PDA

View Full Version : Diffraction limits for wide angle lenses?



joshdaskew
29-Apr-2010, 06:36
Hi, I am using a Chamonix 45N-1 and lenses that range from a 65mm Nikon to a 300mm Nikon. I have read in a few posts that diffraction of lenses often kicks in at around f22 or so. For most purposes, this aperture gives me a sharp image and enough depth of field to render the scene sharp from front to back (without really worrying about movements). For lenses on the wide end (I have Nikon 65mm f4 and a Schneider Super Angulon f8) can I effectively shoot more wide open (say f16 or f11) and still obtain sharp depth of field from front to back? Does diffraction start at a more wide open aperture with these lenses? Or is diffraction a common thing that is standard amongst all lenses, provided the format is consistent? I am mainly shooting architecture with these wide angle lenses.

Sorry, I am not sure if that is really phrased well. Thanks for any information, would be greatly appreciated! Best Regards Josh

Jack Dahlgren
29-Apr-2010, 07:01
The longer the focal length the smaller the optimal working aperture. So I'd be more comfortable using f/16 on a 90mm than I would on a 300mm, but "optimum" depends on more than just sharpness and depth of field. Wind or subject movement combined with lighting conditions may dictate a different aperture. And sometimes you do not want the maximum amount of sharpness.

carverlux
29-Apr-2010, 07:11
Hi, I am using a Chamonix 45N-1 and lenses that range from a 65mm Nikon to a 300mm Nikon. I have read in a few posts that diffraction of lenses often kicks in at around f22 or so. For most purposes, this aperture gives me a sharp image and enough depth of field to render the scene sharp from front to back (without really worrying about movements). For lenses on the wide end (I have Nikon 65mm f4 and a Schneider Super Angulon f8) can I effectively shoot more wide open (say f16 or f11) and still obtain sharp depth of field from front to back? Does diffraction start at a more wide open aperture with these lenses? Or is diffraction a common thing that is standard amongst all lenses, provided the format is consistent? I am mainly shooting architecture with these wide angle lenses.

Sorry, I am not sure if that is really phrased well. Thanks for any information, would be greatly appreciated! Best Regards Josh

Josh,

You may find Bob Atkins' calculator handy not only in determining the Diffraction Limit for any lens but its inextricable relationship with Depth of Field and Circle of Confusion. Here it is: http://www.bobatkins.com/photography/technical/depth_of_field_calc.html. I am sure you will find a few AHA! moments once you spend some time with it.

Enjoy!
Carver

Brian Ellis
29-Apr-2010, 07:21
Unless you're making really large prints - maybe something in the 30x40 range - diffraction isn't a major problem with prints from 4x5 negatives. I wouldn't worry about it. It's usually more important to get the depth of field you need, even if that involves f64 or smaller, than to lose needed depth of field for the sake of whatever minor effect diffraction will likely have on your print.

There's also another lens aberration the name of which escapes me that increases as the aperture becomes larger (sorry I can't remember the name but I'm sure one of the optical experts here will know). So while photographing at a larger aperture may decrease diffraction it increases this other aberration. Thus there's a trade-off in finding the optimum aperture in a given situation, which is why some people (me included) put a mm scale on their cameras, measure the distance the lens travels when focusing on the near and far, set the lens (or back, if focusing with the back) midway in between, etc. etc., a topic beyond the scope of your question.

It's easy enough to see the effects of diffraction for yourself. Just make two negatives of the same scene with one or more of your lenses, one negative at say f16 or f22 and another at say f64. Make prints from them at different sizes. See whether there's a size at which the effects of diffraction bother you. If so, that's the diffraction limit for that lens at that print size.

rdenney
29-Apr-2010, 07:43
For lenses on the wide end (I have Nikon 65mm f4 and a Schneider Super Angulon f8) can I effectively shoot more wide open (say f16 or f11) and still obtain sharp depth of field from front to back? Does diffraction start at a more wide open aperture with these lenses? Or is diffraction a common thing that is standard amongst all lenses, provided the format is consistent?

