Approve the Cookies
This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and our Privacy Policy.
OK
Index  •   • New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Guest
New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Register to forums    Log in

 
FORUMS Photography Talk by Genre Astronomy & Celestial Talk 
Thread started 12 Feb 2010 (Friday) 13:19
Search threadPrev/next
sponsored links
(this ad will go away when you log in as a registered member)

EOS 550D, might it be a good astro camera?

 
MintMark
Senior Member
Avatar
385 posts
Joined Apr 2009
Location: Hampshire, England
     
Feb 12, 2010 13:19 |  #1

I was pleasantly surprised to see the announcement of a new EOS rebel. My 1000D is great, but I often wonder if higher ISO and sensor resolutions would help. The 7D is the top of the range crop body with all the features, but this new 550D seems to bring some of them down into a small rebel sized body.

In particular, 18Mpixels and ISO up to 6400 seems good. It also has an uncrippled movie mode where you can control the exposure... so that might be good for lunar and planetary? Heavier than my 1000D, but still less than a 50D or 7D.

Obviously we'll have to wait for proper reviews... but I think it looks interesting.


Mark

  
  LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as a registered member)
DonR
Senior Member
262 posts
Likes: 4
Joined Dec 2009
Location: Georgia, USA
     
Feb 13, 2010 09:50 |  #2

I'll take a crack at this, Mark, even though I don't have a 550D and don't know anyone who has used one yet.

I suspect the 550D will make an excellent astro camera, merely because it's a modern Canon DSLR. Since the 20D, Canon has done nothing but make successive generations better than the previous for astrophotography. The most significant improvements have been in the area of noise, and I suspect that will be the case again with the 550D. There is nothing about 6400 ISO that makes it better for astrophotography, but the fact that it is offered on a camera implies that noise is well controlled in the sensor.

Similarly, more pixels does not in itself make a camera better for astrophotography. More pixels in the same space means smaller pixels, and smaller pixels means fewer photons strike each pixel during a given exposure. Fewer photons means that to achieve the same sensitivity, each pixel must be more sensitive and less susceptible to noise. These two factors, sensor noise and sensitivity, are the areas where Canon has continued to make advancements over the past years.

If you will be mounting the camera on a telescope, the smaller, lighter body of the 550D compared to the 7D will be an advantage.The most limiting factor in resolution for earth-bound astrophotography, though, is aperture, not pixel size, as the pixel size of modern astro cameras approaches and even surpasses the resolution limits imposed by atmospheric seeing.

In astrophotography, especially photography of faint deep space objects, the goal is to collect as many photons as possible in order to maximize the signal to noise ratio of the image. Higher ISO doesn't help here, as the only factors influencing the number of photons collected are the brightness of the subject, the exposure time and the aperture of the lens (or telescope). For most of us, collecting as many photons as possible usually means dialing down the ISO so that we can expose longer without overwhelming the sensor with sky glow (light pollution). I, for one, have never used the ISO 1600 setting on my 350D, so I'm sure I would have no use whatsoever for ISO 6400. Even with lower noise in the 550D, the overwhelming effect of sky glow would mean I would have to limit the exposure times too severely. Most of the noise in deep space images is "shot noise", which is a result of the very low brightness of the subjects, not a factor of the sensor or camera electronics, and shot noise is reduced only by longer exposures and by stacking multiple images.

Don




  
  LOG IN TO REPLY
hollis_f
Cream of the Crop
Avatar
10,649 posts
Gallery: 1 photo
Likes: 85
Joined Jul 2007
Location: Sussex, UK
     
Feb 13, 2010 11:44 |  #3

DonR wrote in post #9601846 (external link)
In astrophotography, especially photography of faint deep space objects, the goal is to collect as many photons as possible in order to maximize the signal to noise ratio of the image. Higher ISO doesn't help here, as the only factors influencing the number of photons collected are the brightness of the subject, the exposure time and the aperture of the lens (or telescope). For most of us, collecting as many photons as possible usually means dialing down the ISO so that we can expose longer without overwhelming the sensor with sky glow (light pollution). I, for one, have never used the ISO 1600 setting on my 350D, so I'm sure I would have no use whatsoever for ISO 6400. Even with lower noise in the 550D, the overwhelming effect of sky glow would mean I would have to limit the exposure times too severely. Most of the noise in deep space images is "shot noise", which is a result of the very low brightness of the subjects, not a factor of the sensor or camera electronics, and shot noise is reduced only by longer exposures and by stacking multiple images.

