Jump to content

Help: Confused about 1/3 hyperfocal rule of thumb


stsadasdsad

Recommended Posts

I'm struggling with landscape sharpness from front to back in a scene like the attached. I've tried the 1/3 rule of thumb for hyperfocal, but just realized I may have been misinterpreting that. I'm also trying manual focus and I'm a little confused about elements of it. Can someone help me with feedback on the below scenarios? I'd appreciate it!

 

Auto-Focus Mode

- I shot both of these in AF-S (stationary) single point focus.

- These are both F8, ~1/160s, ISO 100, no tripod, Sigma 10-20 at 10mm

- I think I've been misunderstanding the rough 1/3 of the way into the picture shorthand rule of thumb for hyperfocal. I had been interpreting it as the lower horizontal third in the viewfinder (in this shot, corresponds to the immediate right of the foreground tree shadow), but now I'm wondering if the 1/3 way focus point should be somewhere between the foul pole and the infield dirt because, if you factor in total distance to background tree line, that might be a more accurate rendering of 1/3 the total distance. Is that right? Where would you place you focus point in this picture for hyperfocal sharpness?

- I've attached one where the autofocus point was to the right of the foreground tree shadow and another where it was between infield and outfield. Despite what I said above, I don't notice any practical difference in sharpness.

 

Manual Focus Mode

- I'm trying to learn how to do this, as I've read recommendations to use it for landscape sharpness.

- In other shots, I tried to manually focus around where I thought the hyperfocal was, but, in the sunlight, I couldn't see enough in Live View to tell whether it was indeed sharp in that area.

- If you were doing Manual Focus in the sunlight, how would you go about a shot like this? (e.g. I think I read a suggestion somewhere about setting the focus ring to just short of infinity for landscapes and not worrying about ever dialing in focus, or something like that, but wasn't sure if that was correct or feasible)

 

betweeninfieldandfoulpole.thumb.jpg.bf7f7e62436f82ce15f734fa4f8cbace.jpg

 

torightofshadowtreeinforeground.thumb.jpg.4ef94d592e8c382234bfc9d2bcf674f7.jpg

Link to comment
Share on other sites

Many modern lenses don't have the scales on them like the old lenses did. I shall go ahead anyway.

 

hassybg.jpg.6b104ca6aa8cf6fc254e48dc23f385eb.jpg

 

 

The lens of the left is focused at the horizon (infinity). At f/22 everything is in acceptable focus from about 18 feet to infinity.

The focus ring on the right has been moved so f/22 is below the infinity mark.

At f/22 everything is in acceptable focus from infinity to 9 feet. (Above 22 on the left of the scale.)

The hyperfocal distance is at the center pointer, about 18 feet.

  • Like 1
James G. Dainis
Link to comment
Share on other sites

That rule of thumb is meant to be applied to the subject distance from the camera, not its position in the frame.

 

However, the one-third/two-thirds 'rule' is a bit nonsensical anyway; after all, what's one third of infinity? And depth-of-field varies between being symmetrical fore and aft of the subject at lifesize magnification, to grossly asymmetrical at distances of a hundred metres or so.

 

Depth-of-field also varies greatly with the focal length of lens used, so that any rule of thumb quickly becomes useless when very wide angle or telephoto lenses are used.

 

The lens you've used for those examples is classed as a super-wide angle, where depth-of-field is very deep and the 1/3rd-2/3rds 'rule' simply doesn't apply. Not in any meaningful way.

 

I suggest you forget that rule of thumb, and take the guesswork out of it by using a depth-of-field app on your phone. There are quite a few available, but the free app called HyperFocal Pro appears to do everything needed, and with quite a nice graphical representation of depth-of-field.

 

It's a pity that many modern lenses either don't have a distance scale, or one that's too sketchy to be of much use.

Edited by rodeo_joe|1
  • Like 1
Link to comment
Share on other sites

 

I suggest you forget that rule of thumb, and take the guesswork out of it by using a depth-of-field app on your phone. There are quite a few available, but the free app called HyperFocal Pro appears to do everything needed, and with quite a nice graphical representation of depth-of-field.

 

 

This is all super helpful! If I can tap your expertise further, this photo example actually hits on a challenge I've been facing with hyperfocal experimentation to date.

 

I've been using PhotoPills, and, for a Nikon D7500 at 10mm at F8, it's telling me that my hyperfocal is 2 feet. My problem (as shown in my attached shots) is that the ground 2 feet in front of the camera is not in frame at 10 mm. So what I've been doing instead has been single point autofocusing on the nearest ground in front of me using the bottom middle single focus point and then recomposing up. The dual problem there would seem to be: 1) that groundpoint in frame is likely to be, say, 10-20 feet rather than the hyperfocal 2 feet and 2) the up-and-down autofocus and recomposing across that plane isn't likely to be helping my sharpness issues. So, in this example photo, how would I best achieve the hyperfocal?

