Beginner Hyperfocal Distance

[Dominic A]

I've recently been trying out the difference between focusing on what I need and hyperfocal distance. I've not been seeing very much of a difference though, and sometimes even worse. I want everything from a certain point up to my image in focus, but I think focusing on the actual point is actually better than calculating hyperfocal distance to find a distance in between. I'll post two images, first being hyperfocal and second being focusing on the object itself and you'll see a blurriness on the image. Could anyone please help me to see if there's a way to maybe... Fix this? Or if hyperfocal is actually useful.

Hyperfocal:
http://imgur.com/b0kpZ2K

Autofocus:
http://imgur.com/fC4bPm8

 

[woof woof]

I suppose a lot could depend on how closely you look at the pictures. Isn't there really only one plane of focus? So aim for front to back sharpness and look closely and you may be disappointed.

Another technique you could look at is Merklinger method. If you Google it you should find it pretty quickly. Here's a link for you....

https://luminous-landscape.com/digital-focusing-part-two/

"Set the lens to infinity then divide the focal length of the lens by the aperture. This will give you, in millimeters, the subject resolution limit from the far distance to right in front of the camera. For example, say you are using a 35mm lens focused on infinity, and the aperture is f/11, then 35 over 11 is three (give or take). Now any object in the scene larger than 3mm will be identifiable, provided it’s within the resolution capabilities of the lens and sensor. So even though you’re focused on infinity, a 3mm wide blade of grass in the immediate foreground will still be distinguishable as a discreet and distinct object. It may be a bit soft, but it’s still identifiable."

The link for Part 1 seems to be broken so here's another...

https://luminous-landscape.com/digital-focusing-part-one/

Hope that helps... but I often decide to forego hyperfocal or Merklinger and focus on my subject

 

[sk66]

For hyperfocus to be of much use you need to be using a wide FL, a smaller aperture, and you need to have something very close to the lens you need in reasonable focus. Otherwise, you are almost always better off focusing on the subject.

 

[Dominic A]

Ah, for Woof Woof I'll definitely give that a go and thank you for bringing that up! Didn't even know it existed!

Sk66 - Thank you as well, I didn't really bring that into considering and I'm guessing that hyperfocal would be better if there was say... A rock in the middle and I needed that to be in focus alongside the background?

Thank you for all the help as well, greatly appreciated!

 

[woof woof]

For hyperfocus to be of much use you need to be using a wide FL, a smaller aperture, and you need to have something very close to the lens you need in reasonable focus. Otherwise, you are almost always better off focusing on the subject.

I'm not so sure that I agree with any of this

Hyperfocal tables are there for just about any focal length but the proviso is that they're for specific image sizes and viewing distances and they sort of work...

You don't necessarily have to use a wide angle lens to use hyperfocal technique. I have a lot of hyperfocal shots taken with 50 to 100mm focal lengths and you don't always need a small aperture.

One thing that I think is maybe often overlooked is that you might not need a depth of field that goes from 1mm in front of the lens to infinity and if you can move the starting point out (how close is the nearest object?) and the end point in (how far away is the most distant object?) you may be able to use a wider aperture than if you just blindly thought.... I need near front to infinity...

 

[sk66]

I'm not so sure that I agree with any of this

You will never get very close hyper focus w/ lenses 50mm and longer... not without going WAAAY into diffraction. And if the near stuff you want to keep is 50-100ft away, do you really need "infinity?" If you don't, then focusing on the subject (or "~1/3 into the scene") generates essentially the same results.

The "problem" with hyperfocus is that you are sacrificing "actual focus" of the subject for a larger area of "acceptable focus," and that's not a compromise that's usually worth it... unless it's a landscape type shot with very near FG interest. And to get under 12ft w/o significant diffraction you need a lens wider than 50mm.

 

[Willid1]

I have never seen a hyperfocal distance calculator that seemed to match what I experience.

In the film days sharpness was assumed to be judged on a reasonably small print - certainly no bigger than 10x8 inches so clicking 100% on a digital image and scrolling round thinking " oh this is not very sharp is not the same thing.

For real front to back everything sharp you probably need to do something like focus stacking with a series if shots blended together - not that I have ever tried this.