Hmmm, I'm not sure I follow this argument. Let me explain -

I can see how sky glow will affect the length of exposure, with high ISO meaning one needs to use shorter exposures than with low ISO. But those shorter exposures at high ISO means that there are more shots taken over a given time. And stacking more shots means knocking the signal:noise ratio down.

So, all other things being equal, you can take 10 shots of 60 seconds each in a 10 minute period at ISO 100, or you can take 40 shots of 15 seconds each at ISO 400.


Frank Hollis - Retired mass spectroscopist
Give a man a fish and he'll eat for a day. Teach a man to fish and he'll complain about the withdrawal of his free fish entitlement.
Gear Website (external link)

  
  LOG IN TO REPLY
DonR
Senior Member
262 posts
Likes: 4
Joined Dec 2009
Location: Georgia, USA
     
Feb 13, 2010 12:30 |  #4

hollis_f wrote in post #9602398 (external link)
Hmmm, I'm not sure I follow this argument. Let me explain -

I can see how sky glow will affect the length of exposure, with high ISO meaning one needs to use shorter exposures than with low ISO. But those shorter exposures at high ISO means that there are more shots taken over a given time. And stacking more shots means knocking the signal:noise ratio down.

So, all other things being equal, you can take 10 shots of 60 seconds each in a 10 minute period at ISO 100, or you can take 40 shots of 15 seconds each at ISO 400.

Your argument is valid for relatively bright subjects, Frank, but it falls apart if you're imaging a faint subject. This becomes apparent if you consider an extreme example.

Lets say you are trying to image the Horse Head Nebula, a notoriously faint subject. Lets assume that your sky is dark enough to expose for five minutes at ISO 800 - your images will all capture enough photons from the nebula to define the subject, so the nebula will be recognizable, though noisy, in each five minute sub-exposure. Now, if you try a 1 second exposure at ISO 800, the probability is that you won't capture any photons from the nebula at all, and you certainly will not define the subject. Repeating those one second exposures 300 times and stacking them will help very little, because the signal to noise ratio in many of the one second exposures will be zero, meaning those exposures don't count towards the total integration time. And raising the ISO won't help either, because increasing the ISO doesn't capture more photons.

Now consider that even in relatively bright subjects (like M42 for example), there are faint details that are no brighter than the Horse Head Nebula. While you may capture enough photons in one second exposures of M42 at ISO 800 to produce a recognizable image, you certainly won't be able to resolve the faint details that would be apparent in longer exposures, and stacking multiple short exposures won't help with those faint details.

If this seems counter-intuitive to you, just try it - I have tried it. The bottom line is that stacking as many exposures as you can acquire is important, but the length of the exposures is important too, and if your goal is to capture faint details, there is no substitute for long exposure times.

Don




  
  LOG IN TO REPLY
swag72
Senior Member
465 posts
Likes: 1
Joined Nov 2008
Location: Olocau, Spain
     
Feb 13, 2010 14:24 |  #5

Don,

I am interested to know how you work out how many exposures you need and at what length.

For example - If youa re taking a pic of the Orion nebula, how would you go about deciding the above - What process would you go through to arrive at the best possible number and lengths of exposure.


Catch my pictures on Flickr http://www.flickr.com/​photos/swag72/ (external link)

  
  LOG IN TO REPLY
DonR
Senior Member
262 posts
Likes: 4
Joined Dec 2009
Location: Georgia, USA
     
Feb 13, 2010 15:03 |  #6

swag72 wrote in post #9603150 (external link)
Don,

I am interested to know how you work out how many exposures you need and at what length.

For example - If youa re taking a pic of the Orion nebula, how would you go about deciding the above - What process would you go through to arrive at the best possible number and lengths of exposure.

The first factor you have to consider is how long you can track the subject without drift, tracking error or periodic error of the mount causing streaking, blurring or trailing. Preferably the answer to this question is "indefinitely". If this is not the case you should work on that first. A decent quality equatorial mount, adequately polar aligned, is the starting point. Then, depending on your focal length and image sensor size, it may be necessary to enhance the tracking accuracy by guiding, either manually or automatically.

Once you have the tracking/guiding worked out so that it is no longer a limiting factor, You need to take into consideration the brightness of the subject, i.e., how long does it take to capture enough photons from the subject to produce a recognizable image? This will not depend on the ISO setting, but only on your aperture and exposure time, along with the brightness of the subject. If you capture enough photons to define the subject, you can enhance the brightness and contrast of a low ISO shot after the fact in software. The answer to this question may be subjective, however, because many deep sky objects have a wide range of brightness levels, so you need to decide whether you are aiming to capture the brightest areas at the expense of the faintest, or vice versa. Usually you will end up compromising due to the other limiting factors.

