Hyperfocal distance

[roysmith]

I´ve been reading of the hyperfocal distance theory, and did some tests with my A700+SAL18-250 but it doesn´t seem to work, everything looks out of focus... I used an hyperfocal calculator (http://www.dofmaster.com/dofjs.html) and as I understand the correct hyperfocal distance for my A700@18mm f/8.0 is 2.04m. I dialed that on my focus ring (well, kind of, as closest match is 3.2m) and took the pic (everything in manual mode of course, and using mlu). After that took the same pic but with autofocus, and autofocus looks always a lot sharper... am I doing something wrong? Or I´m just missing the right spot on the focus ring?

Thanks!

[Dr. Harout]

Roy, if the table says that everything will be in focus from 2m to infinity (example) at a given aperture that doesn't mean that the focus point should be at 2m.
To be more precise, let us say when you focus your lens at 5m and set the aperture on f/11, the hyperfocal distance would be from, say 2.5m to infinity (I'm saying those numbers just to give you an idea. So in this last case you will put the focus point on 5m and not 2.5m (but objects at 2.5m will be in focus).
One more thing, being in focus doesn't mean razor sharp. Closer to the focal point sharper it is (with pixel peeping of course).

[roysmith]

So I'm not supposed to focus at the hyperfocal distance? Where should I focus then?
I thought the hyperfocal distance was mostly meant to know where to focus in order to get the maximum DOF, not just to know the DOF for a given combination of focal lenght+aperture+circle of confusion, which anyway I suppose it will be dynamic and change depending on the focus distance.
Now I'm really confused

Anyway thanks for the help, Dr.!

[UrsaMajor]

I´ve been reading of the hyperfocal distance theory, and did some tests with my A700+SAL18-250 but it doesn´t seem to work, everything looks out of focus... I used an hyperfocal calculator (http://www.dofmaster.com/dofjs.html) and as I understand the correct hyperfocal distance for my A700@18mm f/8.0 is 2.04m. I dialed that on my focus ring (well, kind of, as closest match is 3.2m) and took the pic (everything in manual mode of course, and using mlu). After that took the same pic but with autofocus, and autofocus looks always a lot sharper... am I doing something wrong? Or I´m just missing the right spot on the focus ring?

I'll assume that the calculator you are using works correctly - which is not a guarantee, as there is a lot of misinformation on the Internet. I'll also assume that the scale on your lens is accurate, and that both the calculations and your lens scale are in the same units (Metric or English).

The thing to remember is that any calculated Hyperfocal distance such as this - right or wrong - is based upon an assumption of what is "acceptably sharp". That is why that calculator has a provision for setting a value for the "circle of confusion". If you are using the default value for the circle of confusion, try changing it to a smaller value to see if the results are more satisfactory.

When you focus at a given distance, objects at that distance should be as sharp as your lens can deliver. Objects either closer than that distance or farther away will be less sharp, as the lens is not optimally adjusted for those distances. The greater the deviation from the focus distance, the more an object at that distance will be out of focus. (A point object will start appearing as a larger and larger circle the more you move away from the focus distance.)

Even though a point object starts going out of focus as soon as you move even a tiny amount away from the focus distance, your eye and brain will not notice this until the circle produced by that misfocused point reaches some given size. That size varies with several factors, including individual judgement. This is why I suggest changing the value used for the "circle of confusion" in that calculator.

There is also the possibility that the programming of that calculator produces erroneous results. I assume that the creator of the program is using the correct equations, as those come from fundamental physics that has been widely known for many decades, but there is always a chance that a mixture of English and Metric units has erroneously been used in the calculations, or that there is an error in the actual computer code.

With best wishes,
- Tom -

[Dusty]

Roy, don't let a supposed hyper-focal distance get you too wrapped up. As Doc said, focus is sharpest at the focal point. The one time you may change that is when you're shooting something (scenics work best) far away or with wide lenses.

Both far distance and use of wider lenses increases apparent DOF, so there's no need to focus at infinity using f16 on a 10mm lens. You're blowing all that DOF in front of the lens. Back it up a bit and get more of the front in focus, unless that's not desired for the shot. Old rule of thumb was that DOF is 1/3 in front and 2/3 in back of the focus point, but what that distance is is controlled by Aperture, lens length and focus point. The calculator changes that, but then, rules of thumb are just that, i.e., approximations.

Unless I'm working wide of very long, I'll focus on the subject and let DOF fall where it may, (although I usually shoot in A mode to control it!)

Dusty

[roysmith]

Ok, I'm starting to understand now... I think
Most of my pictures are landscapes, and so far haven't noticed any problems with DOF (I use mostly f/8 @ 18 to 28mm). I recently read about the hyperfocal distance and that this is commonly used by landscape photographers, so I wanted to give it a try. So far nothing seems improved nor sharper, but will keep trying as the focusing ring on the SAL18250 is not very precise and that may be the problem...
By the way, the CoC value I'm using is 0.02mm, which accordingly to that website is the correct one for the A700. It can be wrong so I'm going to look and read a bit more about that too.

Thanks a lot for your help!