D

 

[woof woof]

You will never get very close hyper focus w/ lenses 50mm and longer... not without going WAAAY into diffraction. And if the near stuff you want to keep is 50-100ft away, do you really need "infinity?" If you don't, then focusing on the subject (or "~1/3 into the scene") generates essentially the same results.

The "problem" with hyperfocus is that you are sacrificing "actual focus" of the subject for a larger area of "acceptable focus," and that's not a compromise that's usually worth it... unless it's a landscape type shot with very near FG interest. And to get under 12ft w/o significant diffraction you need a lens wider than 50mm.

Yet again I find myself disagreeing

Of course you can use 50mm and longer lenses on a 35mm camera and use hyperfocal technique. I have a metric ton plus of images at 100mm taken when using hyperfocal or Merklinger method on my pc

Hyperfocal and Merklinker and any other similar methods out there such as focusing 1/3 into the scene are always going to be fudge of sorts but the key is to think about image size and viewing and when/if you do so the fudge/trick might work but of course once you look closely at 100%+ on the screen it can all fall apart, as David points out above, but if the only other option is focusing on your subject one of the former techniques may be preferable as what are things that are not on the same plane of focus as the thing you focus directly on going to look like? They could be more out of the plane of focus. The key is to think about the final image and viewing and base your decision of what to do on that but of course the same could be said for a shot where you focus on the subject.

With diffraction the theory is that diffraction kicks in a fx but in reality what matters is the effect on the final image. Again I have a metric ton + of images which should show diffraction but do you know how many people have said "You know woof woof, that shot is ruined by diffraction..."? The answer is none

In real world shooting with a 35mm camera the normal working aperture range could be... f5.6 to f11? That wouldn't be too unusual would it? There are cameras that'll show diffraction way before f11 but in the final image will the theory matter?

Just for fun here's a DoF table from a reputable source, 50mm/35mm camera...



and here's one for 100mm...



I've never done focus stacking either but I can see the advantages. Macro and product shots spring to mind... but I suppose there are lots of uses for it.

 

[HoppyUK]

Hyperfocal focusing is the setting that delivers maximum depth-of-field, in terms of total distance. As such, the far distance must always be infinity. In other words, if you don't want the far distance to appear sharp, then you're not using hyperfocal focusing. In the DoF tables above, only the bottom line - Hyperfocal Distance - applies.

In practise, Steven is right - hyperfocal distance focusing is simply impractical with longer lenses, and the real limit is a wide-angle or at least semi wide-angle lens. If you take f/11 as Alan suggests (as an acceptable diffraction limit) and apply that to the 50mm lens table, HFD is 24ft and therefore the nearest subject that will appear sharply rendered will be 12ft away*. Is that close foreground? Maybe. But with a 100mm lens, HFD jumps to 97ft with nearest sharp subject therefore 48ft away, nearly 15m - that rules out close foreground subjects being anywhere near acceptably sharp.

*Handy fact - when HFD is set, the nearest object that will be rendered acceptably sharp is always at exactly half the HFD. Eg, with a HFD of say 10ft, the nearest sharp object will be 5ft away.

Edit: oops! Typos.

 

[snerkler]

For me those scenes that the OP posted are not ideal for using this technique (although that doesn't mean that you can't). I personally reserve hyperfocal distance for landscape photography, and is the technique I apply all the time when shooting your 'traditional' landscape.

OP without seeing the EXIF it's difficult to really comment on the sharpness of the image, we don't know how fast the shutter speed was and whether these were handheld? But for scenes like the above I would always focus on what you want in focus. Hyperfocal distance is most useful when you want focus to infinite, where as the scene you have your scene is only several metres.

 

[Dominic A]

Ah, so you would recommend it if the scene was the more traditional sense of landscape. Thanks for all your help guys, it's been a massive help and I will certainly give your guys' opinions and techniques a go. Thank you for all the help.

