I'm sorry to bring this up AGAIN.

[travellingcello]

Ok so the age old argument of sensor sizes - I've never thought about this in all that much detail until now, only what i thought I knew and what I observed in my own comparisons having used APS-C cameras in the past and now FF.

Is what this guy saying credible? I realise that times have changed a bit and top-of-the-range APS-C cameras like the 7D have closed the gap but: ( http://www.ophrysphotography.co.uk/pages/tutorialfullframeandcrop.htm )

I've seen the arguments where 5DII users argue that they can crop down to the equivalent APS-C sizes and still get the same quality as a 40D etc. etc, but in 'idiots terms' can somebody explain to me what is myth and what is true? All those threads I have seen on those topics get sidetracked and are inconclusive.

The reason I'm bringing this subject up again is because I was going through some of my old files and suprised again by the lack of detail I could see in files I took with APS-C cameras such as the 50D and 60D in comparison to FF images I have taken recently - I'm thinking in particular in terms of wildlife photography where I had assumed the APS-C cameras would be comfortable doing, and provide better IQ.

I'm also confused now by this paragraph in the 'Wildlife Photography' free ebook that was downloaded I think by many here as well...

'A fast, high-quality APS- C camera is a good choice, and has the advantage of providing 50 percent more range than the same lenses can produce when used with a full-frame camera. A 400 mm lens used with an APS-C camera will magnify your subject just as much as a 600 mm lens used with a full-frame camera.'

Can someone explain that to me? He says magnify the picture - but isn't that all it is literally if you can crop down a FF picture to get a picture as good as if not better than APSC?

I know that this has to do with sensor sizes, pixel density and all but that makes me wonder why a lot of people like me didn't understand why we bought APS-C cameras in the first place.

My own current understanding of the role of is APS-C cameras is that they are a good compromise, as they can be quick (FPS, lenses can be cheaper/good bang for buck and overall provide comparable IQ to a FF camera but the end result IQ is always going to be better on a cropped FF image???

I probably sound very confused and I am, someone explain please!

 

[33L]

I think one of the main reasons for a crop camera is the cost advantage on both lens and body and still produce great results.

 

[petersmart]

Ok so the age old argument of sensor sizes - I've never thought about this in all that much detail until now, only what i thought I knew and what I observed in my own comparisons having used APS-C cameras in the past and now FF.

Is what this guy saying credible? I realise that times have changed a bit and top-of-the-range APS-C cameras like the 7D have closed the gap but: ( http://www.ophrysphotography.co.uk/pages/tutorialfullframeandcrop.htm )

I've seen the arguments where 5DII users argue that they can crop down to the equivalent APS-C sizes and still get the same quality as a 40D etc. etc, but in 'idiots terms' can somebody explain to me what is myth and what is true?

Well the article above pretty much explains it - crop a FF image down to a crop equivalent and the picture and IQ will be about the same.

The reason I'm bringing this subject up again is because I was going through some of my old files and suprised again by the lack of detail I could see in files I took with APS-C cameras such as the 50D and 60D in comparison to FF images I have taken recently.

Is everything the same especially the lenses?

If you are using better lenses on the FF camera that can explain a lot.

I have owned a 350D, 450D and now the 1Ds MkII and always got good image detail from all three.

I'm also confused now by this paragraph in the 'Wildlife Photography' free ebook that was downloaded I think by many here as well...

'A fast, high-quality APS- C camera is a good choice, and has the advantage of providing 50 percent more range than the same lenses can produce when used with a full-frame camera. A 400 mm lens used with an APS-C camera will magnify your subject just as much as a 600 mm lens used with a full-frame camera.'

Can someone explain that to me? He says magnify the picture - but isn't that all it is literally if you can crop down a FF picture to get a picture as good as if not better than APSC?

In effect - Yes.

But don't forget that many people with a crop camera will then also crop the image as well giving the effect of a much longer reach lens still.

You can of course crop a FF camera to the equivalent of a 1000mm lens but other factors also come into play including noise, degree of blur etc.

The only true way is the way the writer of that article did - static subjects on a tripod.

I know that this has to do with sensor sizes, pixel density and all but that makes me wonder why a lot of people like me didn't understand why we bought APS-C cameras in the first place.

Well the reason I started with a 350D was that it was the best I could afford at the time.

My own current understanding of the role of is APS-C cameras is that they are a good compromise, as they can be quick (FPS, lenses can be cheaper/good bang for buck and overall provide comparable IQ to a FF camera but the end result IQ is always going to be better on a cropped FF image???

You last statement contradicts itself - they can't "provide comparable IQ to a FF camera" if the "IQ is always going to be better on a crop camera."

Everything else being equal (and they seldom are in the REAL world) IQ from a FF camera SHOULD be better than the IQ from a crop camera and in the hands of a good photographer the pictures USUALLY should be, but as so many pics on here prove that does not always apply and a good photographer can produce better pics from a crop camera than a poor photographer using a FF camera.

.

