Say What? The 5DS, f/11 and Confusing Circles

[Canon Rumors]

I enjoy reading things that make my head hurt, it usually means I’m learning something. Roger at <a href="http://www.lensrentals.com" target="_blank">LensRentals.com</a> is always good at it, I’m now going to have to pay attention to Alex Cooke at F-Stoppers. He has written a great article about diffraction the how it pertains to the resolution of the upcoming Canon EOS 5DS and EOS 5DS R.</p>
<p>From Alex:</p>
<blockquote><p>On a 5D Mark III, the size of the Airy Disk begins to exceed the size of the circle of confusion just after f/11. This means the 5D Mark III reaches its <strong>diffraction limit</strong> at that point, the point at which diffraction begins to become visible when viewing an image at 100% at a typical viewing distance. This is different from the <strong>diffraction cutoff frequency</strong>, the point at which airy disks completely merge and no amount of stopping down will improve resolution. Think of the space between the diffraction limit and the cutoff frequency as the space of diminishing returns. On the other hand, the 5DS reaches its diffraction limit just before f/8, slightly over a full stop sooner than the 5D Mark III. This might have landscape photographers and those who rely on having a large depth of field worried.</p></blockquote>
<p><strong><a href="https://fstoppers.com/education/5ds-f11-and-confusing-circles-68177" target="_blank">Read the full story</a> | Canon EOS 5DS $3699: <a href="http://adorama.evyy.net/c/60085/51926/1036?u=http://www.adorama.com/results/canonnewfeb" target="_blank">Adorama</a> | <a href="http://www.bhphotovideo.com/c/product/1119026-REG/canon_0581c002_eos_5ds_dslr_camera.html/BI/2466/KBID/3296" target="_blank">B&H Photo</a> | <a href="http://www.amazon.com/gp/product/B00T3ERPT8/ref=as_li_tl?ie=UTF8&camp=1789&creative=390957&creativeASIN=B00T3ERPT8&linkCode=as2&tag=canorumo-20&linkId=C3LAZKJCU4IRBJUF" target="_blank">Amazon</a></strong></p>

 

[dolina]

So this will make the camera lousy for landscapes? Can you do landscapes at f/11? I normally shoot wide open unless I am panning or want "to be there".

 

[NancyP]

Naah. Those that don't have the TS-E 24 will want one. I want one.

 

[quod]

So this will make the camera lousy for landscapes? Can you do landscapes at f/11? I normally shoot wide open unless I am panning or want "to be there".

Landscapes on full-frame and APS--C are generally shot from f/8 to f/16. It looks like 5DS folks will need to work on their focus stacking technique.

 

[TeT]

So this will make the camera lousy for landscapes? Can you do landscapes at f/11? I normally shoot wide open unless I am panning or want "to be there".

Landscapes on full-frame and APS--C are generally shot from f/8 to f/16. It looks like 5DS folks will need to work on their focus stacking technique.

read the article ppl... your comments make it obvious that you did not...

 

[mb66energy]

So this will make the camera lousy for landscapes? Can you do landscapes at f/11? I normally shoot wide open unless I am panning or want "to be there".

Landscapes on full-frame and APS--C are generally shot from f/8 to f/16. It looks like 5DS folks will need to work on their focus stacking technique.

One more reason I bought two second hand 5D classic. A very limited but capable camera - if you learn to master it. Still learning.
Another reason: The lenses I own are good but might resolve 20 MPix with 5Ds(R) bodies - while resolving 11-12 MPix on the 5D classic.
Third reason: I like to shoot with two identical bodies with two lenses, e.g. 2.8 24 and 2.0 100 or 2.8 100 Macro and 5.6 400. The price of 5Ds made it 7 times more expensive and 7.2 Euro is close to a Pentax 645Z with the 120mm Macro.
Fourth reason: Decision between 5Ds and 5DsR is not easy - at least for me.
Fifth reason: Need for IS or tripod. Just the 5D classic is very challenging compared to the little bit mushy IQ of 40D and 600D which hided camera shake.

But: The 5Ds(R) are shurely great cameras for those who go to the limits - e.g. with OTUS lenses or yet to be released Canon lenses ...

 

[quod]

So this will make the camera lousy for landscapes? Can you do landscapes at f/11? I normally shoot wide open unless I am panning or want "to be there".

Landscapes on full-frame and APS--C are generally shot from f/8 to f/16. It looks like 5DS folks will need to work on their focus stacking technique.

read the article ppl... your comments make it obvious that you did not...

Perhaps you should read my response carefully, or better yet, answer it yourself. I stand by my response to the question presented.

 

[candc]

another thing to consider is that the blurring from diffraction can it be sharpened in post with good results. motion blur, oof blur, etc. is more problematic.

 

[msm]

I do not fully understand why resolution is said to be so important for landscape shooters. Having shot a A7R with 36mpixels for a year I find it quite hard to realize this resolution from corner to corner, I almost always end up with a trade off between foreground in focus, background in focus or diffraction. I usually choose the latter, a little diffraction blur sharpens up nicely in post. Lens curvature does not exactly help in this matter either.

Of course you can focus stack, but seeing as you also usually need to bracket it is more work than I bother. Stacking is also a process which can introduce ugly artefacts.

But all in all I think around 20mpix actually works great for landscape because I can usually use all this resolution while with higher resolutions I feel must often be downscaled to look good from corner to corner.