 

[woof woof]

In practise, Steven is right - hyperfocal distance focusing is simply impractical with longer lenses, and the real limit is a wide-angle or at least semi wide-angle lens. If you take f/11 as Alan suggests (as an acceptable diffraction limit) and apply that to the 50mm lens table, HFD is 24ft and therefore the nearest subject that will appear sharply rendered will be 12ft away*. Is that close foreground? Maybe. But with a 100mm lens, HFD jumps to 97ft with nearest sharp subject therefore 48ft away, nearly 15m - that rules out close foreground subjects being anywhere near acceptably sharp.

As I seem to be happy to disagree with people on this issue I'm happy to do so again, to a point or at least and state a case for longer lenses.

Richard, you may accept that "in practice" will depend upon the image size and viewing and also the scene. With the longer lenses near things which may be in the frame with a wider lens may not be in the frame so the 12 or 48ft may not be a limiting factor and indeed for compositional purposes relatively far objects could become your near objects when using the longer lens depending on the scene and composition. The distances (12 or 48ft or whatever) may sound a lot but depending upon your scene it may not matter. Diffraction may or may not be a real world issue in the finished shot and personally I've found that more often than not it's much less of an issue than the theory and figures would suggest. What matters is the final image, I have a lot of pictures taken with longer lenses which display (good enough) front to back DoF but I agree with snerkler really, none of this is too applicable to the OP's example shots. 100mm is no good here, best to shoot with a wide angle and smallish aperture

I'm tempted to ask why the far distance must always be infinity but I'm happy to accept that if it isn't it's not hyperfocal, maybe it's zone focusing, it doesn't really matter what it's called if it works for the shot.

 

[HoppyUK]

<snip>
I'm tempted to ask why the far distance must always be infinity but I'm happy to accept that if it isn't it's not hyperfocal, maybe it's zone focusing, it doesn't really matter what it's called if it works for the shot.

The hyperfocal distance focusing setting starts with infinity and works back towards the camera. If it doesn't include infinity, it's not HFD. That doesn't mean other methods are invalid, or that there aren't better techniques for certain situations, just that they're not HFD. http://www.cambridgeincolour.com/tutorials/hyperfocal-distance.htm

 

[woof woof]

So what term are we happy with for front to back dof which doesn't go to from near front to infinity? Zone?

Not that I spend too much time with photographers and I do try to avoid technical discussions when I am with them but I do think that people can sometimes use hyperfocal technique, near front to infinity, and therefore smaller apertures when the composition doesn't require it. Maybe a conversation for another day/thread.

 

[StephenM]

Personally, I use "depth of field" which was defined (when I first learned it) as the distance between the nearest and furthest points that were acceptably sharp. Maximum depth of field was then stated to occur when the lens was focused on the hyperfocal distance. And knowing the hyperfocal distance, you could then calculate approximate values for the depth of field when focusing on any whole integer fractions of this (I may have the wrong term - I mean 1/2, 1/3, 1/4 etc of the hy[erfocal distance).

 

[simonkit]

It's useful to know about and occasionally useful in practice but I wouldn't use it all the time, as you've found sometimes it's just not relevant to the type of shot you're trying to achieve. Ultimately the only part of the shot which is 100% sharp is the focus point. If thinking about landscapes and real world shooting, anyone looking at your images doesn't necessarily expect to see the distant mountain 100% sharp, infact it wouldn't look quite right if it was because to our eyes it never is. A foreground object 2 feet away is more critical though so personally I just adjust the focus point to achieve the balance of sharpness/ DOF I want in each shot.. I do sometimes use hyperfocal as a good start point but make adjustments from there

Simon

 

[snerkler]

So what term are we happy with for front to back dof which doesn't go to from near front to infinity? Zone?

.

As Stephen says I would personally just use the term depth of field.

 

[Steve Smith]

Hyperfocal is from a pre-determined point up to infinity. What you want is depth of field with the furthest distance at (or just past) your subject.


Steve.

 

[sk66]

*Handy fact - when HFD is set, the nearest object that will be rendered acceptably sharp is always at exactly half the HFD. Eg, with a HFD of say 10ft, the nearest sharp object will be 5ft away.

Another handy fact: No matter how far beyond the HFD you focus, the near DOF limit will never be farther than the HFD, and the far limit will always be infinity.
Swagging the numbers: a 50mm @ f/11 on FF has a HFD of ~25ft. No matter how far beyond the HFD the focus is placed the DOF will never be less than 25ft-infinity.

