Pixel Density and Reach--Any Definitive Tests?

I like to shoot wildlife, but can't afford 600mm or 800mm lenses. As such, I find that I'm focal length limited much of the time. To date, I've subscribed to the view that in such situations, a sensor with a higher pixel density will put "more pixels on the duck," and give me a reach advantage over a lower density sensor. So, I bought a crop sensor 7D in preference to a lower pixel density full frame camera believing that I'd get better/higher resolution pics, at least in focal length limited situations, from the 7D. And I've been very happy with my choice.

But there's another view, which goes something like this: Not all pixels are created equal, any any reach advantage the 7D might have from its higher pixel density sensor is more than made up for by the 5DII or 1DIV's less noisy pixels. In other words, although the full frames may put fewer pixels on the duck, they're better pixels.

I've seen all kinds of theoretical discussions of this issue, with folks on both sides making good points. But what I haven't seen are any actual test results that I consider definitive. Anyone out there who has either done some testing themselves or can point to testing done by others that sheds some light on this issue?

Thanks for any responses, and please let's keep the discussion civil. I know this can be a hot topic with some people.

Optical resolution from the lens degrades as you crop more deeply because of the increased magnification required for a given final image size. In my opinion, this tends to make the choice of lens more important than the camera. Magnification is exactly the same regardless of whether the camera is FF or APS-C, as long as we’re talking about crops and the same final image size.

Sure, the camera/sensor will make some difference, but probably not as great as the lens. I’ve seen numerous comparisons of images from the 7D and 5D2 where the differences were difficult to detect. On the other hand, it’s often easy to spot differences between various lenses of the same focal length when viewing extreme enlargements, even when different cameas are used. Today's cameras and sensors are that good.

Phil Hill said:

Optical resolution from the lens degrades as you crop more deeply because of the increased magnification required for a given final image size. In my opinion, this tends to make the choice of lens more important than the camera. Magnification is exactly the same regardless of whether the camera is FF or APS-C, as long as we’re talking about crops and the same final image size.

Sure, the camera/sensor will make some difference, but probably not as great as the lens. I’ve seen numerous comparisons of images from the 7D and 5D2 where the differences were difficult to detect. On the other hand, it’s often easy to spot differences between various lenses of the same focal length when viewing extreme enlargements, even when different cameas are used. Today's cameras and sensors are that good.

Click to expand...

I agree completely with what you've said. But the question I'm asking assumes that you're using the best and longest focal length lens you can afford, and you're still focal length limited--your subject doesn't fill the frame of your 5D II. The question then is, will you get better results shooting the same subject, from the same position, with the same lens and at the same focal length, with a 7D with it's higher density sensor than you would shooting with the 5D II and cropping the image to the same angle of view as you'd get with the 7D.

There seems to be a lot of disagreement on this, and what I'm looking for is some sort of definitive testing that would help to settle the question.

The reason I'm asking is that with a 5D III on the horizon, if it turns out that the 7D's high pixel density advantage turns out to be a myth as some claim, then I may very well get one of those new 5D IIIs.

If you are Pixel peeping, then there might be better pixels, but on the great scheme of things, it might not render a worse image than the 5D, if you process your images correctly, with the right amount of sharpening and noise reduction to the size you will be printing or displaying your images.

Pixel peeping is the fastest road to disappointment on any gear, cameras or lenses. You will never view the image you are displaying in your computer at that magnification. You must ask yourself first, what is my primary display target? Is it print or web? if print, which size? if is web, then definitively anything smaller than a 800x600px image render is more than enough, then no worries. If printing, and depending on the size, you can render awesome good images with a 6-8MP sensor or crop. Again, depending on the size and the intended viewing distance.

Don't get too discouraged with specs. It might not be that important when you take into account the final result.

People around here are making impressive images with their 300D, 20D, XSs, or whatever old camera you might imagine. There is no comparison on "better pixels", but in the final product, those pixels might not be even noticeable.

bhollis said:

I like to shoot wildlife, but can't afford 600mm or 800mm lenses. As such, I find that I'm focal length limited much of the time. To date, I've subscribed to the view that in such situations, a sensor with a higher pixel density will put "more pixels on the duck," and give me a reach advantage over a lower density sensor. So, I bought a crop sensor 7D in preference to a lower pixel density full frame camera believing that I'd get better/higher resolution pics, at least in focal length limited situations, from the 7D. And I've been very happy with my choice.

But there's another view, which goes something like this: Not all pixels are created equal, any any reach advantage the 7D might have from its higher pixel density sensor is more than made up for by the 5DII or 1DIV's less noisy pixels. In other words, although the full frames may put fewer pixels on the duck, they're better pixels.

