Depth of field vs diffraction

[Clafleur]

I am looking to increase depth of field in landscape photography for systems without access to good tilt-shift lenses.  I understand the loss of other qualities that contribute to overall image quality inherent in choosing smaller sensor over larger sensor cameras.  However, if you set that aside and concentrate solely on maximizing depth of field, it would appear that smaller sensor cameras have an inherent advantage.  However, it also appears that the smaller the sensor and the more densely packed the sensels, the wider the f-stop at which diffraction begins to degrade image sharpness, thus negating at least some of the improvement in image sharpness from the greater depth of field.  Is there any short-hand way of measuring this trade-off so that one could determine the optimum sensor size/megapixel count (for any given lens/focal length and f-stop) for maximum depth of field without degradation from diffraction?  Or is the reality that diffraction denies any significant advantage to the smaller sensors and therefore, for maximizing depth of field, in the end sensor size doesn't really matter?

[Slobodan Blagojevic]

Equivalence

[ErikKaffehr]

Hi,

Yes and now. A smaller sensor will have wider DOF. Regarding diffraction it is a bit of nonsense. Diffraction is only dependents of aperture, and affects smaller formats identically. Of course, you enlarge a smaller format more but with a larger format you must stop down more.

The good news is that diffraction responds very well to deconvolution based sharpening, because it can be approximated decently well with a gaussian.

So stop down to f/16 (or f/22 if you must) and use deconvolution based sharpening with a larger radius.

There was a recent article on http://www.onlandscape.co.uk/issues/l clearly stating that Nikon D800 (36MP) at f/22 and optimally sharpened outperformed a Sony Alpha 900 (24MP) at f/8 and also optimally sharpened (it was a pay article).

What I see in my own pictures that stopping down to f/16 is OK. On 4/3 you would get the same DoF at f/8. I would probably sharpen with the following parameters in LR:

Amount 45
Radius 1.3
Detail (100%, that kicks in deconvolution)
Masking 15%

And I would have some luminance noise reduction.

I have written three articles that touch on the issues:
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/68-effects-of-diffraction
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?showall=1
http://echophoto.dnsalias.net/ekr/index.php/photoarticles/29-handling-the-dof-trap


Best regards
Erik

I am looking to increase depth of field in landscape photography for systems without access to good tilt-shift lenses.  I understand the loss of other qualities that contribute to overall image quality inherent in choosing smaller sensor over larger sensor cameras.  However, if you set that aside and concentrate solely on maximizing depth of field, it would appear that smaller sensor cameras have an inherent advantage.  However, it also appears that the smaller the sensor and the more densely packed the sensels, the wider the f-stop at which diffraction begins to degrade image sharpness, thus negating at least some of the improvement in image sharpness from the greater depth of field.  Is there any short-hand way of measuring this trade-off so that one could determine the optimum sensor size/megapixel count (for any given lens/focal length and f-stop) for maximum depth of field without degradation from diffraction?  Or is the reality that diffraction denies any significant advantage to the smaller sensors and therefore, for maximizing depth of field, in the end sensor size doesn't really matter?

[jrsforums]

Eric, cannot seem to get to your articles.

Also, what issue of the magazine was the comparison in.

John

[arentol]

The totality of the system you are using is far more important to landscape photography than the depth of field vs diffraction question itself. In other words while you could probably figure out how much depth of field vs diffraction affects your images based on sensor size, in the end the difference would be incredibly small and other things that aren't a direct function of the sensor size itself would matter far more.

For instance, check out the thread about the Fuji X-10, a 2/3" sensor camera (so f/4 on the X-10 provides roughly the same DOF as f/16 on a full format camera):

http://www.talkphotography.co.uk/forums/showthread.php?t=365827

Look especially at the work and posts by DuncanDisorderly in the first 20 pages or so. You will see that he was able to take shots people with 5d2's didn't even bother trying. The $600 X-10 could handle those situations better than systems (camera + lens) that cost 5 or 6 times as much. Does this mean the X-10 is a better landscape camera than a 5d2 or even a D800? No. What it means is that for certain situations and/or styles of shooting the X-10 is a very capable landscape camera, while in other situations the 5d2 or D800 would be better (likely FAR better).

So you have to consider the totality of each system to determine what is acceptable/preferable for you, and when doing so sensor size should not matter as it relates to the depth of field vs diffraction issue, though it should matter in regards to other things.

[RobbieV]

+1 Erik.

Thank you for the great article and resources.

[ErikKaffehr]

Hi,

Sorry for taking time to respond.

The article is here (needs subscription) http://www.onlandscape.co.uk/2012/08/the-diffraction-limit-how-small-is-too-small/

I have checked my website several times, it seems to work. Please try again.

Best regards
Erik

Eric, cannot seem to get to your articles.

Also, what issue of the magazine was the comparison in.

John

[francois]

…
I have checked my website several times, it seems to work. Please try again.

…

Seems to work perfectly.