting back into the tech market with a certain iPod!! Although I still use a quite superb 1990's CRT Tv for video use :)

[raymo]

Although at risk of hijacking the thread, I'd like to say that in my experience, Sony products, which used to be mid range, seem to have gone downhill in recent times. My daughter's large, heavy, and expensive TV has been lugged to the repair shop three times; each time to have the same component replaced, and my stereo packed up entirely at a young age.
raymo

[gregbradley]

I was drawn like a moth the flame of the Sony alpha series and discovered a camera that was beginning to feel and act like something I would expect in an internet connected world, and was especially interested to discover that you can buy apps for it including star trails and time lapse software.

My dilemma though would be the choice between the A7R and A7S which are so different, and I just feel that the monster pixels of the A7S are a big price to pay for the increased sensitivity, on the other hand...[/QUOTE]

Sony is about to release a 50mp full frame pro E mount camera in Jan/Feb. Its supposed to be a challenge to the Pro DSLRs. Time will tell if its up to the job. I have an A7r and love it for its form and I quite like the Zeiss lenses. D800e is a bit of a smoother camera but harder to get the shot in my opinion. A7r can be complex or act like a point and shoot which I like.

See the video of the Northern Lights Mike Salway linked. That is shot with an A7s. Also I have seen some comparisons between high ISO video from Canon 5D111 Canon C100, GH4 and A7s and the A7s beats them all easily although colour may not be as good as the Canons. A7s is the class leader for low light. For daylight still very good but 12mp isn't going to show 36mp resolution.


Greg.

[bwana]

There was a research paper published about two years ago (from a university in Italy) regarding sensor technology that hinted that microchip technology had advanced to the state where it was feasible to have each photosite on a bare sensor determine the frequency/energy of the light being captured and output a color value & luminosity directly without the necessity of bayer filters or a lot of the current nonsense associated with capturing a color image...

Sounds a bit far fetched but when you think about it, it is probably quite doable if one can reduce what is essentially a spectrograph to the size of a photosite on a sensor.

I for one would love to test this sensor!

bwa

[bwana]

The A7S and A7R are indeed different, in resolution as well as noise control. A repost I recently dropped to DPReview:

"Up to about ISO 1600 (maybe 3200) I think the image quality off the Sony A7R is better than the Sony A7S, being that image quality is a blend of both resolution and noise control. And yes, I realize this sort of shoots down my argument that nobody needs high MPixel cameras, at least for daytime shooting :(.

How I use the A7R and A7S is based very much on how they perform (noise-wise) under the available light. The A7R is pretty good up to about ISO 3200 (can be pushed to ISO 6400). The A7S excels at ISO 3200 up to about 51200. It is interesting that my quality cutoff on both cameras is 1/8th the maximum allowed ISO setting; by design??

I've done a bit of noise analysis on several Sony/Canon cameras using 30 sec Darks at about 20-22C and the Noise Evaluation utility in PixInsight.

My noise evaluation of the Sony A7 II shows it to be about the same as the A7R.

The Canon 6D has better noise control (about 33% better) than the Sony A7R/A7 all the way up to ISO 12800 (the highest I went). Its noise also increases slightly less with increased ISO than does the A7R/A7.

Testing the Canon 60D shows it's noise at ISO 3200 to be about the same as the 6D's at ISO 8000. I even tested my old Canon Rebel XTi. At ISO 1600 it is about a noisy as the 6D at ISO 6400.

I was also provided Darks off a Canon 7D II for testing but I think there was something wrong with the data provided because it tested twice as noisy as the Canon 6D which I find hard to believe? I would have expected it to be about equivalent to the 6D, maybe a touch worse, even considering it has an APS-C sensor and smaller photosites.

The Sony A7S had about 5x less noise than the A7R/A7 II all the way through its ISO range; about a two+ stop advantage." Of course you do gain some noise reduction by downsampling A7R images to A7S size but the A7S, in my opinion, still blows the wheels off the A7R/A7 II!

My testing also indicated the A7S had about 3x less noise than the Canon 6D in the ISO range tested: 3200-12800.

After reading a few reviews on the Canon 7D II, I'm no longer sure whether the Darks off the camera were good or faulty. I've seen reviews ranging from good to poor. Not sure where I stand on the latest edition of the 7D!? However, it really doesn't matter since I will never purchase another APS-C sensor camera after using full frame, and with the Sony A7 series cameras I can shoot in crop mode and still retain the noise advantage of a full frame sensor; win, win!!

Just my ramblings (from Canada)...

bwa

The large number of the images in the following albums were shot with my A7R and/or A7S (both cameras full spectrum modded):
Misty Valley Ranch Albums (https://plus.google.com/photos/116260312230579398213/albums)

[gregbradley]

Wow, imagine that!

Sony has patented a sensor where the colour filter array moves with each shot creating the RGB for each pixel.

Rumour is they are also working on a sensor similar to the Foveon sensor which differentiates between RGB by the depth in the silicone the signal reaches. RGB are read at different depths in the silicone. The problem is Foveon sensors whilst incredibly high res are very poor at higher ISOs.

Greg.

[PeterEde]

no flipout display. Not for me

[bwana]

I wonder what the speed of a moving filter array exposure might be? Sounds like there could be time for movement between the capture of the R, G & B channels, i.e.: motion smear?

As for Sigma's Foveon, if the problem of poor sensitivity at higher ISO's could be resolved, it might have potential. However, in a recent interview Shri Ramaswami w/ Foveon stated, "And of course, there are some disadvantages with our approach. Everyone knows our high ISO, for instance, isn't up to the standards of the competition." And he didn't suggest Sigma/Foveon had any remedy for the problem :confused2:. In fact it may be physics controlling the process which means it is as good as it gets!?

Interesting article at: http://www.mirrorlessrumors.com/new-tfd-foveon-alike-sensor-has-36-layers but a long, long way from market!

Gotta love the technological advances (and sometimes the retreats) in the field of photography...

bwa

[bwana]

Yes, that LiveView screen on the 60D is wonderful!

bwa

[ericwbenson]

There already is a motion "blur"! Check out the attached image I took in 2011 with an EOS 350D + Canon 70-200 f/4 IS (what a *great* lens) and compare the shadow to the real scene (which are separated by about 2 m = 20 nsec).
I think I know what the mechanism behind this is, but I'm not 100% sure ;)

I had not noticed for 3 years until coincidentally the kids where looking thru Picassa today and my wife noticed this very interesting difference. I've been to the Aussie Open twice since and have never managed to replicate the squashed tennis ball, beginner's luck?...maybe next year.

EB

[AstroJunk]

That's a great shot! It really underscores the scanning nature of electronic chips:thumbsup:

[SteveInNZ]

No, it shows the scanning nature of the focal plane shutter.

Steve.

[gregbradley]

That's right.

The shutters have a horizontal slit which varies in width depending on the speed of the shutter.

It goes from top of the sensor down to the bottom. Keep in mind the image is upside down from the lens. So the top of the image here is the bottom of the final picture. That got exposed first and as the shutter moves down the sensor it exposes as it goes. So that means the guy hitting the ball got exposed later than the shadow and he was moving fast so it was a tad later in time. Its called rolling shutter effect.

A development expected of DSLRs in the future is a global shutter where the whole sensor is read out at the same time.

Greg.