That is what makes just focusing on the subject with longer lenses more practical most of the time.... is having the near limit start at 12ft instead of 25ft more important than having your subject actually in focus? Not likely...

 

[snerkler]

Hyperfocal is from a pre-determined point up to infinity. What you want is depth of field with the furthest distance at (or just past) your subject.


Steve.

That's quite a useful cool little fact

OP here's a good online calculator if you don't have one already

http://www.dofmaster.com/dofjs.html

 

[Steve Smith]

And just to add a bit more information, whatever the suggested focus distance is, the area of acceptable sharpness is defined as half that distance up to infinity.

It is a method of maximising the depth of field. As an example, if you had a shot of a distant mountain with some rocks in the foreground, you could set your focus to infinity, but you would be wasting half of your depth of field. The useful half would be from infinity down to somewhere determined by the aperture whereas the other half is from infinity to greater than infinity... which doesn't exist!!

By adjusting the far point of the depth of field to infinity, the near focus point is brought even closer. So with the right aperture, you can get the mountain and the rocks in focus.

It makes much more sense on the lenses we film dinosaurs use as they have depth of field markings for each aperture. By lining up the relevant far point with infinity, you get hyperfocal. No need for charts or calculators.


Steve.

 

[StewartR]

By adjusting the far point of the depth of field to infinity, the near focus point is brought even closer. So with the right aperture, you can get the mountain and the rocks in focus.

Yes indeed, but.... this touches on the problem I have with hyperfocal distance focusing as a concept. By definition, the mountain is right on the edge of the depth of field; in other words it's right on the edge of the zone of acceptable focus. But whose definition of 'acceptable' are you using here? It might not be good enough for you. So using the hyperfocal distance might be a perfect way of ensuring that the focus is not quite right.

 

[simonkit]

Hi Dominic... as you see it can get quite complicated... My advice would be, certainly as a beginner, don't get too obsessed about terminology.. in fact initially I'd try to avoid it as it can just lead to confusion.

Just use the Hyperfocal distance as a general "rule of thumb" and experiment with different apertures and making adjustments to the focus point, as it seems you are, but don't expect too much from the "theory", it's just that.. practice and experimentation will prove far more useful to the point where you can pretty much forget all about hyperfocal distance!

Simon

 

[Steve Smith]

Yes indeed, but.... this touches on the problem I have with hyperfocal distance focusing as a concept. By definition, the mountain is right on the edge of the depth of field; in other words it's right on the edge of the zone of acceptable focus. But whose definition of 'acceptable' are you using here? It might not be good enough for you. So using the hyperfocal distance might be a perfect way of ensuring that the focus is not quite right.

Yes... But you are in charge. And knowing what is going on helps you to take control.


Steve.

 

[sk66]

It makes much more sense on the lenses we film dinosaurs use as they have depth of field markings for each aperture. By lining up the relevant far point with infinity, you get hyperfocal. No need for charts or calculators.

Swagging the HFD is pretty simple... no need for charts or calculators. The hard part these days is actually setting the focus at the right distance...

 

[HoppyUK]

Swagging the HFD is pretty simple... no need for charts or calculators. The hard part these days is actually setting the focus at the right distance...

Given the close distances usually involved with HFD, like a few yards at most, it's easiest to estimate the distance by eye and using single-point AF, focus on something at that distance.

Lens distance scales are notoriously inaccurate, or imprecise (or both) especially with zooms. The reason we don't see DoF scales engaved on lens barrels any more is a) because they're zooms and there's no room for a what would be a complicated set of numbers* and b) since the same lenses are now used on both FF and APS-C format cameras, you'd need at least two DoF scales.

*Some old push-pull zooms used to have DoF scales, eg Vivitar Series-1, as the physical design made it possible.

 

[sk66]

Just estimate "long" if infinity is important... or estimate "short" if infinity isn't very important. If you're good with estimating distances, great; otherwise bracketing might be in order.
The reason I said it's difficult is the short throw AF gearing of modern lenses and their very basic scales... they're fairly worthless. I'm kind of surprised they're still wasting space on them...

