[stevebrot]

Would be nice to see some funky curves for DOF vs. FL vs. format to get an idea of it.

I should not be too difficult to do with Excel or other spreadsheet software if someone had the time . The equations are in the Wikipedia.

Steve

[Frank Fletcher]

Another intuitive way at looking at it (though perhaps not technically correct) is to use the DoF scales that correspond to apertures 1.3 - 1.5 stops larger than the actual.

My personal taste btw is to use 0.018/0.019mm CoCs for the Pentax APS-C system and 0.025ish for the35mm film. I like to err on the side of sharper!

[ChrisN]

I should not be too difficult to do with Excel or other spreadsheet software if someone had the time . The equations are in the Wikipedia.

Steve

Here you go!

[ChrisN]

And a follow-up question or two. For those who use hyperfocal focusing for their landscape photography, how do you set the lens to the hyperfocal distance? It's pretty hard to read, for example, 3.03m on the distance scale on a 31LTD lens. I'm thinking about a bit of string that length with a little flag on the end. Tie one end to the tripod head and stick the flag in the ground to focus on.

Does the camera software get an accurate reading of focus distance? Would it be possible for Pentax to build in a function that automatically sets the focus distance of the lens to the hyperfocal distance for the selected aperture?

[georgweb]

Chris,
thanks for posting the spreadsheet! I have read some more into DOF and try to answer in reverse question order.

Regarding the camera knowing what, I've looked up the EXIF (on a landscape shot in AV mode) from my 6mpix K100D with the 18-55 lens through exiftool(-k).exe, and at the end it says:

".........
Aperture : 9.5
Circle Of Confusion : 0.020 mm
Field Of View : 60.3 deg
Focal Length : 21.0 mm (35 mm equivalent: 31.0 mm)
Hyperfocal Distance: 2.28 m
Light Value : 13.5
....."

Obviously the camera knows something, but is it worthwhile to let it focus right to the edge of the hyperfocal distance?

I'd say no, cause many good people here and elsewhere have pointed out that the FOV is not an even zone of focus engulfed by non-focus zones with a more or less abrupt border zone (my original understanding also).

Rather, there is only one focal plane (also referred to as focal point) which is in focus and the rest is just acceptable sharp or acceptable out-of-focus to be still regarded as sharp, gradually getting more and more out-of-focus from the focal point onwards (or backwards).

In the end it depends on the defenition of OOF or not OOF from the magnification factor that you apply to the original image and the viewing distance. Here we fall victim to large-sreen 100% pixel peeping, which in doubt will be demanded to be sharp

About the funky curves and the 'DOF-shrink' factor for recompiling old manual lens scales to use with APS-C sensors, Bob Atkins (scroll right down to the end) got it allright. The funky curve is rather linear and flat, so a crude factor seems enough to start with, he proposes to cut 1.3 stops off at each end, which is in line with what people have stated elsewere.

Hope this helps a little,
Georg (the other)

[Big Dave]

Steve,
This is true, but when you use the lenses on a APS_C sensor the DOF is greater for a given field of veiw. I find this helpful for general shooting, but a slight draw for portraits. A larger appeture comes in handy in that case.

Dave

35mm film, 43mm lens, f/8, 10' = 9.2'
APS-C, 31mm lens, f/8, 8.64' = 9.22'
APS-C, 35mm lens, f/8, 10' = 9.3'

APS-C, 43mm lens, f/8, 10' = 5.58'
APS-C, 31mm lens, f/8, 10' = 13.4'
35mm film, 31mm lens, f/8, 10' = 34.5'

Online Depth of Field Calculator

Steve

[Maffer]

I have used hyperfocal focusing with the 18-55mm kit lens. But that is major pain in the ass. The distance scale on the lens only applies to the tele-end of the focal range so 18mm which I use for landscape shots usually had to be tested again and again with countless focusing attempts.

