Optimum Lens Aperture

[altosax]

I read these comments in a Canon Professional Network article "tips" section and would welcome comments and responses:

"Any lens will be at its sharpest between two and four stops down from maximum aperture; f/5.6 to f/8 for an f/2,8 lens. Need more depth of field? Don't stop the lens down beyond f/11, check out hyperfocal distance on Google or get a Tilt and Shift (TSE) lens."

altosax

[carregwen]

"Any lens will be at its sharpest between two and four stops down from maximum aperture; f/5.6 to f/8 for an f/2,8 lens."

I think that's wrong. The progression of f-stops is 2.8, 4, 5.6, 8, 11. So if it's between two and four stops it should be f/5.6 to f/11.

[Colin Southern]

It's generally regarded as "2 to 3" stops down, so perhaps just a typo?

I have to say that I don't agree with the recommendation to not stop down below F11 either!

[MrTee]

I have to say that my canon 16 - 35 tends to hold more detail at f11, depth of field is never a real problem at 16mm.

[Colin Southern]

I have to say that my canon 16 - 35 tends to hold more detail at f11, depth of field is never a real problem at 16mm.

Probably because on a FF sensor - focusing at an object 0.77 meters away your DoF ranges from .39m to infinity and beyond

[William W]

[I read this] "Any lens will be at its sharpest between two and four stops down from maximum aperture; f/5.6 to f/8 for an f/2,8 lens. Need more depth of field? Don't stop the lens down beyond f/11, check out hyperfocal distance on Google or get a Tilt and Shift (TSE) lens." [I would appreciate comments] altosax

1. Definitive statements using words such as: “any”; “will”; “sharpest”; “don’t”; “check out on Google” are to be questioned and the source scrutinized.

2. The internet is a reasonable source of information.

3. The internet is a greater source of misinformation.

4. There is an error of Mathematics in the quote (already cited) – the range of two to four stops, beginning at F/2.8 is F/5.6 to F/11 – this shows either poor understanding or sloppy / unprofessional presentation and proof reading, or both – but it is not an hanging offense, of itself.

5. Bottom line:

My guess is the person who wrote this was attempting to be “helpful” but did not have the flying hours up in both Wordsmithery or Photography to really nail, what they wanted to say.

I think it would have been better for the author to be less “school teacher” and write something like this:

"Most lenses will be pretty sharp between two and four stops down from maximum aperture – e.g. f/5.6 to f/11 for an f/2.8 lens.

This is usually referred to the lens’s “sweet spot”. (Which is NOT a technical term - so I placed it in "inverted commas")

Generally, if you close lenses beyond that “sweet spot”, you will likely begin to get Diffraction, which might be a problem as it gets worse (the more you stop down - usually the more diffraction is noticed). It is shown as a general “softness” – a simple explanation.

So, if you want a large Depth of Field instead of only thinking “I must stop the lens down to a really small aperture”, you might consider learning about the Hyperfocal Distance to avoid Diffraction.

Hyperfocal Distance is the technical way of talking about where to focus the lens such that we can use the maximum Depth of Field which is available, for any given Focal Length, at any given Aperture for any given Camera Format (or Film Format).

For every condition, there is one distance the lens can be focused at, to get the maximum Depth of Field – the Hyperfocal Distance.

There are special lenses made by Canon which mimic some of the Camera Movements found on View Cameras and Field Cameras.

These Canon lenses have the prefix “TS-E” and are available in 17mm, 24mm, 45mm, and 90mm (17mm I don't think has been released?).

These lenses manipulate the axes of the lens Elements (bits of glass), inside the lens. With a bit of practice and lots of patience these Lens Movements can allow the final image to have a Depth of Field for any particular scene which is not possible with the normal range of SLR lenses.
One example is for Table Top Food Photography: where a 50mm lens might only be able to compass 12inches DoF - the TS-E 45 can cover about the whole 3ft table, in acceptable focus.

TS-E lenses are quite expensive (compared to a second hand Toyo) or – relatively cheap if you don’t have double dark slides and can't develop film and want the easy of slapping the lens on to your DSLR and carrying a light weight around over your shoulder.

