How to differentiate crop vs. FF

[privatebydesign]

Pro FF:
+DR
+DoF
+Better high ISO performance (pixels are 2x larger on 5D Mkiii than on 70D for example)

Pro APS-C
+much cheaper in general

Crop factor 1.6x is not really an advantage. If your take just the center of the 5D Mkiii frame you get ~9.4MP which is enough to print A2 pictures without quality loss. But on the other side in good like conditions you can crop the image form an APS-C sensor too.

This brings me the following question. For wildlife which is better? APS-C or FF?

And when I want the center 30% of an APSc frame?

Crop wins for focal length limited situations.

Naive and simplistic. What does crop 'win'?

With regards outright IQ technique will have a much larger impact on your images than a crop camera or cropping a ff camera. If you have a ff camera getting a crop camera will give you a very very small realisable resolution difference in focal length limited situations. It will give you different AF, framing, and fps figures which might very well have more impact.

If you have a 1Dx getting a 7D MkII for 'focal length limited situations' will almost certainly make little sense, if you have a 6D or a 5D MkIII it makes a much more complimentary tool.

Here are images I have posted many times, same generation 1Ds MkIII and 7D files from a focal length limited setup, the 7D crop has well over twice the pixels on target as the ff crop, but the differences in non optimally processed images, even at greater than 100%, are minimal.

 

[Lee Jay]

Pro FF:
+DR
+DoF
+Better high ISO performance (pixels are 2x larger on 5D Mkiii than on 70D for example)

Pro APS-C
+much cheaper in general

Crop factor 1.6x is not really an advantage. If your take just the center of the 5D Mkiii frame you get ~9.4MP which is enough to print A2 pictures without quality loss. But on the other side in good like conditions you can crop the image form an APS-C sensor too.

This brings me the following question. For wildlife which is better? APS-C or FF?

And when I want the center 30% of an APSc frame?

Crop wins for focal length limited situations.

Naive and simplistic. What does crop 'win'?

Resolving power and noise.

With regards outright IQ technique will have a much larger impact on your images than a crop camera or cropping a ff camera. If you have a ff camera getting a crop camera will give you a very very small realisable resolution difference in focal length limited situations.

About 50-70% depending on the full-frame camera. I have a 5D. The 7D2 will have about 90% more linear pixel density, and that most certainly will make a huge difference.

I've tested this myself, and even the difference between my T2i, 20D and 5D is a near-linear improvement with pixel density. Teleconverters wouldn't work if this weren't true, but they do work.

 

[Loswr]

If you have a 1Dx getting a 7D MkII for 'focal length limited situations' will almost certainly make little sense...

I'm certainly not planning to get a 7DII, it really offers no meaningful advantage for me.

 

[privatebydesign]

Resolving power and noise.

Funny, that isn't what my actual crops illustrate.

About 50-70% depending on the full-frame camera. I have a 5D. The 7D2 will have about 90% more linear pixel density, and that most certainly will make a huge difference.

I've tested this myself, and even the difference between my T2i, 20D and 5D is a near-linear improvement with pixel density. Teleconverters wouldn't work if this weren't true, but they do work.

Linear pixel density is a figure quoted by people who don't do same generation comparisons. I just posted a genuine unaltered same generation comparison and there is no 50-70% difference in either. On the other hand you believe a comparison is valid if you just go on pixel numbers or pixel size regardless of sensor generation, nobody else does.

The differences you will see between a 5D and a 7D MkII are down to the nine years difference in sensor tech and technique, nothing else. Compare a cropped 6D and a 7D MkII and I am fairly sure we will see the same differences I illustrate, practically none.

Indeed that 6D and 7D MkII is the spiritual upgrade for the thousands of people that owned both the 5D MkII and 7D.

There are real benefits to getting a crop camera, especially one like the 7D MkII for focal length limited situations, but 'Resolving power and noise' are not two of them, resolving power can get a nod in ideal situations, noise never seems to. The true benefits are AF, frame rate, viewfinder framing, and cost.

 

[Lee Jay]

In essence what you are saying is that there's no point to increasing focal length. A 400/5.6 is no better than a 70-200/2.8 according to you. They have the same aperture, and similar optical quality, and an increase in focal length is exactly the same thing as a decrease in pixel size as far as resolving power goes.

Of course, you're just wrong.