As Frank stated, you will always want to take multiple exposures and stack them when using a DSLR, in order to reduce the noise in the final image. There is no upper limit to how many exposures you should take, the more the better. But there is a lower limit, dependent on the subject brightness mainly. I have heard it stated that 14 to 16 sub-exposures are a good minimal number. That's pretty arbitrary, but generally pretty good advice in my experience. I prefer to take 20 or more, and if I can take 100 sub-exposures I will. But if I can't get at least 15 or so, I don't consider the work done - returning to the subject on another night is in order.

Obviously the number of exposures you can take in a night depends on the length of the exposures (among other things), but my philosophy is if I can't capture enough photons in each sub-exposure to at least minimally define the subject I'm wasting my time. Hopefully there will be another night when I can add more sub-exposures.

So I will try to answer your question after all this beating around the bush. For subjects with faint detail, including most nebulae and galaxies, and assuming I am trying to capture the faint detail, I don't like to go less than 3 minutes, and prefer 4 minutes or more. I will then adjust the ISO in order to keep the sky glow from overwhelming the image in the chosen exposure time. I use the histogram (on-camera or in DPP if I have the laptop PC connected) to gauge the sky glow, shooting for the peak of the hump in the histogram, which represents the sky glow, to be somewhere near the middle of the x-axis. Then I will take as many exposures as I can, and if, after stacking and evaluating the results (usually the next day), I feel there's still too much noise, I will try to add more sub-exposures on a subsequent night.

For subjects where most of the detail is not faint, like star clusters, shorter exposures (and more of them) work well. Globular clusters are among my favorite subjects, and two minute exposures are usually plenty for them. You can even make impressive images of many of the globular clusters with 30 second exposures, making them excellent targets for starting out. The brighter globs like M2, M3, M5 and M15 (and the king of all globulars, Omega Centauri if you're far enough south to see it), can yield very satisfying images to the beginner and still offer challenges to experienced astrophotographers.

Don




  
  LOG IN TO REPLY
MintMark
THREAD ­ STARTER
Senior Member
Avatar
385 posts
Joined Apr 2009
Location: Hampshire, England
     
Feb 14, 2010 04:10 as a reply to  @ DonR's post |  #7

I think what Don says is absolutely right. In my original comments about the new camera I should have made it clear that acceptable noise at ISO 6400 is what sounds good. With my 1000D I usually use ISO 800. As for the 18 Mpixels, they are emphasising the gapless microlenses on the sensor which hopefully means that every photon that lands gets recorded somewhere. My hope was that more pixels would lead to more detail in planets, where only a small area of the sensor is used but they are relatively bright (depending on magnification).

As I understand it, the different ISO settings are just multiplying the voltage that results from the photons collected at each site on the sensor. The ISO selection is really just choosing which part of the dynamic range you want to see in the image. As the values are read from the sensor there is an element of noise (read out noise) and the exposure should be long enough so that the features in the image have higher values than the noise. That's why the histogram peak should be away from the left hand side of the graph.

Some things I have read say that this is all that is necessary. Any longer exposures are just risking star trailing, aircraft/satellite trails and vibration accidents. I have even read that it is not necessary for the faintest details to appear in every frame, but you have to combine every frame (averaging rather than kappa-sigma) to detect anything.

I haven't tried longer exposures at lower ISOs, mainly because my setup isn't up to it (hopefully I've found a problem that will help improve this!). Isn't there a possibility that too low an ISO setting will throw away faint detail (by rounding down the values)? I've read threads here that try and work out the "natural" ISO value for each sensor where the values remain unaltered and the dynamic range is the widest. Then the exposure is adjusted to fit with that. That would suggest a fixed ISO, variable exposure time method.

Back to the cameras... I'm also looking at the 50D. It also has the gapless microlenses on its sensor and 15Mpixels for a similar price. It is an older model, but offers plenty of improvements over my 1000D as a daytime camera. I'll wait and see how the low light performance compares for them both.


Mark

  
  LOG IN TO REPLY
DonR
Senior Member
262 posts
Likes: 4
Joined Dec 2009
Location: Georgia, USA
     
Feb 14, 2010 10:25 |  #8

MintMark wrote in post #9606267 (external link)
As I understand it, the different ISO settings are just multiplying the voltage that results from the photons collected at each site on the sensor. The ISO selection is really just choosing which part of the dynamic range you want to see in the image. As the values are read from the sensor there is an element of noise (read out noise) and the exposure should be long enough so that the features in the image have higher values than the noise. That's why the histogram peak should be away from the left hand side of the graph.