 

[artyman]

I think it boils down to pixels on the target, so a crop is going to give you more, however high pixel density and a crop sensor does rely on good technique. A FF sensor may appear to perform better on the noise front and hence give the impression of improved results, but I'm not giving up the 1.6 advantage my 7D gives me. As with most photographs, it's the operator that makes the biggest difference rather than the equipment.

 

[ausemmao]

Crop - cheaper, smaller cameras and total package, bit more DoF.
FF - more expensive, bigger package, bit less DoF, can be used to display how much disposable income one has.

If you operated in circumstances where the differences matter, you wouldn't need to ask the question. If you're asking, going from crop to FF won't appreciably change your images but you may feel better from buying the more expensive tool.

At low ISO if you're able to distinguish the newest crop and FF camera images when printed smaller than 13x19 without a side by side, the photographer has done something wrong with one of the images. High ISO's another story.

 

[Marcus Geezer]

Talking from experience, going from a 400D to 40D to 5D (classic), I hope I can offer something. From a megapixel point of view there is not much difference in the above (10, 10 and 12 respectively) so if looking from that point of view there is not much to gain, and also the 5D I have is actually an earlier model probably more comparable with the 20D or 30D range. What did it give me? Well I take mainly portrait work and my images are smoother with better tones as a result of the FF sensor, my work needs less PP work including contrast boost and sharpening and there is just a something about the images generally which makes them more pleasing to the eye (at least to mine).

It depends what you use your camera for, how you present your images, and what the images are to be used for to be honest.

 

[arad85]

I've seen the arguments where 5DII users argue that they can crop down to the equivalent APS-C sizes and still get the same quality as a 40D etc. etc, but in 'idiots terms' can somebody explain to me what is myth and what is true? All those threads I have seen on those topics get sidetracked and are inconclusive.

You have to understand there is a complete optical system at work here. It's not just the sensor that affects what you see, but the lens in use (and so how hard it has to work to provide contrast and detail (i.e. the things we perceive as sharpness) as well as the rendering on output (whether you render to print, or to screen) and how far away you view that output.

What should be obvious is that if you take a picture of an object from a particular position with a given lens on FF and crop, the image that is captured will cover the same physical area on both sensors. That is, a FF sensor just captures a wider field of view than a crop but that any object in the scene will cover the same sensor area in both cases.

If you think in terms of rendering NOT in terms of pixels and pixel peeping, but in terms of area magnified and viewed at a distance (as you would with a photo) things (IMHO) become easier to understand. Let's take an example: say you photograph a bird on both FF and crop cameras with the same lens and it covers an area of the sensor that is 6mm x 4mm on the sensor. Clearly, this will cover more of the crop sensor (that is ~22x15) than the FF sensor (36x24) but it still covers the same sensor area. In both cases, if you crop to the same field of view and then print it as an 6"x4" picture, you have magnified the resulting on-sensor image from 6mm x 4mm -> 6" x 4" or a 25.4x enlargement. They have both been magnified by the same amount to produce the image.

The only time pixels come into it is whether you can resolve the effect of individual pixels when you view the picture. This generally does not happen unless you zoom in and pixel peep, so consequently, for a given subject, there is very little to choose between FF and crop. In fact, there shouldn't be unless you get to the point of seeing the effect of the individual pixels (which might be possible when cropping really hard and printing large).

If you DO zoom in and pixel peep, the lens, focus and technique begins to become more important. With high pixel densities, the more the lens has to work to produce contrast (and hence sharpness). Depending on the lens and shooting conditions, these might actually work against you.

I'm also confused now by this paragraph in the 'Wildlife Photography' free ebook that was downloaded I think by many here as well...

'A fast, high-quality APS- C camera is a good choice, and has the advantage of providing 50 percent more range than the same lenses can produce when used with a full-frame camera. A 400 mm lens used with an APS-C camera will magnify your subject just as much as a 600 mm lens used with a full-frame camera.'

They are both right and wrong! The APS-C sensor gives the same field of view that a lens 1.6x longer would give on a full frame sensor. In terms of the image it projects onto the sensor, a 400mm lens will project a given object at a given size whether it is on a FF, APSC, MFT, medium format camera - whatever the sensor it is the object will be projected at the focal plane at exactly the same size. What changes is how many pixels it covers and this is determined by the pixel density. Traditionally, crop cameras have a higher pixel density so people think you are getting "more pixels on the subject". Yes, this is clearly true, but whether getting more pixels on the subject is important is determined by how you are viewing the rendered output. Also, you have to consider whether the lens is capable of allowing the higher pixel densities to capture the finer detail and you also have to consider the more you magnify the image as projected on the sensor, the more critical focus is and the more apparent it is that a lens only focuses at a single distance.

My own current understanding of the role of is APS-C cameras is that they are a good compromise, as they can be quick (FPS, lenses can be cheaper/good bang for buck and overall provide comparable IQ to a FF camera but the end result IQ is always going to be better on a cropped FF image???