But there is another kicker – to get the full value of the TS-E lenses (especially the 17 and the 24, and somewhat the 45 as well), you really have to pony up for a 5D; 5D MkII; or a 1 series that has the larger sensor . . . well perhaps one of those APS-H cameras would be OK), you really need to use the full Image Circle of the lens to get the maximum bang for the buck, of the Lens’s Movements.

So really using F/11 or F/16 ain't that bad . . . . if you need to”

WW

[William W]

Probably because on a FF sensor - focusing at an object 0.77 meters away your DoF ranges from .39m to infinity and beyond

that's why I buy Canon L series too, Colin - they focus beyond infinity - and I like that.

WW

[Colin Southern]

17mm I don't think has been released

It's out here now

[Dave Humphries]

Hi Will,

A couple of points might benefit from a little more qualification in your excellent reply;

Hyperfocal Distance is the technical way of talking about getting the maximum Depth of Field for any given Focal Length, at any given Aperture for any given Camera Format (or Film Format).

Using Hyperfocal Distance doesn't per se increase, or maximise, DoF, however it does allow the photographer to place the DoF where it can best serve the subject, if they know the closest and farthest subject points they wish to be sharp and can refer to tables or use a calculating engine to come up with the answer.

As we know, this often means not focusing on any particular part of the composition; foreground or background, but at some distance between (the Hyperfocal distance).

These lenses manipulate the axes of the lens Elements (bits of glass), inside the lens which, with a bit of practice and lots of patience can create extraordinary Depth of Field, not possible with the normal range of SLR lenses.

This makes them sound like magic and able to 'do' anything. It maybe over complicating things, but I'd qualify this with something like "~ can create extraordinary Depth of Field at an adjustable angle which is not parallel to the film/sensor plane. The skill in their use comes in getting this focus plane angle aligned with the subject."

I hope you don't mind me 'sticking my oar in' and if I am wrong, (anyone) please don't hesitate to put me right.

Equally, if my Wordsmithery can be improved - go for it

Now, how many "any"s, "will"s and "sharpest"s did I put in there?

Good post.

Cheers,

[KentDub]

I thought that a FF sensor with a high pixel density (such as the 5dmkII or some of the 1 series cameras) were not diffraction limited on some lenses in some cases (I ran the calculator with my lenses and in a lot of cases it was not diffraction limited). In those cases - does it still hold true to not stop the lens down further (say past f/11 or f/16)?

[Dave Humphries]

Hi Kent,

I haven't done the sums, nor do I have a feel for whether that makes sense, but as it's you Kent, I expect it does (make sense)

Therefore, I would say it doesn't hold true and you may stop down.

Hi Will,

Coming back to this with fresh eyes, the answer is now obvious to me, just add two words to what you wrote to resolve my first point above

Hyperfocal Distance is the technical way of talking about getting the maximum use of Depth of Field for any given Focal Length, at any given Aperture for any given Camera Format (or Film Format).

and 'job done'

Cheers,

[Colin Southern]

I thought that a FF sensor with a high pixel density (such as the 5dmkII or some of the 1 series cameras) were not diffraction limited on some lenses in some cases (I ran the calculator with my lenses and in a lot of cases it was not diffraction limited). In those cases - does it still hold true to not stop the lens down further (say past f/11 or f/16)?

Hi Kent,

I think that this is a good example of theory diverging from practice.

My suggestion is to just not worry about diffraction (I don't) it isn't usually a big deal. If you need to stop down then just stop down (for me the reason is usually more exposure related than DoF related). I think DoF tables should be given more weight than diffraction tables.

[William W]

A couple of points might benefit from a little more qualification in your excellent reply; I hope you don't mind me 'sticking my oar in' and if I am wrong, (anyone) please don't hesitate to put me right.

Crikey! you are kinder than my English Master in High School – he would have used the oar in a much different manner.

Yes I agree 100% on both points. I knew what I meant but the text was NOT accurate. Thanks.

Because this forum allows “clean ups” I will “adjust my dress” such that I am seen in a better light – but more importantly the text is accurate.

Thanks for the heads up, could you cast your eyes over the corrections and let me know - Thanks again

Cheers,

WW

[William W]

It's out here now

Yeah I can't afford it at the moment . . . so I am avoiding the issue of knowing anything about it

[Colin Southern]

Yeah I can't afford it at the moment . . . so I am avoiding the issue of knowing anything about it

Resistance is futile!