I used a DoF calculator (Online Depth of Field Calculator) and then set objects at given hyperfocal distances, used either AF or MF to get the object in focus and then shot series of pics with a scene where there is one object at half of the HD and something far far away (~infinity) and tried to get those rendered with acceptable sharpness. With carefull adjusting of the focus you can get the values you should set the focus on the distance scale for a given aperture.

Using the calculator mentioned above:

F/16, 18mm, K200D CoC says that HD is 1.03m and DoF extends from 0.515m to infinity. I placed object 1.03m away from camera and used AF to get the focus. Focus was halfway between 1 and 2m reading on the distance scale. An object placed approx. 0.5m away from the camera will now be acceptably sharp. Also the targets far away will also be sharp enough. Setting infinity focus will render far away targets okay but that object at 0.5m distance will be much more blurry.

I am fairly new here Big Dave. Thanks for the very useful thread topic.

[stevebrot]

Steve,
This is true, but when you use the lenses on a APS_C sensor the DOF is greater for a given field of veiw. I find this helpful for general shooting, but a slight draw for portraits. A larger appeture comes in handy in that case.

Dave

Ackkk!

This is making my brain hurt, but you are correct.

30mm APS-C, f/8, 10' = 15.1'
45mm 35mm film, f/8, 10' = 8.15'

APS-C wins...

• Same distance, same aperture, equivalent FOV, APS-C has more DOF.
• Same distance, same aperture, same reproduction ratio for the viewed image, APS-C has more DOF.

35mm wins if

• Same distance, same aperture, same reproduction ratio at the film plane, APS-C has less DOF
• Same distance, same aperture, same focal length, APS-C has less DOF.

Edit: The statement below is wrong. For a given lens designed for 35mm, the DOF scale will be more restrictive on APS-C...same reproduction ratio at the film plane...
[Since final image size is what counts, the lens scales should be a conservative estimate of DOF when used on an APS-C camera. There, I said it,and it feels good to get it out.]

Steve

[Frank Fletcher]

Since final image size is what counts, the lens scales should be a conservative estimate of DOF when used on an APS-C camera. There, I said it,and it feels good to get it out.
Steve

Umm... don't follow you again here.

I understand about image size but the DoF scale on a lens designed for 35mm will still be erroneous for an APS-C - erroneous, as in wrong - the opposite of "just" conservative. The scales on a 30mm lens (for 35mm) will still show a DOF of around 46 feet ... on ANY camera.

This is just wrong for an APS-C where the DoF will be around 15 feet. um... that's not conservative - that's 31 feet of extra blur..... *dun dun duuuunn* .... 1.5 times less DoF.

The scales on the _lens_ don't change just because of the camera it's on. This fact is independent of the FoV/Image Dimensions.

[danielchtong]

Hello everyone,
How many of you use zone focusing with your primes? Most of my prime lenses have zone calibrations next to the distance scale on the barrel. I find it very useful, not to have to focus constantly using this feature. It ensures that you don't miss that special shot. It is especially useful in low light where you have a little working distance from the subject. You should try it with your manual focus primes. I think that you will find it useful.

Dave

I use mostly zone focusing with my 16mm Zenitar. It is a manual lens and to me there is only 2 zones depending what I want to focus. I did not use the elaborate tables as listed above as I am fairly familiar with the lens.


This one was obvious. I was shooting while holding/feeding the bird the same time






Shot from floor level





Daniel

[georgweb]

Daniel,
these are so good, I still have to put those filter rings on my Zenitar like you suggested. And you're right, on wide angles there's 2 or max. 3 focus zones, should be quicker than AF once set up correctly.
Could'nt resist, two more examples of zone focus success

Kiron/Vivitar 28/2 on K100D


Enna Lithagon 4/24 on K100D

Georg (the other)

[keithlester]

And a follow-up question or two. For those who use hyperfocal focusing for their landscape photography, how do you set the lens to the hyperfocal distance? It's pretty hard to read, for example, 3.03m on the distance scale on a 31LTD lens. I'm thinking about a bit of string that length with a little flag on the end. Tie one end to the tripod head and stick the flag in the ground to focus on.