My thinking is that producing a working full frame sensor is fundamentally actually quite costly and certainly more than 2.36x the price that the respective areas would suggest. It is this that makes FF cameras more costly and that APS-C are a good cost/quality compromise. If we only had FF cameras, entry into DSLR territory would be £1000+ just for the body, let alone a good quality lens to go on it....

 

[travellingcello]

Thanks for your replies, I think I 've pretty much now understood....!

 

[jamesoliverstone]

Well from real world experience, I have sold my FF D700 and gone to an APS-C Sony A77. I can tell you that there is far more detail in the images than any shot taken on my D700, but then I would expect it from the latest in sensor technology and double the amount of megapixels. But, I do miss the ultra shallow dof and there is a quality to a FF image that you wont get on APS-C... this however might be purely in my head...

The glass makes a big difference though, and most FF users tend to use pro grade glass, which is why a lot of the images from that camera will look stunning.

Tell you what, I have just recieved my 85mm 1.4 through the post this morning, so I will take a picture tonight if I can and post it here so you can see what the quality is like from a high end aps-c sensor... personally, I love it, massive amounts of detail and as sharp as a tack.

 

[DemiLion]

Crop - cheaper, smaller cameras and total package, bit more DoF.
FF - more expensive, bigger package, bit less DoF, can be used to display how much disposable income one has.

Really? Canon 1DMkIII v 5DMkII would indicate otherwise (apart from the DoF), as would several other comparisons. Generalisation can be dangerous at times, especially if envy is part of the equation.

Well from real world experience, I have sold my FF D700 and gone to an APS-C Sony A77. I can tell you that there is far more detail in the images than any shot taken on my D700, but then I would expect it from the latest in sensor technology and double the amount of megapixels. But, I do miss the ultra shallow dof and there is a quality to a FF image that you wont get on APS-C... this however might be purely in my head...

Nope it isn't (the feel and quality that is), and you've pretty much hit the nail on the head there!

 

[HoppyUK]

Thanks for your replies, I think I 've pretty much now understood....!

Andy has got it right, and the guy in your link hasn't taken everything into account and clearly got sensor sizes wrong. The crop factor is a linear measure, and the area measure, the one that matters here, is crop factor squared, ie 1.6 x 1.6 = 2.56 which makes a Canon cropped sensor less than half the area of full frame.

This is what's driving the quality difference, all other things being equal (which they usually are not!) and there are two aspects to it.

The larger area collects more light, more than twice as much, so there is more of the basic material needed to work with. Noise is lower, high ISO better, wider dynamic range, more and shadow detail with cleaner colours etc.

The other thing that most people forget is that there are two basic components to lens sharpness - resolution (the fineness of detail) and contrast (how clearly those details are shown). These are the two axes of a lens MTF graph which is what lens designers use to measure quality.

A fact of optical physics is that when resolution demands go up, contrast goes down, and it is image contrast (not the same thing as pumping up the contrast slider in post processing) that contributes most to perceived sharpness. For a given resolution, or call it pixel count if you like, the lens on a crop format camera has to work twice as hard and so contrast is much lower.

The end result is that full frame always has two big advantages over a cropper. If you compare two cameras with similar pixel counts like a Nikon D700 and D300, or a Canon 7D vs 5D2, then the full framer always wins.

The downside is full frame is expensive, and for telephoto work you need much bigger and more expensive lenses. There quite quickly comes a point where full frame is just too unwieldy and costly, eg compare a 7d with a 300mm f/4 lens and a 1Ds with 500mm f/4. So a better and more practical/affordable compromise is to use a shorter lens on a cropper and try to squeeze as much as you can out of their higher pixel density with the best quality lens you can lay your hands on.

 

[ausemmao]

Really? Canon 1DMkIII v 5DMkII would indicate otherwise (apart from the DoF), as would several other comparisons. Generalisation can be dangerous at times, especially if envy is part of the equation.

Where's the incorrect generalisation?

Would indicate what? WRT price, the 1Ds3 is more expensive than the 1D3 or 4, the 5D is a different class body. If you compare the same class body (D300 vs 700) oh look, the D700 is more expensive. Compare the 5D2 to the 7D (and good luck trying to show the 7D's a lower class body LOL) and again, the 5D2's more expensive.
Can be, not is. The two words don't mean the same thing.
As for showing off income, photography is no different to any other hobby in that respect. Some people's hobby is in the collecting, others in the using, others in the showing. I never said either was good or bad .

 

[DemiLion]

No probs; but if you meant 'in class' comparisons (not that there really are that many (1Dv1Ds and D3svD3x being the only ones that I can think of), then say so from the outset.

 

[ausemmao]

No probs; but if you meant 'in class' comparisons (not that there really are that many (1Dv1Ds and D3svD3x being the only ones that I can think of), then say so from the outset.

Ah, fair enough. I should have been clearer. I get engineer brain and assume only the thing being talked about is changing, otherwise it's too easy to bring out edge cases to support any argument. Otherwise somebody might claim IQ is inversely related to sensor size because a 7D image is better than a DCS760 one .