Estimating extra reach (resolving power) of crop vs FF

[AlanF]

Here is a way of calculating the effective extra reach or resolving power of a crop body versus FF, which will amuse the geeks among us.

Measure the MTF of a lens on the crop (= MTFcrop) and the same lens on the FF (= MTFff). The ratio of the MTFs, MTFcrop/MTFff, gives the relative resolving power of the bodies with that lens. However, the crop body can be placed 1.6x further away to give the same field of view. Therefore, the true effective relative resolving power, R, is given by:

R = 1.6x MTFcrop/MTFff.

Photozone lists measured MTFs for a set of lenses on the 5DII and 50D. I calculated their ratios for the Canon 200mm f/2.8 II, 85mm f/1.2 II and 35mm f/2 at wide apertures below the DLA. MTFcrop/MTFff is very close to 0.726 in all cases.

This gives R for 50D/5DII = 1.16.

So the effective extra reach is 16%. (Based on the ratio of their pixel sizes, a value of 36% is expected.

The dpreview widget gives values for the 5DIII and 7D only for a few lenses. I did the same calculations with the Tamron 150-600mm (between 150-400mm), the Canon 200-400mm and the Sigma 50mm f/1.4 A at wider apertures below the DLA. In all cases, MTFcrop/MTFff is close to 0.742.

This gives R for 7D/5DIII = 1.19.

So, the effective extra reach is 19%. (Based on the ratio of their pixel sizes, a value of 45% is expected).

There are always arguments about using MTFs quantitatively, but I think in this particular calculation it is reasonably valid to use them. It fits in reasonably well with experience - Jon has shown there is better resolving power in photos of the moon with the 7D, but it doesn't look 45% better. And my own experience is that the 7D and 70D aren't much better than the 5DIII, certainly not 1.6x.

 

[privatebydesign]

That might be true in a testing scenario, but few of us shoot in those. Factor in AF, handholding, higher than base iso, less than ideal aperture or shutter speed etc etc etc and the differences become minimal, as so many people who have owned both have attested to.

That doesn't mean there is no point to a 7D, 70D 7D MkII, as a compliment to a 6D etc one might work very well, but the resolution thing really is a red herring unless you are using a heavy tripod, 10X live view manual focus blah blah.........

 

[Deleted member 91053]

I have found little difference in reach when changing sensor size, though I have noticed improved IQ.
I don't carry out scientific comparisons I just go out and see which performs better.
I do most of my photography with the Canon 800mm so if there were much of a difference this is the lens to show it up!
Now if Canon/Nikon made a low MP APSC camera to the same standard as the 1DX/D4 then the situation may be different!

 

[AlanF]

That might be true in a testing scenario, but few of us shoot in those. Factor in AF, handholding, higher than base iso, less than ideal aperture or shutter speed etc etc etc and the differences become minimal, as so many people who have owned both have attested to.

That doesn't mean there is no point to a 7D, 70D 7D MkII, as a compliment to a 6D etc one might work very well, but the resolution thing really is a red herring unless you are using a heavy tripod, 10X live view manual focus blah blah.........

What the estimates tell you is that even if you have a heavy tripod, base iso etc you will gain only a small increase in reach, and not 60%.

 

[Loswr]

This gives R for 7D/5DIII = 1.19.

So, the effective extra reach is 19%. (Based on the ratio of their pixel sizes, a value of 45% is expected).

That might be true in a testing scenario, but few of us shoot in those. Factor in AF, handholding, higher than base iso, less than ideal aperture or shutter speed etc etc etc and the differences become minimal, as so many people who have owned both have attested to.

As I've stated, the most significant advantage to APS-C is lower cost.

 

[privatebydesign]

That might be true in a testing scenario, but few of us shoot in those. Factor in AF, handholding, higher than base iso, less than ideal aperture or shutter speed etc etc etc and the differences become minimal, as so many people who have owned both have attested to.

That doesn't mean there is no point to a 7D, 70D 7D MkII, as a compliment to a 6D etc one might work very well, but the resolution thing really is a red herring unless you are using a heavy tripod, 10X live view manual focus blah blah.........

What the estimates tell you is that even if you have a heavy tripod, base iso etc you will gain only a small increase in reach, and not 60%.

I've been saying that for years and practically nobody agreed with me, it is great to see the winds of change, finally.

I deliberately stayed out of the last 5D MkIII and 7D reach comparison thread, it was interesting that after ironing out some flawed methodology the same conclusions were drawn that I did a long time ago, albeit to the utter disdain and disbelief of the then only 7D owning OP.

I wonder if people will believe me on some of the other contentious stuff I say now too? I doubt it..........

 

[AlanF]

I did the same calculations for the Nikon D800 and D7000. You would expect them to have virtually identical resolving powers since their pixel sizes are 4.88 and 4.78 µM, respectively. And that is what I calculated, giving some credence to the calculations I did for the Canons.

