Small apertures, diffraction limits, and DOF

Started May 1, 2008 | Discussions thread
freddyNZ Senior Member • Posts: 2,611
Re: Small apertures, diffraction limits, and DOF

Jay Levin wrote:

freddyNZ wrote:

Well, that goes to show that professional photographers can either
make mistakes, or don't understand the physics behind photography
very well (unless you have misinterpreted what was written).

And then there is a problem if people believe them. That's plain wrong.

The concept of diffraction in photography was popularized by

Sean T. McHugh, Professor of Chemical Engineering at Cambridge
University. He discusses the physics of diffraction in this article:

However, the practical significance of his article appears at the end
in a section entitled, Notes on Real-World Use in Photography.
Specifically, he states:

"Even when a camera system is near or just past its diffraction
limit, other factors such as focus accuracy, motion blur and
imperfect lenses are likely to be more significant. Softening due to
diffraction only becomes a limiting factor for total sharpness when
using a sturdy tripod, mirror lock-up and a very high quality lens."

I don't agree with the last part of that statement - as it has been presented.

Stopping down to the point where diffraction effect is clearly visible is easy "hand-held" in good light at base iso on a dslr. It might have been far more problematical with slow film, and especially with larger formats used for landscape shooting. If you check MTF charts for lenses tested on most systems, resolution starts falling slightly from as wide as F4 or typically f5.6 when tested on 10mp APS-c bodies. Using f11 or smaller, the effects do start to become clearly visible - IMO "significant" for large prints. The small amount of contrast loss isn't normally too much of a problem as it can be mainly recovered in post-processing. But when detail is lost (and fine detail will certainly be lost at f16 or smaller on an 10mp APS-c system) then it can't be recovered.

I expect that the site references parameters for assessing when a system is "diffraction limited", such as print size, viewing distance, eyesight etc etc.

I suggest that you try some test shots yourself to see what degree of resolution loss you want to put up with for landscape shooting - taking that into account, rather than selectively interpreting advice from others - even if you consider them "experts".

In reality, mirror lock-up is generally inappropriate for shooting
landscapes. According to Bob Atkins:

"In general, for lenses of 300mm and over (and macro work), users
should try to avoid using shutter speeds between 1/30 and 1/4 second
for optimal sharpness. For "normal" work with lenses of 100mm or
less, sharpness loss due to mirror induced vibrations does not seem
to be an issue."

Bob Atkins should have added to the above:

"But when using long lenses and for macro work, it's just amazing how often the needed exposure falls bang-smack in the 1/4 - 1/30th second range. Life's a beach. Lack of MLU in a camera (or shutter delay mode) can be a real problem.

People generally do not use long lenses and mirror lock-up for
landscapes. This only reinforces the point made by Rob Sheppard
about the irrelevance of diffraction for landscape work.

Much landscape shooting is done in less than ideal light, and a sturdy tripod, a camera with MLU can be quite essential. Long lenses are used for landscape.

Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow