Small Apertures & Diffraction. Yeah, Whatever...

[therickman]

Something just dawned on me.

As I'm looking through my various photography magazines, admiring the gorgeous landscape photos, I noticed almost every one of them has been shot at apertures between f/16 and f/22. They are completely sharp, beautifully composed, oh... and all shot by professionals.

Yet on these forums, everybody whines and complains about diffraction at small apertures. "Your photos won't be sharp at f/22!" "Use a wider aperture or it will be blurry!" Blah blah blah. Apparently, the pros in these magazines don't agree. After all, they're out actually taking photos (and getting paid for it) instead of pixel peeping and coming here to whine about it.

--
Insert obligatory quote here...

 

[Bobo Hodls]

Are you sure the format of the images in question was 35mm?

--
...Bob, NYC

'Well, sometimes the magic works. Sometimes, it doesn't.' - Little Big Man

Galleries: http://www.bobtullis.com

 

[John K]

...because I shoot macro. IMHO being able to view an image on a computer screen is causing people to worry about minor details -things that they wouldn't see if they were just looking at prints. The images that you admire are good because they are well composed, and not because they are razor sharp. Judging the quality of a photo by sharpness alone is a mistake -there's so much more to a good photo than the fine details.

I routinely take macro photos at Fstops that people swear won't work because of the "diffraction monster". The best advice I can give you is to put your time and effort into learning composition and let the measurebaters take all the poorly composed razor sharp images that they want...

F11, ISO 100, and 1x:



F11, ISO 100, 2x:



F11, ISO 100, 3x:



F11, ISO 100, 4x:



Look at all that horrible diffraction!

--
Also known as Dalantech
My Blog: http://www.nocroppingzone.com
My gallery: http://www.johnkimbler.com
Macro Tutorials: http://dalantech.deviantart.com/gallery/#Tutorials

Always minimal post processing and no cropping -unless you count the viewfinder...

 

[mbaginy]

Whenever I need to stop down to f11, f16, f22 or more for DOF reasons, I do so. I haven't been able to notice image quality problems in prints or viewing the images "normally".

Only when I begin to pixel peep, do I notice diffraction. I try to shy away from stopping down "too much" and keep in the optimal range for most shots. But then, I also shoot wide open to get the shallow DOF when the images requires it. That's also not ideal (from a sharpness viewpoint).

Format also plays an important role. Think of those large format photographers who stopped down to f64. Following the diffraction facts, those shots should have been visual disasters!

Bottom line for me: use the aperture necessary for the image. It's good to know the details behind settings but I won't be pixel peeping with my A4 or A3 size prints anway.
--
Mike

 

[Chris59]

Chances are these photos were shot on medium or large format cameras. The blurring effects of diffraction are the same at f16 or f22 regardless of format which means that the smaller format cameras like 35mm or APSC (film OR digital) have the same size circle of confusion as their larger cousins. However, this translates to a larger percentage of the whole image which essentially means that an image blown up to the same size print at f22 will, all things being equal, means a less sharp and detailed image from the 35mm camera compared to, say, a 4" by 5" view camera.

 

[Karma Traveler]

Something just dawned on me.

As I'm looking through my various photography magazines, admiring the
gorgeous landscape photos, I noticed almost every one of them has
been shot at apertures between f/16 and f/22. They are completely
sharp, beautifully composed, oh... and all shot by professionals.

Yet on these forums, everybody whines and complains about diffraction
at small apertures. "Your photos won't be sharp at f/22!" "Use a
wider aperture or it will be blurry!" Blah blah blah. Apparently, the
pros in these magazines don't agree. After all, they're out actually
taking photos (and getting paid for it) instead of pixel peeping and
coming here to whine about it.

The effects of diffraction at f/16 and f/22 will not be readily visible in magazine-sized prints. But, hey, don't let a little thing called "logic" stop you front ranting.

 

[jlynch]

If you look closely you will find that many of landscape photographs shown in magazines like Outdoor Photography are shot using 35mm film. The diffraction limitations apply to DSLRs.

--
Jeff Lynch
Serious Amateur Photography
http://jefflynchdev.wordpress.com

 

[Blake Cook]

Chances are these photos were shot on medium or large format cameras.
The blurring effects of diffraction are the same at f16 or f22
regardless of format which means that the smaller format cameras like

I thought diffraction is a result of the physical properties of light related to the actual aperture size/diameter, not f/stop. The larger the sensor the larger the actual aperture diameter for any given f/stop. In other words, f/22 on a 35mm sensor is a much smaller physical diameter translating to more diffraction than the same f/stop on a 4x5.

As and example, small sensor cameras like my Panasonic FZ18 begin to show diffraction at f/4 which has a physical diameter about the size of f/26 on 35mm DSLR.
--
Blake in Vancouver
http://flickr.com/photos/28305360@N00/

Panasonic Stuff, Canon Stuff. Mac Stuff & annoying PC & Windows Immobile PDA POS.