MTFcrop/MTFff with identical lenses is close to 0.63 for all lenses (on the dpr widget), compared with 0.66 expected from the relative pixel heights of both sensors. The relative resolving power of the crop, given the 1.5x less field of view is:

R = 0.945 (compared with R= 1.02, calculated from the ratio of pixel sizes).

The 36.6 mpx ff beats out the 16 mp crop since, when cropped it is at least as good as the crop sensor and has all the advantages of FF when not cropped.

 

[AcutancePhotography]

I wonder if people will believe me on some of the other contentious stuff I say now too? I doubt it..........

If you support your opinions with evidence that can be independently evaluated... yes
If you just state your opinion... no

 

[FEBS]

The dpreview widget gives values for the 5DIII and 7D only for a few lenses. I did the same calculations with the Tamron 150-600mm (between 150-400mm), the Canon 200-400mm and the Sigma 50mm f/1.4 A at wider apertures below the DLA. In all cases, MTFcrop/MTFff is close to 0.742.

This gives R for 7D/5DIII = 1.19.

So, the effective extra reach is 19%. (Based on the ratio of their pixel sizes, a value of 45% is expected).

There are always arguments about using MTFs quantitatively, but I think in this particular calculation it is reasonably valid to use them. It fits in reasonably well with experience - Jon has shown there is better resolving power in photos of the moon with the 7D, but it doesn't look 45% better. And my own experience is that the 7D and 70D aren't much better than the 5DIII, certainly not 1.6x.

So if the new 7D2 might have a 20.2mp in stead of 18mp of 7D then the effective extra reach would be about 25% roughly?

That makes me think about my idea of buying the 7D2. My extra reach will not be that high as expected. Truly, that was something I did find remarkable after I bought a 5D3, that I could crop much more compared to the 7D. This should be the scientific prove for that. If It was really a 1.6 factor, then the details should also stay more present in the photo, and that's not the same if you compare in Lightroom for both cameras. The 5D3 gives more possibility to crop, so that in practice the 60% is only 19% in advance for the 7D.

For sure something to keep in mind.

Thanks for this explanation Alan.

 

[Mt Spokane Photography]

I did the same calculations for the Nikon D800 and D7000. You would expect them to have virtually identical resolving powers since their pixel sizes are 4.88 and 4.78 µM, respectively. And that is what I calculated, giving some credence to the calculations I did for the Canons.

MTFcrop/MTFff with identical lenses is close to 0.63 for all lenses (on the dpr widget), compared with 0.66 expected from the relative pixel heights of both sensors. The relative resolving power of the crop, given the 1.5x less field of view is:

R = 0.945 (compared with R= 1.02, calculated from the ratio of pixel sizes).

The 36.6 mpx ff beats out the 16 mp crop since, when cropped it is at least as good as the crop sensor and has all the advantages of FF when not cropped.

The issue with the D800 was that even carefully done tests by experienced testers, DPR, for example, had to use extreme measures to get the resolution the sensor is capable of. Ordinary users do not stand a chance, since even the tiniest vibration blurs some of the pixels. You must use high shutter speeds, and for telephoto shots, where you want more reach, its even more difficult. Turn up the ISO so you can use a faster shutter speed, stop down to increase lens resolution, and you loose resolution due to noise.

The gain is real, but almost unattainable, even by experts.

 

[NancyP]

Don't knock "lower cost", not to mention lower weight. Not everyone is going to be forking over $6,600.00 for an action camera. A reasonably priced "mini-1DX"-spec'ced (wrt fps and buffer) 7D2 is a very attractive compromise for those who ALSO don't have a Big White yet.

 

[privatebydesign]

Nancy, nobody does.

And focal length limited doesn't mean there isn't a lens available, it just means you don't have it to use, it doesn't matter if you don't own it or you don't have it with you.

I rarely take longer than a 70-200 with me unless I know I will need it, but that 200 is my limit whether I have a 7D or a 5D MkIII. AlanF has given us a sound mathematical explanation as to why the pixel numbers don't add up to the resolution differences we see.

My take was back in the 7D and 5D MkII days when people owned both because they thought the 7D gave them a lot more resolution over a cropped 5D MkII, but it doesn't, it does give them better AF and frame rate, but not a noticeable increase in resolution from real world shooting. My suggestion at that time to people wanting a 5D MkII and a 7D was to not get either but get a 1Ds MkIII instead, it covered both cameras stills capabilities for around the same money. AlanF has just presented us with a real equation as to why the "pixels on duck" meme was a complete fallacy.

 

[RustyTheGeek]

Don't knock "lower cost", not to mention lower weight. Not everyone is going to be forking over $6,600.00 for an action camera. A reasonably priced "mini-1DX"-spec'ced (wrt fps and buffer) 7D2 is a very attractive compromise for those who ALSO don't have a Big White yet.

LOL Nancy! ;D Funny, I was thinking the same thing! I really like nuero but it's easy to make his statement when one owns the gear he does. (And over the years I've spent more than I care to admit as well.)