I've seen all kinds of theoretical discussions of this issue, with folks on both sides making good points. But what I haven't seen are any actual test results that I consider definitive. Anyone out there who has either done some testing themselves or can point to testing done by others that sheds some light on this issue?

Thanks for any responses, and please let's keep the discussion civil. I know this can be a hot topic with some people.

Click to expand...

--
Martin Ocando
-------------------------

Martin Ocando said:

If you are Pixel peeping, then there might be better pixels, but on the great scheme of things, it might not render a worse image than the 5D, if you process your images correctly, with the right amount of sharpening and noise reduction to the size you will be printing or displaying your images.

Pixel peeping is the fastest road to disappointment on any gear, cameras or lenses. You will never view the image you are displaying in your computer at that magnification. You must ask yourself first, what is my primary display target? Is it print or web? if print, which size? if is web, then definitively anything smaller than a 800x600px image render is more than enough, then no worries. If printing, and depending on the size, you can render awesome good images with a 6-8MP sensor or crop. Again, depending on the size and the intended viewing distance.

Don't get too discouraged with specs. It might not be that important when you take into account the final result.

People around here are making impressive images with their 300D, 20D, XSs, or whatever old camera you might imagine. There is no comparison on "better pixels", but in the final product, those pixels might not be even noticeable.

Click to expand...

Martin,

Thanks for your response. You seem to be suggesting that any differences are negligible--just a matter of pixel peeing--and I shouldn't worry about it. If you're correct--that it doesn't much matter whether you use the 5DII or 7D in focal length limited situations--then you're effectively arguing that the 7D's supposed reach advantage is a myth, or at least is so minor as to be inconsequential.

Is that what you're saying? if so, have you done or seen any testing to back up that view?

Liquidstone, If you're out there reading this, I suspect you have some practical experience with this issue, and I'd be very interested in your views.

Not exactly. There are inherit specific advantages in pixel density on FF cameras over APS-C, and there are even more advantages on lack of leaking from one photosite to the next, signal to noise ratio, dynamic range, etc. But it the long run, it all depends on the intended target of your images.

If you never intend to print your images at more than 8x10 letter sized prints, there will be a big difference and advantage if you use a FF 5DMkII, but you can get almost the same results with a good capture of a 7D, correctly processed for NR and sharpening for 8x10.

I mean, my target right now is to get a FF camera in the near future, but not because I plan to print big, but because I rather have a FF RAW of my images, that later I can decide on printing them big, than an APS-C image that might hurt my image quality in big prints. But that is because I'm a passive photographer. My main target is landscape and some wildlife, but not Kenya like safari wildlife, but some birds, iguanas, squirrels, macros, etc. Nothing that I will be needing a machine gun kind of a camera.

I think the definitive choice between a 7D and a 5D MkII is on the particular needs regarding high speed shooting and AF capabilities for fast action, rather than actual RAW image quality and pixel density. There is no point in having a 5D MkII to shoot fast moving wildlife or sports, if you will not be able to focus properly on your subject, or get enough frames to capture all the different postures of your animals or players.

bhollis said:

Martin Ocando said:

If you are Pixel peeping, then there might be better pixels, but on the great scheme of things, it might not render a worse image than the 5D, if you process your images correctly, with the right amount of sharpening and noise reduction to the size you will be printing or displaying your images.

Pixel peeping is the fastest road to disappointment on any gear, cameras or lenses. You will never view the image you are displaying in your computer at that magnification. You must ask yourself first, what is my primary display target? Is it print or web? if print, which size? if is web, then definitively anything smaller than a 800x600px image render is more than enough, then no worries. If printing, and depending on the size, you can render awesome good images with a 6-8MP sensor or crop. Again, depending on the size and the intended viewing distance.

Don't get too discouraged with specs. It might not be that important when you take into account the final result.

People around here are making impressive images with their 300D, 20D, XSs, or whatever old camera you might imagine. There is no comparison on "better pixels", but in the final product, those pixels might not be even noticeable.

Click to expand...

Martin,

Thanks for your response. You seem to be suggesting that any differences are negligible--just a matter of pixel peeing--and I shouldn't worry about it. If you're correct--that it doesn't much matter whether you use the 5DII or 7D in focal length limited situations--then you're effectively arguing that the 7D's supposed reach advantage is a myth, or at least is so minor as to be inconsequential.

Is that what you're saying? if so, have you done or seen any testing to back up that view?

Click to expand...

--
Martin Ocando
-------------------------

Martin Ocando said:

I think the definitive choice between a 7D and a 5D MkII is on the particular needs regarding high speed shooting and AF capabilities for fast action, rather than actual RAW image quality and pixel density. There is no point in having a 5D MkII to shoot fast moving wildlife or sports, if you will not be able to focus properly on your subject, or get enough frames to capture all the different postures of your animals or players.