Does the camera software get an accurate reading of focus distance? Would it be possible for Pentax to build in a function that automatically sets the focus distance of the lens to the hyperfocal distance for the selected aperture?

You dont focus on the hyperfocal distance itself. You put your infinity mark on one of your aperture marks and are in focus down to the other mark. That's the traditional way of using the scales for PAS photography.

IMO landscape photography is better done by critical manual focussing using a magnifier. If you are using a tripod, it makes no sense trying to use PAS technique (or lack of), how would that be of any benefit?

If you are just taking landscape snaps, in some kind of a rush, it would be better to use the auto-focus kit zoom that almost certainly came free with your camera. Buying manual prime lenses only makes sense if you are going to focus manually, for which purpose you have a viewfinder and your eyesight, which will always outperform hyperfocal guesswork.

[Frank Fletcher]

You dont focus on the hyperfocal distance itself. You put your infinity mark on one of your aperture marks and are in focus down to the other mark. That's the traditional way of using the scales for PAS photography.

If you don't have scales (a modern zoom), or are using a custom table/ring/software, you do focus on a hyperfocal distance.

I use this all the time. It's exactly 1/3 deep into the area of critical focus. I don't look it up, I just know the 1/3 rule and use it.

IMO landscape photography is better done by critical manual focussing using a magnifier. If you are using a tripod, it makes no sense trying to use PAS technique (or lack of), how would that be of any benefit?

I see you have a tilt/shift lens in your lineup. Perhaps you come from a large format background. The manual focusing, chimping the lcd comment makes sense in that light. Personally, the LCD is too small and not bright enough for me - especially in full sunlight. It's different if I'm shooting 4x5.

You know, I rarely use manual focusing for my landscape photography. The autofocus does an adequate job and the tolerances are far greater than, for instance, macro.

If you are using a tripod, you are already beyond "Point and Shoot". If you are using hyperfocal scales, tables, etc - you are beyond PAS.

If you are just taking landscape snaps, in some kind of a rush, it would be better to use the auto-focus kit zoom that almost certainly came free with your camera.

What does the auto-focus kit zoom have to do with hyperfocal technique? Also HF is not something people often do in a rush. If one were in a rush, they'd likely just use f/22 and forget about it. You might as well do that with a prime anyway if you're considering an el-cheapo zoom. Your results, even with (pr especially with) diffraction will be better on the prime.

Even if I were to use the auto-focus zoom kit, I'd still need HF (or LCD chimping) to get the whole scene in focus. Neither zoom nor autofocus are relevant factors.

Buying manual prime lenses only makes sense if you are going to focus manually, for which purpose you have a viewfinder and your eyesight, which will always outperform hyperfocal guesswork.

Often times, the viewfinder on the K10d is too dark to adequately judge focus across the frame using analog DoF preview, or stopping down manually.

Furthermore, if I'm using hyperfocal technique, I _know_ that my subject/s will be adequately in focus. There is no guesswork. Or am I just guessing that the sun will rise tomorrow morning?

Why do prime lenses come in automatic focusing these days? That wouldn't make sense if they were just for manual focusing.

Personally, when I'm backpacking, my walk-around lens is the DA 40mm limited. It is capable of fully automatic use. I most often preset my aperture and speed manually - but use the autofocus.


One of the beautiful things about a prime is that when using HF, all you have to worry about is distance and aperture. If you know your camera and lens. all the other variables go away.

[Frank Fletcher]

Knowledge of DoF, and the use of scales/tables/etc, does not presuppose HF and citical focus to infinity.

This is especially true of near-focus work. I might want to know the near and far limits are so that my subject is fully in focus and the background still has a pleasant bokeh.