 

[HoppyUK]

Just estimate "long" if infinity is important... or estimate "short" if infinity isn't very important. If you're good with estimating distances, great; otherwise bracketing might be in order.
The reason I said it's difficult is the short throw AF gearing of modern lenses and their very basic scales... they're fairly worthless. I'm kind of surprised they're still wasting space on them...

It sounds like a contradiction, but you can actually put a number on infinity, at least as far as focusing is concerned. But there are problems in practise. I don't know about you, but I can estimate a few feet or yards quite accurately (the near distance of HFD) or pace it out if you like, but longer distances are much harder. The other problem is, infinity changes a lot with focal length.

I don't think there's any alternative to having a reference chart for HFD focusing. I have just a handful of HFD numbers, which is all you need, stuck inside the lens cap. The process is 1) decide on the closest object you want sharp, 2) double it, estimate that distance visually, and then focus on it with centre-point AF. That's HFD set. 3) Refer to the table and set the aperture required against that distance and the lens focal length. Done.

There are lots of DoF and HFD calculators available for smartphones, eg DOFmaster http://www.dofmaster.com/iphone.html

 

[snerkler]

The process is 1) decide on the closest object you want sharp, 2) double it, estimate that distance visually, and then focus on it with centre-point AF. That's HFD set. 3) Refer to the table and set the aperture required against that distance and the lens focal length. Done.

There are lots of DoF and HFD calculators available for smartphones, eg DOFmaster http://www.dofmaster.com/iphone.html

Found this confusing at first as after you say to focus on the object that's HFD set I thought to myself no it's not, but then you went on to say you then need to choose the aperture for that distance, obviously only then does step 2 become the HFD. Helps if I read it all first lol.

I actually do it differently to this, I choose my focal length and aperture first and then use a calculator/chart to work out the HFD. As I'm generally using between 18-24mm I know that near distance is very small so all the foreground is sharp. Same result, different method

 

[HoppyUK]

Found this confusing at first as after you say to focus on the object that's HFD set I thought to myself no it's not, but then you went on to say you then need to choose the aperture for that distance, obviously only then does step 2 become the HFD. Helps if I read it all first lol.

I actually do it differently to this, I choose my focal length and aperture first and then use a calculator/chart to work out the HFD. As I'm generally using between 18-24mm I know that near distance is very small so all the foreground is sharp. Same result, different method

Haha, I see what you mean, but actually HFD is set at stage 2) for that subject, regardless of lens or aperture, because HFD is always exactly double the distance of the nearest object you want sharp, regardless of anything else. But yes, you then have to select the right aperture to bring everything into focus at the focal length you're using. And it doesn't matter in what order you do these things.

For anyone interested, these are the numbers I have inside the lens cap of my 17-40mm lens (on full-frame*) used for shooting landscapes. It's easy enough to guess any in-between settings as while the numbers in DoF tables looks very precise (they have to be, to make the maths work) this is actually not a very critical process. Depth-of-field doesn't suddenly go from sharp to blurred, it's very gradual, and if you're a couple of feet out it probably won't make any visible difference.

..................f5.6........f11........f22
17mm.......5.6ft......2.8ft.......1.4ft
24mm.......1.2ft......5.7ft.......2.9ft
35mm........24ft.......12ft..........6ft

The other thing I do, if the situation allows, is choose an f/number one stop higher than the table suggests (though it's best to avoid the highest f/numbers to minimise softening from diffraction). This will more than cover any errors, but more to the point the sharpness standards for DoF were drawn up many moons ago when we were all shooting film and a 10x8in print was regarded as a big enlargement. In other words, they're pretty loose and you will get better sharpness at the close and far distances by stopping down a bit more. Oh, and BTW, if you crop an image, more than just a trim, that is effectively a change of sensor format and it changes the DoF!

*Choose the ones you want for your camera/lens from www.dofmaster.com

 

[sk66]

I don't think there's any alternative to having a reference chart for HFD focusing.