Click to expand...

I think this is a very good point. When I bought my 7D, I did so primarily to shoot wildlife--a purpose for which I saw the 7D particularly well suited, not only for the reasons you've mentioned (better/faster AF and burst speed), but also for the 7D's supposedly greater reach (due to it's higher pixel density sensor).

At the same time, I recognize that a full frame sensor will be better at high ISOs, and will also allow me to shoot with shallower dof--both of which I consider significant advantages. And I suspect that a new 5DIII will have AF and burst capabilities much closer to the 7D's. So if a 5DIII really does give me AF and burst capabilities on a par with the 7D, it would be very tempting. The only thing that would likely stop me is if the 7D would still give me significantly greater reach in focal length limited situations.

And that's why I'm interesting in seeing some tests.

Not true; the IQ difference between the 7D and 5D mk II are much smaller than you make them out to be. When I bought my 7D, I had trouble deciding between the 7D and 5D mk II. Because the IQ is so similar, I didn't even consider that to be a factor. I weighed the wider angle and DOF advantage of the larger sensor in the 5D vs. the advanced AF and ~100% viewfinder and the 8-8.5fps of the 7D, and in the end, I chose the 7D. I'm still going to get a 5D though; I have been waiting for the new model to come out to see if the AF improves. If it does, I will buy one, but I am not going to get rid of or stop using the 7D.
--
Caution: Do not stare into laser with remaining eye.

Well, is not exactly reach. You are simply cropping your images with an APS-C sensor. I mean, magnification remains the same in comparison to a FF camera, but indeed it gives the "appearance" of increased reach.

I don't have any idea about what Canon might be doing to the 5D MkIII, but trying to get it closer to the 7D seems a bit out of their intended market target. I mean, I really hope they improve the AF and FPS, but I won't be too sad if they just keep it the same. The 5D MkII is an outstanding camera as it is, anyways. What I do hope is that he MkIII bring prices lower to the MkII

bhollis said:

Martin Ocando said:

I think the definitive choice between a 7D and a 5D MkII is on the particular needs regarding high speed shooting and AF capabilities for fast action, rather than actual RAW image quality and pixel density. There is no point in having a 5D MkII to shoot fast moving wildlife or sports, if you will not be able to focus properly on your subject, or get enough frames to capture all the different postures of your animals or players.

Click to expand...

I think this is a very good point. When I bought my 7D, I did so primarily to shoot wildlife--a purpose for which I saw the 7D particularly well suited, not only for the reasons you've mentioned (better/faster AF and burst speed), but also for the 7D's supposedly greater reach (due to it's higher pixel density sensor).

At the same time, I recognize that a full frame sensor will be better at high ISOs, and will also allow me to shoot with shallower dof--both of which I consider significant advantages. And I suspect that a new 5DIII will have AF and burst capabilities much closer to the 7D's. So if a 5DIII really does give me AF and burst capabilities on a par with the 7D, it would be very tempting. The only thing that would likely stop me is if the 7D would still give me significantly greater reach in focal length limited situations.

And that's why I'm interesting in seeing some tests.

Click to expand...

--
Martin Ocando
-------------------------

kimvette said:

Not true; the IQ difference between the 7D and 5D mk II are much smaller than you make them out to be. When I bought my 7D, I had trouble deciding between the 7D and 5D mk II. Because the IQ is so similar, I didn't even consider that to be a factor. I weighed the wider angle and DOF advantage of the larger sensor in the 5D vs. the advanced AF and ~100% viewfinder and the 8-8.5fps of the 7D, and in the end, I chose the 7D. I'm still going to get a 5D though; I have been waiting for the new model to come out to see if the AF improves. If it does, I will buy one, but I am not going to get rid of or stop using the 7D.
--

Click to expand...

Actually, I didn't think I was making the IQ difference out to be one way or the other. I'm just trying to find out whether a crop sensor with a higher pixel density (such as the 7D's) offers a significant reach advantage over full frame sensors in focal length limited situations.

More pixels are better. Case in point: Nikon's V1:

http://www.lensrentals.com/blog/2012/01/a-bit-of-3250mm-fun

I was trying to answer as best I could, but maybe I wasn’t clear. As long as you’re using recent cameras of similar generations, pixel density is going to have less impact than you seem to expect.

So, for example, I think you’d get better cropped results with a Rebel and a top-quality lens than with a 5D2 and an average lens. And with the same lens mounted on each, I’ve never seen enough difference to justify a decision based on a comparison of highly-cropped images (at modest ISO speeds).