 

[privatebydesign]

In essence what you are saying is that there's no point to increasing focal length. A 400/5.6 is no better than a 70-200/2.8 according to you. They have the same aperture, and similar optical quality, and an increase in focal length is exactly the same thing as a decrease in pixel size as far as resolving power goes.

Of course, you're just wrong.

No, that is absolutely not what I am saying.

What I am saying is 'in same generation sensors the differences between a crop camera, and cropping a ff camera to get the same fov, especially appropriate in focal length limited situations, shows negligible difference in resolution even in ideal situations'. You can only test this with same (or very close) generation sensors, 6D and 7D MkII would be valid, 5D MkII and 7D would be valid, etc etc, if you don't have two same generation sensors you cannot test this and any 'illustrations' are entirely bogus, as all yours have been when we have danced this dance previously. Specifically in this thread I am referencing a 7D MkII and current FF Canon cameras, the 6D, 5D MkIII and 1DX, now obviously none of us have a 7D MkII yet, but when we do I am fairly sure we will see the same thing.

So, same lens, same place, same subject magnification, crop ff to crop camera framing and look at the difference, exactly as I illustrated above, when you do that in realistic enlargements there is no visible difference, when you take enlargement up to well over 100%, as my crops do, the differences are just visible. The crop camera has slightly more resolution and more noise, if you mitigate the noise you lose the detail, if you sharpen the FF you get more noise, if you normalise them it is a wash even at these extreme enlargements.

This has been true for several generations of sensors and I tested it myself because I wanted a 7D, but after doing this chose to not get one. Now I would hope a 7D MkII would show an improvement over the MkI results, but it looks like I'd still be better off getting a 1D MkIV !

It is funny, I have had this similar conversation here several times, normally with people who don't own both, they argue endlessly about the 'resolving power' of the crop camera, the 'pixels on target advantage' etc, invariably when they get a FF camera they too don't see much difference.

Neuro, who used to own the 5D MkII and 7D, sold the 7D after getting his 1DX and doing similar tests, AlanF used to categorically state the difference was massive until he got his 5D MkIII and on and on............

Show me some 5D MkIII-7D MkII, or 5D MkII-7D >100% crops using the same lens from the same place to prove I am wrong, and my 1Ds MkIII and 7D crops which illustrate my specific point perfectly are also wrong.

 

[Lee Jay]

What I am saying is 'in same generation sensors the differences between a crop camera, and cropping a ff camera to get the same fov, especially appropriate in focal length limited situations, shows negligible difference in resolution even in ideal situations'. You can only test this with same (or very close) generation sensors, 6D and 7D MkII would be valid, 5D MkII and 7D would be valid, etc etc, if you don't have two same generation sensors you cannot test this and any 'illustrations' are entirely bogus, as all yours have been when we have danced this dance previously.

Baloney. I specifically went out and proved you wrong by shooting intentionally at highly non-ideal settings, and still the difference is quite substantial. Here it is again. The differences are much larger using ideal settings.

 

[Loswr]

What I am saying is 'in same generation sensors the differences between a crop camera, and cropping a ff camera to get the same fov

Baloney. I specifically went out and proved you wrong

So, he states same generation sensors and cropping the FF sensor to the crop FoV, and you prove him wrong by showing (jugding by the image titles and pixel dimensions) a comparison between cameras from 2004, 2005, and 2010 (a 5-6 year technology gap), and upscaling some of the images.

That was a complete fail. Care to try again?

 

[Loswr]

Neuro, who used to own the 5D MkII and 7D, sold the 7D after getting his 1DX and doing similar tests, AlanF used to categorically state the difference was massive until he got his 5D MkIII and on and on............

I initially compared the 7D to the 5DII, and came to the same conclusion you have – APS-C image compared to FF image cropped to APS-C FoV, the APS-C has slightly more detail and slightly more noise, and one can be traded for the other so in practice, the only difference is in the MP of the resulting image, and if ~8 MP is sufficient for your output there's no advantage to using the crop body from a sensor IQ standpoint. Still, the 7D offered significant advantages in other areas - AF, frame rate, weather sealing, etc. The 1D X obviated all of those advantages, so I sold the 7D.

 

[privatebydesign]

Neuro, who used to own the 5D MkII and 7D, sold the 7D after getting his 1DX and doing similar tests, AlanF used to categorically state the difference was massive until he got his 5D MkIII and on and on............