Depth of field is a function of focal length and magnification, and whether a given aperture provides enough depends on the scene at hand. DOFMaster, which assumes an 8x10 print as the standard, calculates a depth of field from a little over 6 feet to infinity when focused at about 12 feet, using a 65mm lens on 4x5. If you print to 16x20 and expect the same close scrutiny, the depth of field ranges from 25 feet to infinity when focused at 50 feet.

At f/22, the depth of field increases to about 12.5 feet to infinity when focused at 25 feet, using the tougher 16x20 print-size standard.

That should give you an idea.

Take a look at DOFMaster.com, but remember that it is based on a standard 8x10 print, which is too small and too forgiving for most large-format applications.

Regarding diffraction, it has some effect at all apertures, and the issue is whether the effect of diffraction becomes a constraint. But I would say that, in general, don't worry about it. Insufficient depth of field will cause a lot more blurriness than diffraction, even at f/45. I have certainly made images at f/45, and printed them large with good sharpness, using 90mm lenses on 4x5. So, use the largest aperture that provides the depth of field you need, as viewed with a strong loupe on the ground glass, and don't worry about diffraction.

Remember also that the movements make it possible to adjust the focus plane, which can help you get the things you want in focus without depending on depth of field to do so. Depth of field describes the distance in front of and behind the focus plane that will appear to be sharp given a stated standard of sharpness. The focus plane will be sharp no matter what, even if you tilt it using camera movements. You'll be limited by the coverage of your lenses somewhat, but it doesn't take much tilt at short focal lengths to have a big effect.

Here's an example, made with a 121mm f/8 Super Angulon:

http://www.rickdenney.com/images/ConcepcionSTwr032793-9_lores.jpg

The lens was tilted to the right very strongly to get the window opening at right into the focus plane along with the tower. The left side was outside the depth of field, but that was not a problem for me. The aperture was f/45, and the image is plenty sharp.

Rick "for whom diffraction is what it is" Denney

timparkin
29-Apr-2010, 08:05
Does diffraction start at a more wide open aperture with these lenses?

No... diffraction is a constant regardless of aperture

If you use f/22, you can get a typical enlargement of 9x for typical circles of confusion..

If you use f/64 you can get a typical enlargement of 3.2x

So, yes, diffraction still has a large effect for 4x5 ...

However, like a previous poster has mentioned, it's always a trade off. Only stuff on the focal plane can be enlarged 9x at f/22 - if something is out of focus, it's out of focus regardless of diffraction..

Tim

p.s. - great page on this very site - http://www.largeformatphotography.info/fstop.html

Lynn Jones
29-Apr-2010, 10:05
Josh,

Stop paying attention to "psuedo science" and deal with reality. Realistically, good quality lenses will start to significantly soften with an aperture of about 2mm. We don't care about focal length or anything else.

Lynn

timparkin
29-Apr-2010, 11:27
Josh,

Stop paying attention to "psuedo science" and deal with reality. Realistically, good quality lenses will start to significantly soften with an aperture of about 2mm. We don't care about focal length or anything else.

Lynn

So for a Canon G1 at a focal length of 6.1mm, f/6 s on the edge of diffraction but on a Schneider 1000XL, f/500 is well within diffraction limits...

You'll have to excuse me if I don't agree with this..

Emmanuel BIGLER
29-Apr-2010, 11:38
For lenses on the wide end (I have Nikon 65mm f4 and a Schneider Super Angulon f8) can I effectively shoot more wide open (say f16 or f11) and still obtain sharp depth of field from front to back?

Hello !
Each of your excellent wide-angle lenses probably has a best f-stop as recommended by the manufacturer.
For the Schneider Super Angulon f-8, I assume a 75 or 90mm, the best f-stop is probably 22.
If you shoot at f/11 or f/16, you may get a slightly sharper image at the centre, du to slightly less difraction, but this will be very hard to notice, you'll loose some sharpness an probably some light in the corners. So it is a trade-off between coverage and sharpness at the centre, you have to somewhat sacrifice the centre in order to get the full coverage (about 100 for the f/8 "old" SA series).
I have seen pictures taken with a 90mm f/8 super angulon on 6x9, at f/11 those images are as sharp as you might dream of, but cropped by the 6x9 (56x82 mm) format you do not use the corners.

