Image quality: Extender on FF vs crop

I was wondering. How does a 7D with a 70-200mm compare to a FF with 70-200mm and a 2x Extender III?
Obviously we are taking about f/5.6 vs f/2.8, but the difference in high ISO performance should make up for that.
So about the sharpness?

7D will have always advantage with 1.6 crop factor and pixel density in center of the frame.

I would probably only go for a 1.4x extender, last month I was thinking of 2x, didn't want to risk focus speed and IQ.
2x extender will de grade the lens no matter what camera you choose.

2x good for a prime lens with f/2.8 or f/4
for a zoom a zoom i use 1.4x
that's what I know
i remember the canon extender x2 on my prime lens 200 f/2.8L II with FF is better than when I use it on a crop body

For me my 70D produces similar if not a little better IQ with the bare lens than my 6D with an 1.4x TC. But they are very close, and actually other factors would decide which combination I'd use.

The situation is similar when I compare the 70D + 1.4x + lens vs 6D + 2x + lens. The 70D +1.4x tc is a little better.

Plus the angle of view is always smaller with the 70D combos than with the 6D + tc (or stronger TC) combos, e.g. 672mm equiv. vs 600mm (70D + 1.4 vs 6D + 2x)

I have a sharp Sigma 120-300 2.8 OS that I use with Sigma 1.4x and 2x TC (very very good copies, my 2x is sharper than 2 copies of Canon 2xIII I tested).

Back to your original question, I'd rather use the 7D + 70-200 (and perhaps a 1.4x TC if needed), instead of a FF with a 2x TC.

uh-oh, the old 'debate' again.

Any lens can deliver only a certain amount of optical detail resolution to the focal plane, regardless of the actual size of the frame at the focal plane. The 70-200mm lens delivers identical optical performance to the T3i as it does to the 5DIII.

Any lens' optical resolution is 'divided by' the degree of magnification in order to achieve the same final print size. Where FF takes 16.9x magnification to make a 16" x 24" final print, APS-C requires about 27.3x magnification to produce the same size print. 24mm (FF) * 16.9 = 405.6mm, 14.9mm (APS-C) * 27.3 = 406.8mm (16" = 406.4mm)

Assuming I used 160mm FL on 20 MPixel FF body, and set the zoom to 100mm FL on 20 Mpixel APS-C body, I end up with the same amount of subject and surroundings filling each frame. So 16.9x FF print will intially appear to be identical in content with 27.3x APS-C print. And (assuming identical detail resolution at both FL -- yes, probably not true in reality!), the FF print will appear to be sharper than the APS-C print because it has 1.6x greater detail resolution per inch on print. Assuming identical pixel count for both bodies, APS-C has even less detail than FF, just as 135 has less detail than large format.

This is because (assuming the mythical spectacular lens with 1000 lines/millimeter detail at focal plane) 1000/16.9 is greater than 1000/27.3 ! This principle has long been the explanation for why Large Format outperformed the 135 format for detail, and why some magazines REFUSED TO USE the smaller 135 format images!

A telextender will degrade the image the same factor regardless of image size. photozone.de used to publish test results with the Canon 1.4x which showed a loss of about -10% in detail resolution of the primary lens. (1000*0.9)/16.9 is still larger than (1000*0.9)/27.3
So if we assume a -20% for a TC2x, and 400mm on 20 Mpixel FF vs. 260mm on 20Mpixel APS-C for identical framing, (1000*0.8)/16.9 is still larger than (1000*0.8)/27.3

I tested this not to long ago out of my own curiosity, just for reference sake.

Lens-70-200F/4 non-IS
Bodies- 60D and 6D
Extender- Canon 1.4X II

I was quite impressed with the results.

60D

IMAGE LINK: https://www.flickr.com​/photos/lenzfreak/1234​4771543/

IMAGE LINK: https://www.flickr.com​/photos/lenzfreak/1233​7233504/

The F stops here wrote in post #16779385
I tested this not to long ago out of my own curiosity, just for reference sake.

I was quite impressed with the results.

Aside from the topic at hand, some awesome images!

Heya,

When it comes to this question, I really suggest you simply get a longer telephoto lens, instead of trying to make a 200mm lens into a 400mm lens. Extenders are not benign. It depends on what you want to do and what your threshold is on acceptable quality from your 70-200 F2.8, when you see the results.

Very best,

palad1n wrote in post #16779199
7D will have always advantage with 1.6 crop factor and pixel density in center of the frame.