I initially compared the 7D to the 5DII, and came to the same conclusion you have – APS-C image compared to FF image cropped to APS-C FoV, the APS-C has slightly more detail and slightly more noise, and one can be traded for the other so in practice, the only difference is in the MP of the resulting image, and if ~8 MP is sufficient for your output there's no advantage to using the crop body from a sensor IQ standpoint. Still, the 7D offered significant advantages in other areas - AF, frame rate, weather sealing, etc. The 1D X obviated all of those advantages, so I sold the 7D.

I am sorry, I knew that from earlier threads, I didn't mean to imply anything other than your having found the same thing when doing the same limited test.

As I said, for those that had a 7D and 5D MkII (and unlike you didn't go to a 1DX) the spiritual upgrade is the 6D and 7D MkII, they offer very complimentary feature sets that maximise the crop camera advantages in birding, sports, action and focal length limited scenarios. It seems 1 series cameras really do do it all though.........

 

[jepabst]

One is not better than the other. Honestly, just depends on what you are shooting. The extra reach on the 7D from the 5D combined with a 200mm f/2.8 means you have 320mm reach for shooting sports. However, the 20mm on the 7D shooting landscapes means you are shooting at 32mm instead of 20mm. Do you want to shoot wide, long - what's your lens budget? This question is eternally pointless. They both make beautiful images.

 

[AcutancePhotography]

This question is eternally pointless. They both make beautiful images.

And that, I feel, is just the reality of the issue.

 

[privatebydesign]

One is not better than the other. Honestly, just depends on what you are shooting.

True.

The extra reach on the 7D from the 5D combined with a 200mm f/2.8 means you have 320mm reach for shooting sports.

Not true. Look at my crops.

However, the 20mm on the 7D shooting landscapes means you are shooting at 32mm instead of 20mm.

The EF-S 10-22 is generally considered a better performing lens than either the 16-35 or 17-40, so no lack of ultra wide for crop and the price difference is less than $50 for FF. Indeed the 24-105 f4 L IS is generally much cheaper than the directly comparable EF-S 17-55 f2.8 IS.

This question is eternally pointless.

Not true.

They both make beautiful images.

True

 

[Sporgon]

What I am saying is 'in same generation sensors the differences between a crop camera, and cropping a ff camera to get the same fov, especially appropriate in focal length limited situations, shows negligible difference in resolution even in ideal situations'. You can only test this with same (or very close) generation sensors, 6D and 7D MkII would be valid, 5D MkII and 7D would be valid, etc etc, if you don't have two same generation sensors you cannot test this and any 'illustrations' are entirely bogus, as all yours have been when we have danced this dance previously.

Baloney. I specifically went out and proved you wrong by shooting intentionally at highly non-ideal settings, and still the difference is quite substantial. Here it is again. The differences are much larger using ideal settings.

Lee, I think another area you are going wrong with this comparison is that the 5D requires a fair amount of careful sharpening, whereas ( if I remember rightly back to 2004), the 20D did not. Bear in mind that back in the day those two cameras were aimed at different typical users. If you tried to optimise the 5D first I don't think there would be such a difference, and as the other guys have said, they are referring to current, or current-but-one generations cameras.

What your test shows is that the later generation is better. No surprise there.

 

[jepabst]

One is not better than the other. Honestly, just depends on what you are shooting.

True.

The extra reach on the 7D from the 5D combined with a 200mm f/2.8 means you have 320mm reach for shooting sports.

Not true. Look at my crops.

However, the 20mm on the 7D shooting landscapes means you are shooting at 32mm instead of 20mm.

The EF-S 10-22 is generally considered a better performing lens than either the 16-35 or 17-40, so no lack of ultra wide for crop and the price difference is less than $50 for FF. Indeed the 24-105 f4 L IS is generally much cheaper than the directly comparable EF-S 17-55 f2.8 IS.

This question is eternally pointless.

Not true.

They both make beautiful images.

True

I'm not going to argue with you.

 

[privatebydesign]

One is not better than the other. Honestly, just depends on what you are shooting.

True.

The extra reach on the 7D from the 5D combined with a 200mm f/2.8 means you have 320mm reach for shooting sports.

Not true. Look at my crops.

However, the 20mm on the 7D shooting landscapes means you are shooting at 32mm instead of 20mm.