[Chris Malcolm]

I found when shifting to digital from film and getting better lenses than I'd ever had for film that I was capable of taking much sharper larger images than I ever had with film. But only occasionally with good luck until I'd learned the new relationships of shutter speed, stability, and focus accuracy demanded by these higher resolutions. My new higher standards of sharpness and resolution also meant that I had to revise the old ideas about DoF, which my higher standards had now made much shallower.

[Johnc]

For those interested, there is a free DOF calculator for Windows Mobile at http://winmodof.free.fr/en/
Downloaded a few days ago.
Interesting.

[harvey]

Ok, I'm starting to understand now... I think
Most of my pictures are landscapes, and so far haven't noticed any problems with DOF (I use mostly f/8 @ 18 to 28mm). I recently read about the hyperfocal distance and that this is commonly used by landscape photographers, so I wanted to give it a try.

Do you experiments with a scene with close foreground detail. For example:

• Field of flowers with you standing in the flowers
• Wall or fence in the foreground
• Lake/seascape/river bank with close rocks in the foreground

And see how much forground you can get in focus while having everything to infinity acceptably sharp.

Harvey

[01af]

circle of confusion n. -- A group of photographers desperately trying to understand depth-of-field.


Yes, 0.02 mm is the 'correct' maximum diameter of the circle of confusion for APS-C format, and slightly more than 2 m is about right for the hyperfocal distance of an 18 mm lens on an APS-C-format camera at f/8 (yielding DOF from 1 m to infinity).

Still, relying on DOF for good sharpness is not a good idea, particularly when expectations on sharpness are high. Near the borders of the DOF range, i. e. at the front limit and the rear limit of the DOF, sharpness will be just-so acceptable---read: far from really good. And focussed at the hyperfocal distance, infinity is at the rear limit, so things at far distance won't be sharp at all.

In fact, far things are small, so in order to get them acceptably sharp in an image, sharpness at infinity mustn't be 'acceptable' (by whatever standards) but near-perfect. Think of this as 'relative sharpness'---the smaller an object gets depicted (i. e. the farther away it is), the better the sharpness must be for the object to be recognizable. When accepting this thought, the concept of hyperfocal focussing immediately becomes nonsense. If things at the horizon are important in your image's composition then by all means do focus at infinity. Do NOT rely on hyperfocal focussing! This basically is the essence of Harold Merklinger's tremendously worthwhile article 'The Ins and Outs of Focus' ... essential reading for every landscape and architectural photographer.

Hence, the generalized rule for proper focussing is: Choose one thing in your frame as the primary subject, then focus on that.

-- Olaf

[roysmith]

Ok I get it now. I´m happy to go back to infinity focussing without feeling guilty. Plus it´s easier to remember and faster
I may use this hyperfocal distance only in special cases when "infinity focused DOF" is not enough.
Thanks a lot for the info!

[Birma]

Very interesting article - thanks for the link Olaf.

[WaltKnapp]

Still, relying on DOF for good sharpness is not a good idea, particularly when expectations on sharpness are high. Near the borders of the DOF range, i. e. at the front limit and the rear limit of the DOF, sharpness will be just-so acceptable---read: far from really good. And focussed at the hyperfocal distance, infinity is at the rear limit, so things at far distance won't be sharp at all.

In fact, far things are small, so in order to get them acceptably sharp in an image, sharpness at infinity mustn't be 'acceptable' (by whatever standards) but near-perfect. Think of this as 'relative sharpness'---the smaller an object gets depicted (i. e. the farther away it is), the better the sharpness must be for the object to be recognizable. When accepting this thought, the concept of hyperfocal focussing immediately becomes nonsense. If things at the horizon are important in your image's composition then by all means do focus at infinity. Do NOT rely on hyperfocal focussing!

Hence, the generalized rule for proper focussing is: Choose one thing in your frame as the primary subject, then focus on that.

-- Olaf

The bottom line of that is that there is no DOF advantage in FF vs APS since in both cases only the exact focus plane will be in sharpest focus (as shown by pixel peeping). So by your definition DOF is no shallower in FF than APS. In fact DOF does not exist!

I think I'll stick with DOF does exist and control of DOF is an important part of composition. Including sensable use of hyperfocal focussing.

Walt

[01af]

So by your definition DOF is no shallower in FF than APS. In fact DOF does not exist!

Huh!? Are you nuts? I never 'defined' such a nonsense. Of course DOF does exist. However DOF is DOF and the plane of focus is something else.

... control of DOF is an important part of composition.

Sure it is. Still, DOF is no alternative to proper focussing.

-- Olaf

[David Kilpatrick]

Both Walt and Olaf are nuts - very useful expert nuts! Depth of Focus (image plane) versus Depth of Field (subject plane) is an issue many who are not nuts (in the best sense) struggle to grasp. When macro comes into the equation and the two are equal for a small 'window' round 1:1 (with a symmetrical or simple lens system) it's even harder to explain but very easy to demonstrate.

Hyperfocal distance is dependent on many assumptions and conditions for final image use. I think it's almost 'broken' in high res digital photography. We are now able to see the image at such a high magnification we either have to rewrite old rules, or start looking at prints not pixels.

Anyway, good to log on and see a lively argument... keep it civil folks!

David