Not necessarily true to say "APS-C will always have advantage over FF".

Let us compare 30D vs. 5DII:

• Pixel density is identical at 156 pixels/mm.
  
• Same subject from same distance will be identically sized on the sensor of both cameras; the 5DII simply captures more surrounding objects within its larger frame area. An 7.5' object at a distance of 100' will fill the approx 15mm APS-C frame height using 200mm lens, but the 7.5' object will also be 15mm on the FF sensor -- we also capture 4.4' more of the surrounding area.

Wilt wrote in post #16779417
Not necessarily true to say "APS-C will always have advantage over FF".

Let us compare 30D vs. 5DII:

• Pixel density is identical at 156 pixels/mm.
  
• Same subject from same distance will be identically sized on the sensor of both cameras; the 5DII simply captures more surrounding objects within its larger frame area. An 7.5' object at a distance of 100' will fill the approx 15mm APS-C frame height using 200mm lens, but the 7.5' object will also be 15mm on the FF sensor -- we also capture 4.4' more of the surrounding area.

your statement is true if you compare the typical 2014 APS-C dSLR vs. the typical 2014 FF dSLR, but it is not a universally true statement!

i didn´t mentioned that I mean only current DSLR on market.
Good point, Thank you!

As mentioned, I've had much better results with a 1.4X and a 70-200mm f2.8 then with a 2X on my 7D. In my view, I find 2X work best on Primes.

TJays wrote in post #16779612
As mentioned, I've had much better results with a 1.4X and a 70-200mm f2.8 then with a 2X on my 7D. In my view, I find 2X work best on Primes.

That doesn't make a whole lot of sense given the 70-200mm 2.8 II is as sharp or sharper than most primes.

Wilt wrote in post #16779362
A telextender will degrade the image the same factor regardless of image size. photozone.de used to publish test results with the Canon 1.4x which showed a loss of about -10% in detail resolution of the primary lens. (1000*0.9)/16.9 is still larger than (1000*0.9)/27.3
So if we assume a -20% for a TC2x, and 400mm on 20 Mpixel FF vs. 260mm on 20Mpixel APS-C for identical framing, (1000*0.8)/16.9 is still larger than (1000*0.8)/27.3

Hmm.
Looking at Dxomark scores for example. A 70-200mm II IS comes in at 21P-Mpix on a 5D3 and 12P-Mpix on a 7D. This would imply a near 40% resolution loss by going to a crop camera. And if it is true that the resolution impact of a 2x Extender III is 20%, this would suggest that an Extender is a much better option than a crop camera.
The F stops here's photos above also confirm this.

MalVeauX wrote in post #16779398
Heya,

When it comes to this question, I really suggest you simply get a longer telephoto lens, instead of trying to make a 200mm lens into a 400mm lens. Extenders are not benign.

Not practical for my needs given the cost, and size/bulk of these lenses.

Mornnb wrote in post #16780315
Hmm.
Looking at Dxomark scores for example. A 70-200mm II IS comes in at 21P-Mpix on a 5D3 and 12P-Mpix on a 7D. This would imply a near 40% resolution loss by going to a crop camera. And if it is true that the resolution impact of a 2x Extender III is 20%, this would suggest that an Extender is a much better option than a crop camera.
The F stops here's photos above also confirm this.

I was simply comparing lens on FF vs. same lens on APS-C and the impact of taking ALL of the lens resolution and getting its detail to the 16x24" print...with all else being equal, but assuming you use the FL appropriate to the format, so that subject fills frame in each case at different FL. In this comparison, it is simply the enlargement mag factor which enters the consider, but does not factor in the ability of the sensor to capture the image.

Then there second situation as tested, which is a different test. There is the additional issue of losses due to actual capture, and here you have consideration of 21Mpixel vs. 12Mpixel for the full image. Of course, higher MPixel wins, assuming the subject fills the frame in each case.

This also really is the 'optical vs. digital zoom' issue! Should I crop to leave (fewer than captured) pixels, or should I use a telextender that zooms in a comparable amount...optical zoom is always better than lopping off pixels.

Mornnb wrote in post #16780315
Not practical for my needs given the cost, and size/bulk of these lenses.

Heya,

Take a long hard look at the weight and size of a 70-200 F2.8 lens, and a teleconverter. It will be probably about as heavy, if not heavier, and as big, if not maybe a hair shorter, than one of these 400mm lenses.

But if you are dead set on using 200mm + teleconverter, and you want it ignore that, by all means, carry on.

Very best,