The EF-S 10-22 is generally considered a better performing lens than either the 16-35 or 17-40, so no lack of ultra wide for crop and the price difference is less than $50 for FF. Indeed the 24-105 f4 L IS is generally much cheaper than the directly comparable EF-S 17-55 f2.8 IS.

This question is eternally pointless.

Not true.

They both make beautiful images.

True

I'm not going to argue with you.

Expressing an opinion is not necessarily arguing. This is a forum, a place for discussion ideas and thoughts. I expanded yours with my own and gave comments with easily verifiable facts to support my differing opinions, hardly an argument.

 

[Bob Howland]

I have both a 7D and a 5D3 and use and like both, but for different things. I use Raw (almost) exclusively. Simply stated, the trade-off is (1) higher pixel density (i.e., smaller pixels), (2) larger sensor and/or (3) high frame rate. At any given price point, choose any two of the above.

The 5D3 gives better image quality, especially in very low light. Its Auto-ISO max is set to 12800 whereas the 7D's is set to 3200. In particular, the 5D3 seems to have more DR. Furthermore, regardless of what Canon says, I don't believe that the 7D2 image quality will match or exceed the 5D3's (or 6D's) image quality.

The 7D uses smaller pixels than the 5D3. Thus, using a given lens focal length at a given distance from the subject, the 7D uses more pixels to make the image, unless the subject is so large that it "overfills" the 7D's field of view. A FF camera with the same pixel density as the 7D would have about 47MP. To maintain the same data frame rate as the 7D through the image processing circuitry would require dropping the FPS from 8 to about 3.2. Raising the frame rate back to 8 would requiring using better, faster and, therefore, more expensive circuitry.

Also, the smaller sensor of the 7D should have cost implications regardless of the number of pixels. The mirror and related components can be smaller and lighter and, in principle, cheaper and easier to produce.

So, what do I use and where? My default camera is the 5D3. The 7D is used almost entirely to photograph racing cars and motorcycles with a 100-400, 70-200 or 300 f/2.8 with or without TCs. When doing that, the 5D3 is usually also at hand, with a shorter lens attached. This is a role which the 5D3 can play but its predecessor, a 5D, couldn't. I suspect a 6D can't either. Anyway, unlike some, I find the extra "reach" to be real and advantageous. I suppose I could buy a 600 f/4 or Sigma 200-500 f/2.8 and a 1Dx , but then I'd have to carry them.

 

[privatebydesign]

Anyway, unlike some, I find the extra "reach" to be real and advantageous.

Got any comparison images to back that up Bob?

 

[Lee Jay]

Lee, I think another area you are going wrong with this comparison is that the 5D requires a fair amount of careful sharpening, whereas ( if I remember rightly back to 2004), the 20D did not.

Other way around, actually.

Bear in mind that back in the day those two cameras were aimed at different typical users. If you tried to optimise the 5D first I don't think there would be such a difference, and as the other guys have said, they are referring to current, or current-but-one generations cameras.

Each had their own optimized sharpening already applied.

What your test shows is that the later generation is better. No surprise there.

The 20D and the 5D were the same generation, and the T2i was really not much better per unit of sensor area, except for the higher pixel count.

 

[Lee Jay]

What I am saying is 'in same generation sensors the differences between a crop camera, and cropping a ff camera to get the same fov

Baloney. I specifically went out and proved you wrong

So, he states same generation sensors and cropping the FF sensor to the crop FoV, and you prove him wrong by showing (jugding by the image titles and pixel dimensions) a comparison between cameras from 2004, 2005, and 2010 (a 5-6 year technology gap), and upscaling some of the images.

The images are all shown with the same sensor area. The only meaningful difference is how that area is divided into pixels. As expected, the one with the smaller divisions wins in each case, and the 20D and 5D were of the same generation and same sensor performance per unit of area.

I'll say it again - smaller pixels are the same thing as a longer focal length, as any astrophotographer knows. That's why the astro folks generally talk about "image scale" which is measured in arc-seconds per pixel, rather than sensor size, pixel count and focal length. They know what actually matters - aperture diameter for light gathering and image scale for resolving power, subject to the limits set by diffraction and the atmosphere.

Saying smaller pixels don't help with "reach" (resolving power) is the same as saying a longer focal length doesn't help with reach, and it's just as wrong.