Note: I just purchased a Canon Refurb 70D 18-55 Kit for $836 from Canon because I don't want to spend the huge $$ on the likely over $2K 7D2. (And I think it will likely follow the same path as the previous 7D/60D releases and have a similar if not the same sensor anyway.) My reason? More reach with better crop sensor and higher fps (than the 60D). That's about it.

 

[wsmith96]

That might be true in a testing scenario, but few of us shoot in those. Factor in AF, handholding, higher than base iso, less than ideal aperture or shutter speed etc etc etc and the differences become minimal, as so many people who have owned both have attested to.

That doesn't mean there is no point to a 7D, 70D 7D MkII, as a compliment to a 6D etc one might work very well, but the resolution thing really is a red herring unless you are using a heavy tripod, 10X live view manual focus blah blah.........

What the estimates tell you is that even if you have a heavy tripod, base iso etc you will gain only a small increase in reach, and not 60%.

edited for asking my question within the quote - sorry

Reach or resolution?

 

[AlanF]

That might be true in a testing scenario, but few of us shoot in those. Factor in AF, handholding, higher than base iso, less than ideal aperture or shutter speed etc etc etc and the differences become minimal, as so many people who have owned both have attested to.

That doesn't mean there is no point to a 7D, 70D 7D MkII, as a compliment to a 6D etc one might work very well, but the resolution thing really is a red herring unless you are using a heavy tripod, 10X live view manual focus blah blah.........

What the estimates tell you is that even if you have a heavy tripod, base iso etc you will gain only a small increase in reach, and not 60%.

edited for asking my question within the quote - sorry

Reach or resolution?

In the following context they are the same: if you can achieve 1.6x greater resolution with the crop, a subject 1.6x further away will have the same apparent resolution as for the FF and so you have 1.6x the reach.

 

[privatebydesign]

It is a great shame you chose 1.6 as your example number there Alan, it will confuse people all the more as it is the "crop factor".

You point out that the difference in resolution is between 1.19 and 1, depending on the sensors being compared, that is the point, the best you can hope for with a crop camera is between nothing and 19% more "reach", not the 60% that last comment implies.

 

[Andyx01]

I'm sure this thread is full of bad information all around, and I'm not even going to read the previous posts.
If you want to know how to convert back and forth here are the facts.

FF is 2.56X larger than APSC (1.6X crop factor - 1.6 * 1.6 = 2.56)
If you have an APSC Body with a given lens, and want to obtain the same framing, depth of field, and ISO noise on a FF use the following math:
Crop to Full Frame:
Length * 1.6
Aperature * 1.6
ISO * 2.56

i.e. a
100mm f/2.8 lens @ ISO 400 on a Crop body camera will take the same picture as
160mm f/4.48 lens @ ISO 1024 on a Full Frame body.

To convert the other way, the math is:
Length / 1.6
Aperature / 1.6
ISO / 2.56

i.e. a Full Frame with an 85mm f/1.8 lens shooting at ISO 320 =
a Crop body 53mm f/1.1 lens shooting at ISO 125

As far as cropping a Full frame image down to an APSC size, the pixels on target are FF MP divided by 2.56
i.e. 22MP / 2.56 = roughly 8.6MP on target.

If your lens has the resolving power to see the added resolution, then there ya go, use a crop body for extra reach.

 

[privatebydesign]

I'm sure this thread is full of bad information all around, and I'm not even going to read the previous posts.
If you want to know how to convert back and forth here are the facts.

FF is 2.56X larger than APSC (1.6X crop factor - 1.6 * 1.6 = 2.56)
If you have an APSC Body with a given lens, and want to obtain the same framing, depth of field, and ISO noise on a FF use the following math:
Crop to Full Frame:
Length * 1.6
Aperature * 1.6
ISO * 2.56

i.e. a
100mm f/2.8 lens @ ISO 400 on a Crop body camera will take the same picture as
160mm f/4.48 lens @ ISO 1024 on a Full Frame body.

To convert the other way, the math is:
Length / 1.6
Aperature / 1.6
ISO / 2.56

i.e. a Full Frame with an 85mm f/1.8 lens shooting at ISO 320 =
a Crop body 53mm f/1.1 lens shooting at ISO 125

As far as cropping a Full frame image down to an APSC size, the pixels on target are FF MP divided by 2.56
i.e. 22MP / 2.56 = roughly 8.6MP on target.

If your lens has the resolving power to see the added resolution, then there ya go, use a crop body for extra reach.

Well maybe you should have read some of it Andy, if you had you would have realised we are not talking about equivalence, which you have outlined reasonable well. We are talking about comparable resolution capabilities, the bit you sum up as "If your lens has the resolving power to see the added resolution", that is the bit we are talking about.

 

[AlanF]

I'm sure this thread is full of bad information all around, and I'm not even going to read the previous posts.

What can one say to that?

 

[Don Haines]

I'm sure this thread is full of bad information all around, and I'm not even going to read the previous posts.

What can one say to that?

I can say "go get an SX-50"