Martin Ocando said:

Well, is not exactly reach. You are simply cropping your images with an APS-C sensor. I mean, magnification remains the same in comparison to a FF camera, but indeed it gives the "appearance" of increased reach.

Click to expand...

Don't agree. Let's say you're shooting with a 5DII in a focal length limited situation. Your subject is too small or too far away to fill the 5DII's frame, and only fills a portion of the frame/sensor equivalent to the same size as the 7D's smaller sensor. So, you could shoot with the 7D without any cropping, or shoot with the 5DII and crop down to the same angle of view. Same result? Not necessarily. The cropped 5DII image will only be comprised of 8 megapixels, while the 7D's image (due to its higher density sensor) will be comprised of 18 megapixels. Common sense would seem to suggest that the 7D image should be significantly sharper. Hence, the 7D's supposed reach advantage.

The question is, does this bear out in practice, or are other factors at work that tend to balance things out.

I've seen all kinds of theoretical discussions going both ways. But what I haven't seen are any definitive tests.

Put differently.... You’re either going to capture an image that will stand up to that extreme magnification or you’re not. Pixel density isn’t likely to be the limiting factor at that point, at least not in a significant way.

Phil Hill said:

Put differently.... You’re either going to capture an image that will stand up to that extreme magnification or you’re not. Pixel density isn’t likely to be the limiting factor at that point, at least not in a significant way.

Click to expand...

So you're saying that you don't think the pixel density difference in the sensors will make much difference in the final image. Understand, what we're talking about is a final image comprised of 8MP from the 5DII (closer to 4MP from a 5D) versus an 18MP image from the 7D (due to its significantly higher pixel density).

I appreciate your opinion, which is shared by some, but by no means all. That's why I'm interested in seeing some good test results.

bhollis said:

Actually, I didn't think I was making the IQ difference out to be one way or the other. I'm just trying to find out whether a crop sensor with a higher pixel density (such as the 7D's) offers a significant reach advantage over full frame sensors in focal length limited situations.

Click to expand...

It does.

--
Caution: Do not stare into laser with remaining eye.

bhollis said:

Liquidstone, If you're out there reading this, I suspect you have some practical experience with this issue, and I'd be very interested in your views.

Click to expand...

With sharp lenses (like the big whites), the 7D's smaller pixels translate to more reach, at least at lower ISOs (400 and below).

Here's my test between the 7D and 5D2 when shot from the same distance, using the same lens.

http://www.pbase.com/liquidstone/7dvs5d2

I reckon I can live with a 72 MP 7Dxx - I can then perhaps minimize the use of TCs, and reap the benefit of fast AF unslowed by TCs.

--
Romy



PHILIPPINE WILD BIRDS (Over 260 species captured in habitat, and counting.)
http://www.romyocon.net

Long Lens Imaging (1000 mm and beyond)
http://www.longlens.net

You are right, but that is not what I'm talking about. Is not more reach what you are acquiring, is just cropping the same FF image with a smaller frame, but with more pixels. Magnification and DOF remains the same.

I mean, a 250mm lens on an APS-C camera will not give you the shallower DOF and perspective compression as a 400mm lens on a FF camera. It will be of a 250mm lens. Apparent focal length is the same, but the image will be different in DOF and perspective.

bhollis said:

Martin Ocando said:

Well, is not exactly reach. You are simply cropping your images with an APS-C sensor. I mean, magnification remains the same in comparison to a FF camera, but indeed it gives the "appearance" of increased reach.

Click to expand...

Don't agree. Let's say you're shooting with a 5DII in a focal length limited situation. Your subject is too small or too far away to fill the 5DII's frame, and only fills a portion of the frame/sensor equivalent to the same size as the 7D's smaller sensor. So, you could shoot with the 7D without any cropping, or shoot with the 5DII and crop down to the same angle of view. Same result? Not necessarily. The cropped 5DII image will only be comprised of 8 megapixels, while the 7D's image (due to its higher density sensor) will be comprised of 18 megapixels. Common sense would seem to suggest that the 7D image should be significantly sharper. Hence, the 7D's supposed reach advantage.

The question is, does this bear out in practice, or are other factors at work that tend to balance things out.

I've seen all kinds of theoretical discussions going both ways. But what I haven't seen are any definitive tests.

Click to expand...

--
Martin Ocando
-------------------------

liquidstone said:

With sharp lenses (like the big whites), the 7D's smaller pixels translate to more reach, at least at lower ISOs (400 and below).

Here's my test between the 7D and 5D2 when shot from the same distance, using the same lens.

http://www.pbase.com/liquidstone/7dvs5d2

I reckon I can live with a 72 MP 7Dxx - I can then perhaps minimize the use of TCs, and reap the benefit of fast AF unslowed by TCs.

Click to expand...

Thanks. That's very helpful.