 

[therickman]

Are you sure the format of the images in question was 35mm?

They're all crop and full frame sensor digital cameras. For instance, in the January 2009 issue of "Photography Monthly," there are beautiful landscapes shot with the 5D and 1DIIN, all are at f/16-22.

The effects of diffraction at f/16 and f/22 will not be readily visible in magazine-sized prints. But, hey, don't let a little thing called "logic" stop you front ranting.

I'm sure the photographers planned for these images to be printed at sizes much larger than a common magazine, yet they still used very small aps. But, I guess that wasn't a "logical" decision on their part. Heck, they're only getting paid for their images of beautiful, exotic travel locations.

--
Insert obligatory quote here...

 

[Ioannis Doukakis]

I have run some tests on diffraction using an EF 100mm Macro f/2.8 lens.

Yes, after a specific aperture, sth between f/11 and f/16 depending on sensor size and lens, we start to get diffraction.

But I NEVER saw ANY diffraction problems exceeding one pixel. So in the focused areas you just have each pixel "bleed" its colors into the neighbouring ones. You HAVE to have a very contrasting scene to see this.

On the other hand, if you see professionals' images, most of them are heavily Photoshop processed. Any diffraction issue can easily be diminished using an unsharp-mask tool, setting a sub-pixel radius. But if you have something totally out of focus, you cannot recover it using today's sharpening tools.

 

[Ioannis Doukakis]

Diffraction is property of light, has to do with the size of the pixel (minimum detal) compared to the size of the aperture. It has nothing to do with the medium that we record on, whether it is film or CMOS sensor.

The smaller the pixel size on sensor, the earlier you will see diffraction problems.

 

[Bobo Hodls]

I asked, because that had confused me too in the past. Most times when I'd see F22 apertures or so, they turned out to be MF or LF captures.

Diffraction exists, but we have a lot of leeway in manipulating the pixels that matter. Where extreme fidelity is crucial, one will take into account diffraction. But often it's not the gorrilla in the room that some make it out to be. Too, one's discernment of fidelity at varying output sizes plays into this.

At 8x10 print sizes, or equivalent viewing on screen or in publications, one very well can put diffraction considerations aside. OTOH, if one of those images turned out to be something that called for a much larger presentation, not having taken diffraction into consideration may be found to be regrettable. (I coulda been a condender! Aaaaargh!).

--
...Bob, NYC

'Well, sometimes the magic works. Sometimes, it doesn't.' - Little Big Man

Galleries: http://www.bobtullis.com

 

[MM1]

Are you sure the format of the images in question was 35mm?

They're all crop and full frame sensor digital cameras. For instance,
in the January 2009 issue of "Photography Monthly," there are
beautiful landscapes shot with the 5D and 1DIIN, all are at f/16-22.

So there you have it - the diffraction effect depends also the size of the pixel. As far as I know, it's perfectly safe to use f/16 on 5D, whereas you get the diffraction blur on 50D at f/16.

--
Cheers,
Martin

 

[Ben_Egbert]

I tend to agree diffraction is way overblown. I use f16 all the time, and hardly ever see the diffraction effect, but I sure can see a poor dof or soft corners if I try to shoot at wider apertures.

On the other hand, there are diminishing returns above f16. Try it yourself, do a hyper focal calculation for a scene at 35mm f16 and then see how little you gain at f22.

Most of my landscapes are at f11, f13 and f16 depending on focal length. I stop down my 17-40 more than required for DOF because the vignetting is not gone until f11, and the corners need stopping down long after DOF is sufficient.

For a 70mm landscapes, f22 may have some value, but I typically don't use 70mm when I have near subjects.

--
http://www.pbase.com/roserus/root

Ben

 

[black0]

Nice shots, but I hardly think f11 and images scaled for the web is what we are talking about here... I'm not saying you are wrong, but this would have been far more convincing if these were all full crops at f22. If you are going to convince the measurbators, you have to play by their rules.

Diffraction does happen - that is just physics. It is worth knowing about, so you can make informed decisions. But it is just that - a decision, and the only way it can really be performed is to get out there and try stuff, see what it looks like, and decide if you personally are happy with the results. Going back to the original parent post, I don't put a whole lot more stock in the settings used by "professionals" than I do by the advice of folks on forums. it may suggest things to try that I haven't thought of before, but really, doesn't it still come down to what you produce with your eye and your camera?

 

[Yehuda Katz]

of a problem.
Yehuda

I asked, because that had confused me too in the past. Most times
when I'd see F22 apertures or so, they turned out to be MF or LF
captures.