Close, but I think the term "amplifying" is more appropriate than "multiplying". Multiplying implies a precise mathematical conversion, and that isn't the case here. The charge that accumulates in each well of the sensor is converted to a voltage at the end of the exposure, and the voltage at each site is amplified before being fed to the A/D converter. There is some variability in the amplification process, which is one reason that higher ISO means noisier images. But this variability, and the other noise components introduced by the sensor and electronics, including read-out noise, are quite small compared to "shot noise", which is caused by the nature of light itself. Photons arrive at the sensor in discrete packets rather than in a continuous stream. In order to produce an image that accurately reflects the relative brightness of different parts of the image, you need to collect enough photons at each site on the sensor so that the number is representative of the relative illumination of each site. Shot noise is due to the natural variability in the rate of photon arrival at a given spot, and in photography it is inversely proportional to the square root of the level of illumination.

Imagine that your job is to count the number of cars that pass a point on the freeway in an hour. Imagine that the traffic is moderately heavy and steady, but not bumper to bumper and not perfectly uniform. You could count the number of cars that pass you in 10 seconds and multiply by 360, or you could count the number of cars that pass you in 10 minutes and multiply by 6. Obviously the latter case will give you a better approximation. The best value, though, would be obtained by counting the cars that pass you in an hour.

The effect of raising or lowering the ISO is similar to this. Experienced terrestrial photographers know that shooting at lower ISO yields cleaner, less noisy images, and this is due to the fact that reducing the ISO results in a longer exposure time, higher illumination level, and lower shot noise. Doubling the exposure time reduces shot noise by a factor of the square root of two.

MintMark wrote in post #9606267 (external link)
Some things I have read say that this is all that is necessary. Any longer exposures are just risking star trailing, aircraft/satellite trails and vibration accidents.

Definitely, if trailing and vibration are issues, you will achieve better results by keeping the exposures shorter. But your objective should be to eliminate these issues. I have been doing this for a while, and I have worked out the issues to the point that shots ruined by trailing, vibration or slipping of the mount are quite rare. This has been by far the biggest challenge I have faced, but it is achievable.

As far as satellites and airplanes, you are at their mercy, but with enough subs they disappear from the stacked results. This is where kappa-sigma stacking really helps.

MintMark wrote in post #9606267 (external link)
I have even read that it is not necessary for the faintest details to appear in every frame, but you have to combine every frame (averaging rather than kappa-sigma) to detect anything.

It's definitely true that the faintest details won't stand out in every frame, but for the best results they should be represented in at least the large majority of the frames. It may be necessary to stretch the image contrast to see them though.

My home site suffers from pretty severe light pollution, so faint objects like the Horse Head are especially challenging. I have found that in order to achieve a decent composite image it is necessary to push the exposure times until the sky glow "seems" to overwhelm the subject in individual subs. Attached is an example: an 8 minute sub at ISO 400, where the sky glow hump was well to the right of the middle of the histogram. The nebula is barely discernible, but along with it is a contrast enhanced view of the same image, showing that the nebula is definitely there in the data. Keep in mind that this shot was made with an LP filter, which gives an indication of how bright the sky glow was on this night.

MintMark wrote in post #9606267 (external link)
I haven't tried longer exposures at lower ISOs, mainly because my setup isn't up to it (hopefully I've found a problem that will help improve this!). Isn't there a possibility that too low an ISO setting will throw away faint detail (by rounding down the values)? I've read threads here that try and work out the "natural" ISO value for each sensor where the values remain unaltered and the dynamic range is the widest. Then the exposure is adjusted to fit with that. That would suggest a fixed ISO, variable exposure time method.

Canon doesn't publish this information, but my understanding is that the "natural" ISO value for their cameras is always the lowest ISO value available, i.e., ISO 100 on the Canon DSLR's I am familiar with. If this is correct, then the amplifier isn't used or is used at minimum gain at ISO 100, but every photon counts towards the image. Unfortunately, shooting at ISO 100 for most deep space objects would result in excessively long exposure times - not from the standpoint of image quality, but purely from a practical standpoint. I guess what I'm saying is I just don't have that much patience, which is why I got rid of my film SLR's. But it would be an interesting experiment to acquire some long exposures of DSO's at ISO 100 for comparison.

Don


HOSTED PHOTO
please log in to view hosted photos in full size.