Regarding shorter focal lengths like the 65, it is a general rule that the recommended f-stop is wider, e.g. 11 or 16 when focal lenghts are shorter. However for their 65mm grandagon N Rodenstock still recommend 16-22 ; check for the 65mm Nikon, for example here the general spreadsheet quotes the image circle @f/16, which is probably the best f-stop.
http://www.largeformatphotography.info/lenses/LF4x5in.html
For the shorter Rodentock apo-grandagon series (35, 45, 55) the recommended f-stop is 8-11, not 22.

Chauncey Walden
29-Apr-2010, 12:04
Lynn's saying that diffraction becomes serious at a 2mm aperture goes along with something I heard years ago called the "Rule of 4" (which is just a rule of thumb, of course.) That is you will be happiest if you limit your minimum f/stop to the focal length divided by 4, which also goes along with what Emmanuel was saying about the shorter focal lengths. So, 4mm not too bad, 2mm bad at least as far as large format lenses are concerned.

timparkin
29-Apr-2010, 12:08
Lynn's saying that diffraction becomes serious at a 2mm aperture goes along with something I heard years ago called the "Rule of 4" (which is just a rule of thumb, of course.) That is you will be happiest if you limit your minimum f/stop to the focal length divided by 4, which also goes along with what Emmanuel was saying about the shorter focal lengths. So, 4mm not too bad, 2mm bad at least as far as large format lenses are concerned.

That means f/128 would be OK for my Nikkor T*ED 500?

Not being argumentative - just slightly dissapointed when people 'poo poo' science and then suggest blatantly incorrect rules of thumb.. (and I don't mean you - the rule works from 90 to 210 - sort of... )

Tim

ic-racer
29-Apr-2010, 12:33
My rule of thumb is to use the same F stop Number across all focal lengths for a format. So in 35mm f8 for the wide angles, normals and teles.
4x5 F22 for all and 8x10 f45 for all.

You can do all the math but one way to understand it is that although your aperture is physically smaller at the same F stop Number with the WA lens, the WA lens minimizes diffraction effects by making the circles of confusion smaller. Similar with a long lens; although the physical aperture is bigger, the diffraction effects are magnified.

The above is for comparisons within a format. For cross-format comparisons, you need to define your minimum circle of confusion size, viewing distance, focal spread etc. to get your aperture. The ones I listed above are just 'usual case' settings for me.

rdenney
29-Apr-2010, 13:12
I see a lot of worshiping at the altar of sharpness, but maybe I'm missing the point.

To me, the aperture has a huge impact on the gross look of an image. If I want selective focus, I use a wide aperture. If I want lots of depth of field, I stop down as needed to achieve it. The difference between those extremes is absolutely profound, altering the look of the image completely. That's what I mean by gross, in the German sense of large.

Other gross effects on the look of an image include how the subject moves or doesn't move during the exposure. Often, the gross look I desire in terms of movement is in conflict with the gross movement I desire in terms of aperture. I may need f/45 to get the depth of field, but the resulting 20-second exposure might allow the slightest breeze to ruin my image. Some days you just can't win, and some visualizations are just not possible with large format. But subject movement is a gross effect that profoundly affects the look of an image.

Vignetting is a gross effect, though it is sometimes correctable and sometimes it just doesn't matter that much. Assuming I've met my needs for the above, I might stop down to minimize vignetting.

Lens aberrations and diffraction are a fine effect, it seems to me. I live with (and maybe even desire) lens aberrations when using selective focus, and I live with diffraction if necessary to achieve the necessary depth of field. We can say when diffraction will be significant, but it is only significant if we know beforehand what standard we must meet. Most of the time, my standard is "the best I can get and still achieve the gross effects important to my visualization". Diffraction effects smoothly increase as we stop down. As with the edges of the depth of field, it's not like we take the next little incremental step and suddenly the red neon DIFFRACTION HERE! light comes on. It's just that we cross a standard that might have been arbitrary and based on an assumed print size and standard of sharpness that, if necessary, we could usually compromise slightly.

But the time we spend pondering both make it seem like we must believe lens aberrations and diffraction are gross effects, and depth of field and background blur are fine effects.