Diffraction exists, but we have a lot of leeway in manipulating the
pixels that matter. Where extreme fidelity is crucial, one will
take into account diffraction. But often it's not the gorrilla in
the room that some make it out to be. Too, one's discernment of
fidelity at varying output sizes plays into this.

At 8x10 print sizes, or equivalent viewing on screen or in
publications, one very well can put diffraction considerations aside.
OTOH, if one of those images turned out to be something that called
for a much larger presentation, not having taken diffraction into
consideration may be found to be regrettable. (I coulda been a
condender! Aaaaargh!).

--
...Bob, NYC

'Well, sometimes the magic works. Sometimes, it doesn't.' - Little
Big Man

Galleries: http://www.bobtullis.com

 

[jlynch]

Yes, as a degreed engineer I am well aware that diffraction is an optical phenomenon but the diffraction limited aperture on a DSLR is not. There is a fairly good tutorial on this here.

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

--
Jeff Lynch
Serious Amateur Photography
http://jefflynchdev.wordpress.com

 

[Hugowolf]

The diffraction effects don’t vary from system to system, but some camera systems report the effective aperture and some only report the nominal aperture. For example, to focus to 1:1, macro lenses generally change the effective focal length of the lens and this affects the relative aperture. On Nikon systems the effective aperture is reported in the viewfinder, on Canon cameras the nominal value is reported.

The Nikon 105 mm f/2.8 VR macro lens has a maximum aperture of f/2.8 and a minimum aperture of f/32 nominally, but at 1:1 these change to f/5 – f/57 and it is these values that are reported in the viewfinder and stored in the EXIF data.

This isn’t much of a problem at normal shooting distances, but close up it can make a considerable difference. The aperture that is reported on Canon systems can be deceptive and can foster some confusion (and many arguments). A nominal f/16 at 1x magnification will show noticeable diffraction effects, whereas f/16 at only 1:4, the diffraction effects will be hardly noticeable.

The effects of diffraction are clearly visible with the 100 mm f/2.8 and 25 mm of extension at closest focusing distance, even at f/11 (with FF sensor).

Brian A.

 

[Bobo Hodls]

of a problem.
Yehuda

Last night I was giving a introductory run to two different papers. The default settings shows the more exotic paper to render the fine details much better at a 6" distance - I mean Mr. Magoo could see the difference. When viewed from the more proper arm's length, the differences start to vanish (still, the more exotic paper shows an intangible edge, but only side by side). While the softer print is quite acceptable for framing (it will be the default stock), which paper do you think would be chosen for a gallery showing?

Because of the greater ability to close-scrutinize developing details that is is afforded to the masses today through digital (not just pro's with a good darkroom setup and print lab relationship), the bar is raised. Take two astounding natural light compositions, the one with the finer resolved detail will be noticed by those who have a developed discernment. It's up to the photographer to decide the balance for his intended audience, giving proportional weight to the balance of composition, subject presentation, and fidelity of output. It's possible to have all to a degree, but if one suggests all is required in every subject of pursuit, that would take us to a different topic, for which there is no answer.

Again, diffraction isn't a deal killer, but like choosing a shutter speed or aperture, it's a consideration to take into account according to the image's purpose or intent.

--
...Bob, NYC

'Well, sometimes the magic works. Sometimes, it doesn't.' - Little Big Man

Galleries: http://www.bobtullis.com

 

[Yehuda Katz]

you want to blur something in motion), when dealing with macros you have to make a decision, more diffraction or more DOF. I'll always go with more DOF.

Take care,
Yehuda

of a problem.
Yehuda

Last night I was giving a introductory run to two different papers.
The default settings shows the more exotic paper to render the fine
details much better at a 6" distance - I mean Mr. Magoo could see the
difference. When viewed from the more proper arm's length, the
differences start to vanish (still, the more exotic paper shows an
intangible edge, but only side by side). While the softer print is
quite acceptable for framing (it will be the default stock), which
paper do you think would be chosen for a gallery showing?

Because of the greater ability to close-scrutinize developing details
that is is afforded to the masses today through digital (not just
pro's with a good darkroom setup and print lab relationship), the bar
is raised. Take two astounding natural light compositions, the one
with the finer resolved detail will be noticed by those who have a
developed discernment. It's up to the photographer to decide the
balance for his intended audience, giving proportional weight to the
balance of composition, subject presentation, and fidelity of output.
It's possible to have all to a degree, but if one suggests all is
required in every subject of pursuit, that would take us to a
different topic, for which there is no answer.

Again, diffraction isn't a deal killer, but like choosing a shutter
speed or aperture, it's a consideration to take into account
according to the image's purpose or intent.

--
...Bob, NYC

'Well, sometimes the magic works. Sometimes, it doesn't.' - Little
Big Man

Galleries: http://www.bobtullis.com