  
  LOG IN TO REPLY
flyboy89
Member
Avatar
100 posts
Joined May 2008
Location: Leicester, UK
     
Mar 05, 2010 20:33 |  #9

If we are stacking images in DSS for example, do you think that the Multiple Exposure feature on film SLRs would be similar to this? I think the main problem I would have would be the stars trailing :/




  
  LOG IN TO REPLY
fogboundturtle
Senior Member
735 posts
Likes: 36
Joined Mar 2010
     
Mar 10, 2010 10:13 |  #10

I have an 550D and its as capable of any other DSLR camera for AP. You will always needs to do flats and darks for your stacking. This is the nature of all CMOS Sensor, They heat up, create noise. That noise needs to be remove just like Don is showing in the previous post.

If you have a little budget, you can remove the IR cut filter on your camera and replace with a Baader Ir Cut filter that will allow more infrared band to pass. I suggest you do business with http://www.hapg.org/ca​mera%20mods.htm (external link) if you decide to go that route. he is well known among the community.


Canon 5D Mark III, Canon 70D, Canon EF 24-105L, Tamron 150-600mm, Tamron 70-200 F2.8 DI VC USD, Sony A7r, Sony FE 55mm F1.8

  
  LOG IN TO REPLY
DonR
Senior Member
262 posts
Likes: 4
Joined Dec 2009
Location: Georgia, USA
     
Mar 10, 2010 13:46 |  #11

Hi flyboy89,

I just noticed your post, better late than never.

If you're talking about multiple exposures on the same frame using film, that would probably not be a good idea. My experience with film astrophotography is pretty limited, but I have done a little of it. The biggest difference between film and digital astrophotography is that film loses sensitivity to light pretty quickly as it is exposed. Film is engineered for normal exposures, usually lasting less than 1 second. The sensitivity drop becomes noticeable within a few seconds, and is proportional to the total illumination, not to the exposure time. So the brighter areas of an image saturate more slowly than the darker areas, resulting in the darkest areas (the sky glow in all cases) eventually catching up with the brightest areas. How long it takes for this to happen depends on the magnitude of the sky glow, and is called the "sky fog limit". This film phenomenon is called reciprocity failure, and does not apply to digital sensors, so the term "sky flog limit" also does not apply to digital cameras.

Because of reciprocity failure, film astrophotography is most successful in locations with very dark skies, and exposure times for DSO astrophotography with film are typically much longer than with digital cameras. And because film isn't subject to most of the noise sources in digital cameras, the primary source of noise in film astrophotography is "shot noise", which is the noise caused by the fact that photons don't arrive at a constant rate. Longer exposures decrease the ratio of shot noise to signal, so in film astrophotography it is not uncommon to produce a final image from a single exposure. The source of noise that can still be decreased substantially by stacking multiple film images is the noise contributed by the film grain, and careful choice of film can make this less of an issue.

So, there would be no advantage to shooting multiple short exposures with film. You would be better off with one or a few long exposures from a dark site. Twenty and thirty minute exposures are not uncommon with film, and they can be much longer, upwards of one hour.

Don




  
  LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as a registered member)

12,941 views & 0 likes for this thread
EOS 550D, might it be a good astro camera?
FORUMS Photography Talk by Genre Astronomy & Celestial Talk 
AAA
x 1600
y 1600

Jump to forum...   •  Rules   •  Index   •  New posts   •  RTAT   •  'Best of'   •  Gallery   •  Gear   •  Reviews   •  Member list   •  Polls   •  Image rules   •  Search   •  Password reset

Not a member yet?
Register to forums
Registered members may log in to forums and access all the features: full search, image upload, follow forums, own gear list and ratings, likes, more forums, private messaging, thread follow, notifications, own gallery, all settings, view hosted photos, own reviews, see more and do more... and all is free. Don't be a stranger - register now and start posting!


COOKIES DISCLAIMER: This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and to our privacy policy.
Privacy policy and cookie usage info.


POWERED BY AMASS forum software 2.1forum software
version 2.1 /
code and design
by Pekka Saarinen ©
for photography-on-the.net

Latest registered member is jimlee
942 guests, 312 members online
Simultaneous users record so far is 15144, that happened on Nov 22, 2018

Photography-on-the.net Digital Photography Forums is the website for photographers and all who love great photos, camera and post processing techniques, gear talk, discussion and sharing. Professionals, hobbyists, newbies and those who don't even own a camera -- all are welcome regardless of skill, favourite brand, gear, gender or age. Registering and usage is free.