My f/22 images are sharper than my f/11 or f/45 images with short lenses on 4x5. But I will stop down to get the depth of field I need, or open up to allow a short enough shutter speed or to get narrow enough depth of field when that is what my visualization demands, because those effects can be seen in a snapshot-sized print. It takes a really big print from 4x5 to detect a diffraction problem, and then it's likely only photographers who'll notice it.

Rick "rant over" Denney

timparkin
29-Apr-2010, 13:21
I see a lot of worshiping at the altar of sharpness, but maybe I'm missing the point.

You might be ;-) the original person asked..

can I effectively shoot more wide open (say f16 or f11) and still obtain sharp depth of field from front to back?
yes (I think most people agree on this)

Does diffraction start at a more wide open aperture with these lenses?
No (again, I think this is a basic "law of physics" sort of thing)

is diffraction a common thing that is standard amongst all lenses, provided the format is consistent?
Yes (again - that damned "physics" again.. )

Matus Kalisky
29-Apr-2010, 13:27
Purely from the physics point of view the diffraction effect (or the size of the circle of confusion) dos not depend on the focal length, because we are used to use aperture already normalized to focal length (f/8, f/22, ..).

Kirk Gittings
29-Apr-2010, 13:30
Diffraction limit is not a limit of any kind-it is not like hitting a brick wall. It builds up barely noticeable at small apertures and then increases as you stop down more. It is easiest to see with a DSLR making equivilent exposures as you stop down. But you have to look closely because it is not blatant at first. I don't print large from 4x5 and frankly never worry about it. And I don't hesitate to shoot at f45. Especially now that I scan allot of film-in general I am a minimalist sharpener-but a slight amount of extra sharpening makes up for a slight amount of diffraction.

I remember giving a talk once with Mark Citret. All through his talk he referred to using f45 for many shots and I had mentioned it on a couple of images too. After the talk someone came up and said to us "I thought because of diffraction limits that you couldn't shoot at that small of an aperture? Mark and I scratched our heads. Where do people get this idea? This was maybe 6 years ago and he said he was increasingly hearing this at his public talks.

rdenney
29-Apr-2010, 13:33
OP: can I effectively shoot more wide open (say f16 or f11) and still obtain sharp depth of field from front to back?
yes (I think most people agree on this)

Well, it depends on the definition of front and back. The image I showed as an example barely had sufficient depth of field at f/45.

But the questions suggested someone still on the steepish part of the learning curve, and the adult educator in me demands to address underlying issues as well as stated ones. The question above suggests to me that the OP was worried that diffraction was a bigger effect than depth of field. In just about any measure that is important, it isn't.

Rick "who did advise to use the largest aperture that met other objectives" Denney

Matus Kalisky
29-Apr-2010, 14:17
I have experienced the effect of the diffraction directly only once. I did a close up image of a sun flower which required exposure compensation of 1 1/3 of stop. I did the image twice with affective apertures of f16 2/3 and f/45 1/3. While the second one had definitely deeper depth of field, the first one was definitely sharped - very obvious with a 4x loupe so I guess it would be visible in 8x10" print.

Just for a note - both shots were taken "in studio" and the camera position did not change. The lens was Fujinon 125/5.6 CM-W

I may scan them for comparison.

Mike Anderson
29-Apr-2010, 14:31
....
I remember giving a talk once with Mark Citret. All through his talk he referred to using f45 for many shots and I had mentioned it on a couple of images too. After the talk someone came up and said to us "I thought because of diffraction limits that you couldn't shoot at that small of an aperture? Mark and I scratched our heads. Where do people get this idea? ...

From the internet, where else?

http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml

On the 180mm f/5.6 Rodenstock APO Sironar HR: "I regard f/32 and f/45 as unusable."

...Mike

Ed Richards
29-Apr-2010, 15:07
> On the 180mm f/5.6 Rodenstock APO Sironar HR: "I regard f/32 and f/45 as unusable."

Be fair:-) He is talking about using that lens on an MF digital back. With the level of enlargement that requires, there really is alot of diffusion.