For FF @ f/11 the HFD is the FL as a percentage of the FL... i.e. 50% of 50mm= 25ft. Or, 10% of the FL squared if you prefer... i.e. 5x5=25ft.
If you change the aperture by 2 (i.e. f5.6/22) the HFD changes by 2 (i.e. 50ft/12.5ft) larger aperture=longer HFD; the "in betweens" are proportional.

F/16 works the same for the 1.5x crop factor, and is probably close enough for 1.6/1.7x... it will just be a little short.

The problem with all of this is that if you estimate/focus even a little short you will loose a lot of DOF from infinity. But if you estimate/focus long you only loose a little from the near side.

 

[HoppyUK]

For FF @ f/11 the HFD is the FL as a percentage of the FL... i.e. 50% of 50mm= 25ft. Or, 10% of the FL squared if you prefer... i.e. 5x5=25ft.
If you change the aperture by 2 (i.e. f5.6/22) the HFD changes by 2 (i.e. 50ft/12.5ft) larger aperture=longer HFD; the "in betweens" are proportional.

F/16 works the same for the 1.5x crop factor, and is probably close enough for 1.6/1.7x... it will just be a little short.

The problem with all of this is that if you estimate/focus even a little short you will loose a lot of DOF from infinity. But if you estimate/focus long you only loose a little from the near side.

I would need to have that lot written inside my lens cap too! A table is easier, and I'm not sure your first 50% example works for other focal lengths.

Using the same table for different formats will lead to unacceptable errors.

For typical landscape situations, which is really where HFD technique applies most commonly, accuracy really hinges on your ability to estimate short distances within a foot or so - say in the 4-12ft kind of range. I can do that well enough. The other thing is you need to focus accurately on that distance, which is where using single-point AF does the job for you. Trying to set focusing distance accurately from the distance scale on most lenses these days is frankly hopeless. There's another thing too, though hardly significant except in extremes, is that focusing distances on the lens are measured from the camera's sensor, whereas depth-of-field tables are calculated from the front of the lens.

 

[sk66]

I would need to have that lot written inside my lens cap too! A table is easier, and I'm not sure your first 50% example works for other focal lengths.

No 50% only works for 50mm.
For a 30mm it would be 30% (10ft) or 3x3 (9ft)... actual is 8.8ft. The difference in near DOF when focused at 10ft instead of 8.8ft is around 2".
For something like 24mm I would swag it as 25% of 24 or 2x3 (6ft)...actual is 5.65ft.

 

[snerkler]

Haha, I see what you mean, but actually HFD is set at stage 2) for that subject, regardless of lens or aperture, because HFD is always exactly double the distance of the nearest object you want sharp, regardless of anything else.

Well to be pedantic, no as focus needs to reach infinite for it to be the HFD

If I want something at 2' to be sharp then by your method I'd have to focus at 4' and that won't always give you infinity focus and so is not the HFD as shown here. You need step 3 (i.e. the aperture) to make it the HFD

Sorry, I know this is being really pedantic, but it could lead someone astray

 

[HoppyUK]

No 50% only works for 50mm.
For a 30mm it would be 30% (10ft) or 3x3 (9ft)... actual is 8.8ft. The difference in near DOF when focused at 10ft instead of 8.8ft is around 2".
For something like 24mm I would swag it as 25% of 24 or 2x3 (6ft)...actual is 5.65ft.

Ah right, I get you now. That's quite a handy as f/11 is a good aperture for this

 

[sk66]

Ah right, I get you now. That's quite a handy as f/11 is a good aperture for this

And to get your f/5.6-22 chart numbers you just double/halve the f/11 HFD distance.
No charts/app and I can guestimate it pretty accurately for any lens/FL I choose to use. Which is quite handy with a zoom. For instance, set your 17-35 to 20mm and the f/11 HFD is 4 ft (20% of 20, or 2x2); the f/22 distance is 2ft and the f/5.6 distance is 8ft... your chart doesn't tell you that (actuals are 7.8ft/3.93ft/2ft). And you can even use it for the in-between apertures...i.e. f/8 is ~ 6ft (~1/2 way between the f/11 and f/5.6 distances) and f/16 is ~ 3ft.

F/16 works as the starting point for 1.5x CF (a bit more error for the 1.6/1.7x Canon CF's).