 

[bedford]

I'm having difficulties understanding the reference to the aperture without any mention of the focal length.

Based on my "gut understanding" of physics the amount of diffraction is a function of wavelength and the lenght-scale of the object at which diffraction takes place (this being also true for sound waves).

So, when discussing this phenomenon in a meaningful way, the focal length should always be specified, as the diameter of the hole through which light has to pass is equal to focal-length divided by f-number.

Could someone with a more profound background in physics shed some light on the issue? ;-)

Oliver

 

[scott_m]

I'm having difficulties understanding the reference to the aperture without any mention of the focal length.

Based on my "gut understanding" of physics the amount of diffraction is a function of wavelength and the lenght-scale of the object at which diffraction takes place (this being also true for sound waves).

So, when discussing this phenomenon in a meaningful way, the focal length should always be specified, as the diameter of the hole through which light has to pass is equal to focal-length divided by f-number.

Could someone with a more profound background in physics shed some light on the issue? ;-)

Oliver

Well I wouldn't go so far as to say "profound" but yes you are quite correct. All other things being equal (and you know what that means) a larger aperture will out-resolve a smaller one, which is why astronomers build bigger telescopes - light-gathering power is only part of the equation.

However... the Airy disk formula states resolution in terms of angular separation (useful for astronomy where everything is effectively at infinity) whereas for camera sensors it's a linear measurement (pixel spacing or line pairs per mm). The linear, rather than angular, diameter of the disk then depends on the focal length of the lens (trigonometry) so in this case, the focal length disappears from the formula and you end up back with the f-ratio.

More detail here: http://en.wikipedia.org/wiki/Airy_disk

HTH,
Scott

 

[sanj]

I do not fully understand why resolution is said to be so important for landscape shooters. Having shot a A7R with 36mpixels for a year I find it quite hard to realize this resolution from corner to corner, I almost always end up with a trade off between foreground in focus, background in focus or diffraction. I usually choose the latter, a little diffraction blur sharpens up nicely in post. Lens curvature does not exactly help in this matter either.

Of course you can focus stack, but seeing as you also usually need to bracket it is more work than I bother. Stacking is also a process which can introduce ugly artefacts.

But all in all I think around 20mpix actually works great for landscape because I can usually use all this resolution while with higher resolutions I feel must often be downscaled to look good from corner to corner.

i agree.

 

[marcel]

So this will make the camera lousy for landscapes? Can you do landscapes at f/11? I normally shoot wide open unless I am panning or want "to be there".

Landscapes on full-frame and APS--C are generally shot from f/8 to f/16. It looks like 5DS folks will need to work on their focus stacking technique.

One more reason I bought two second hand 5D classic. A very limited but capable camera - if you learn to master it. Still learning.
Another reason: The lenses I own are good but might resolve 20 MPix with 5Ds(R) bodies - while resolving 11-12 MPix on the 5D classic.
Third reason: I like to shoot with two identical bodies with two lenses, e.g. 2.8 24 and 2.0 100 or 2.8 100 Macro and 5.6 400. The price of 5Ds made it 7 times more expensive and 7.2 Euro is close to a Pentax 645Z with the 120mm Macro.
Fourth reason: Decision between 5Ds and 5DsR is not easy - at least for me.
Fifth reason: Need for IS or tripod. Just the 5D classic is very challenging compared to the little bit mushy IQ of 40D and 600D which hided camera shake.

But: The 5Ds(R) are shurely great cameras for those who go to the limits - e.g. with OTUS lenses or yet to be released Canon lenses ...

I am using my 5d classic again. I think I've learned to appreciate how good the camera is. And is 10 years old....

 

[Sporgon]

F/11 ? That's nothing. Wait until they start talking about the realities of using an f/1.2 or f/1.4 lens on a FF 50 mp sensor.

 

[LDS]

More detail here: http://en.wikipedia.org/wiki/Airy_disk

Be aware that some semplifications in the formulaes are valid only within a given acceptable error. And spatial resolution depends on the distance of the subject (almost always infinity in astronomy, but not in photography), and also the sensor size.

 

[Ivan Muller]

so what?

If you set focus of a FF 24mm lens at 2.4meters and f8, everything will be in focus from 1.17meters to infinity...

and at f5.6 everything will be in focus from 1.8m to infinity...

http://www.cambridgeincolour.com/tutorials/dof-calculator.htm

Its called Hyperfocal distance

And btw I regularly shoot at f22 etc and at A3/2 sizes even at 100% I have yet to be disappointed....and what if the image is just a touch less sharp?

 

[charlesa]

I am confused... or not.

 

[dolina]

For those who reserved what will you be using the 5DS or 5DS R for? Wildlife? Birds? Landscape? Studio?

 

[3kramd5]

So this will make the camera lousy for landscapes?

That has yet to be determined. I doubt it, but if it is lousy for landscapes, it is not because of its high resolution. Hitting the diffraction limit doesn't mean that suddenly your images go to garbage. A 50MP image shot past the diffraction limit will almost certainly outresolve a 20MP image shot wide open.

 

[quod]

So, I should spend $4K on a body so that I hit the unglorious heights provided by hyperfocal focusing and diffraction? I can do this with my 5D3 and not spend an extra dime. If I spend $4K for a 50MP sensor, I am going to squeeze every little bit I can get out of that resolution. To each his own. I aspire for more.