Mike Anderson
29-Apr-2010, 15:48
> On the 180mm f/5.6 Rodenstock APO Sironar HR: "I regard f/32 and f/45 as unusable."

Be fair:-) He is talking about using that lens on an MF digital back. With the level of enlargement that requires, there really is alot of diffusion.

Well that is a poor choice of words ("...unusable..."), and if you're new to photography and trying to learn this stuff you could easily come away misinformed. And I think there's lots of this stuff - stuff that exaggerates the evils of diffraction - floating around. rdenney's post above (http://www.largeformatphotography.info/forum/showpost.php?p=584685&postcount=13) is one of the few things (only thing?) I've read on the internet that puts diffraction in perspective.

...Mike

Mike Anderson
29-Apr-2010, 15:52
I see a lot of worshiping at the altar of sharpness, but maybe I'm missing the point.

...

Rick "rant over" Denney

Rant? What rant? That's a very nice statement. It should be preserved, like as an article on your website.

...Mike

joshdaskew
29-Apr-2010, 17:34
Thank-you so much to everyone who contributed! Greatly appreciated!! Will need a bit of time to digest all those answers and then apply it to my own work.. I was looking for a rather rough guide to diifraction with these wider lenses (and my Super Angulon is a 90mm f8) and have come away with more than that. To Rdenney, yes, I did like your answer and definitely don't get too bogged down in the "rules" of sharpness. I generally shoot wide open on my Rodenstock Sironar - N 150mm for portraits and have recently been experimenting shooting (close to) wide open on my Schneider Xenotar 150mm 2.8, with full knowlege that this is not the sharpest point of the lens. Like you say though, there are a whole bunch of visual factors involved when making these decisions. Having said that though, I think it is important to know the basic parameters and then decide which way to go yourself. Thanks again for all the input. Best regards Josh

Ed Richards
29-Apr-2010, 19:06
That other aberration is spherical aberration, which increases when the lens is wide open. It is not a big issue with modern LF lenses, and is much prized in old lenses by some people we know.:-)

One thing I have noticed with wide, and really wide lenses, is that they can look less sharp and more prone to diffraction. This happens when you look at fine detail like twigs and grass, and forget that what you are seeing on the negative is much smaller than with your normal lens. So you can see visible softening in these superfine details, forgetting that they only look soft because they are smaller than what you think you are comparing them to in your mind. This might only be me, but it took me a while to sort out when I was testing a 47mm XL recently.

Nathan Potter
29-Apr-2010, 20:13
rdenny has folded two of the most significant variables into his original comment - diffraction and depth of field - the two most alluded to in the original thread.

To differentiate a bit between the two is important for clarity. A diffraction limit refers to how finely a point can be focused at the plane of best focus. That point (or diameter) is called an Airy disc and is a function only of the wavelength of light and the f/no. of the lens. The relation is simple, D = 2.44(lamda)N where lamda is wavelength of light and N is the f/no. Secondarily the focal length of the lens enters in, since the f/no = FL/aperture diameter.

Now on either side of the sharpest focal point light rays diverge and the Airy disc becomes fuzzy. The further away you go from the point of best focus the fuzzier the point (Airy disc) becomes. This increasing diameter, fuzzy spot, is referred to as the COC (Circle of Confusion). At any given distance from perfect focus the diameter of the COC is reduced by decreasing the size of the aperture. This effect is what yields an increase in the depth of focus. This is the effect which rdenny refers to as probably being the more dominant cause of apparent loss of sharpness in many image taking situations. I would tend to agree in most situations.

Now another effect alluded to has to do with the degree of optical lens correction or image forming perfection due to the glass and the lens design. Many lenses used at maximum aperture stress the image quality by using the peripheral parts of the glass. As one reduces the aperture diameter the off axis rays are eliminated thus improving the optical image quality. Most modern, high quality lenses are so highly corrected that only small improvements may be realized in reducing the aperture.

So the diffraction limit is least with the lens wide open; the optical correction is best as the lens is stopped down and the depth of field (DOF) is greatest with the lens stopped down. Diffraction competes with DOF and optical correction. The best overall compromise for image quality is often found in the range of f/16 to f/32 for large format lenses.

Nate Potter, Austin TX.