 

And a more theoretical question: Should I mainly be using hyperfocal in cases like the attached photo where there's nothing of particular interest in the foreground, midground, or background and instead you're just documenting the breadth of an entire landscape scene? And let's say 40 feet into this photo, there was a nondescript bush, but your eye would still naturally go in that direction....even if I'm still just documenting the entire scene, should I ditch the hyperfocal and instead set focus on the bush?

 

Thanks again SO much for any help you can provide!

Edited by stsadasdsad
Link to comment
Share on other sites

Set your af/ae lock to af lock only, focus on something, anything two feet away, such as your left hand, engage the af lock, then recompose and shoot.

 

Better would be to focus on the subject you absoloutly want to be sharp, then use your depth of field calculator to select an aperture that will render as much of the scene as you like 'acceptably sharp'

  • Like 1
Link to comment
Share on other sites

Should I mainly be using hyperfocal in cases like the attached photo where there's nothing of particular interest in the foreground, midground, or background

You've just described a rather uninteresting photograph, from a generic standpoint. UWA lenses are useful tools, but only if used correctly. I find I'm more successful for outdoor UWA shots if I use the deep DoF to get very close to a primary subject, and then allow background objects to create negative space behind the primary subject. UWA lenses are a special tool requiring special techniques. I recommend accessing some of the many excellent primers on their use.

  • Like 1
Link to comment
Share on other sites

Set your af/ae lock to af lock only, focus on something, anything two feet away, such as your left hand, engage the af lock, then recompose and shoot.

 

Better would be to focus on the subject you absoloutly want to be sharp, then use your depth of field calculator to select an aperture that will render as much of the scene as you like 'acceptably sharp'

 

I'm not sure I follow. Are you saying to put my hand two feet in front of the lens, focus on it, and then remove it from the frame?

Link to comment
Share on other sites

You've just described a rather uninteresting photograph, from a generic standpoint. UWA lenses are useful tools, but only if used correctly. I find I'm more successful for outdoor UWA shots if I use the deep DoF to get very close to a primary subject, and then allow background objects to create negative space behind the primary subject. UWA lenses are a special tool requiring special techniques. I recommend accessing some of the many excellent primers on their use.

 

I totally get what you're saying....I may be overstating what I mean by "interesting." Check out the attached for a better example....In this case, in my opinion, the whole thing is interesting but there's not one singular element to focus on--it's the sum of the parts. At the same time, since it was a roadside thing with limited time and movement options, I couldn't get down and positioned more dramatically behind the bush in the foreground. So I think I just focused on the bush from where I was, but wasn't happy with the overall end result--it didn't seem sharp throughout the entire picture (and the bush didn't seem so sharp either). If it were you, where would you have focused in this one?

64354866_10107681702443521_3904658006922493952_o.thumb.jpg.bb522eb595b07784d5b575af07b57def.jpg

  • Like 1
Link to comment
Share on other sites

I'm not sure I follow. Are you saying to put my hand two feet in front of the lens, focus on it, and then remove it from the frame?

Yes.

 

Focus on an object at the desired distance.

 

Engage the af lock (not the ae lock).

 

Recompose scene as desired.

 

Take photo.

 

Focal point will be at the previously locked distance, in this case two feet.

 

 

You don't have to point the camera even remotely at the desired scene, as all we're doing is locking the focus at a set distance.

 

This is a common trick for street/candid work, focus on something at the same distance as your subject, lock the focus, then turn and shoot.

 

Incidentally, I'd give the hyperfocal distance a margin of error, if you really need infinity to be good, use the figure for an aperture one stop wider. When shooting at f8, use the hyperfocal distance for f5.6, for example, to keep the mountains sharp, just be aware that you'll lose a bit in the foreground, which may or may not be important.

Link to comment
Share on other sites

 

You don't have to point the camera even remotely at the desired scene, as all we're doing is locking the focus at a set distance.

 

 

Thank you--this explains so much. Even though I do side-to-side/up-and-down autofocus-and-recompose all the time within my landscape/cityscape compositions, I wasn't making the mental connection that it's locking in distance rather than subject (even though it should have been obvious to me). That's why I was thinking I was violating some cardinal rule of the universe by autofocusing on something that wasn't even going to end up within the compositional frame. Thanks much!

Link to comment
Share on other sites

I totally get what you're saying....I may be overstating what I mean by "interesting." Check out the attached for a better example....In this case, in my opinion, the whole thing is interesting but there's not one singular element to focus on--it's the sum of the parts. At the same time, since it was a roadside thing with limited time and movement options, I couldn't get down and positioned more dramatically behind the bush in the foreground. So I think I just focused on the bush from where I was, but wasn't happy with the overall end result--it didn't seem sharp throughout the entire picture (and the bush didn't seem so sharp either). If it were you, where would you have focused in this one?

[ATTACH=full]1300889[/ATTACH]

The DOF focus is fine but the camera appears to be hand held. Use a tripod for landscapes. Then worry about DOF.

  • Like 2
Link to comment
Share on other sites

I totally get what you're saying....I may be overstating what I mean by "interesting." Check out the attached for a better example....In this case, in my opinion, the whole thing is interesting but there's not one singular element to focus on--it's the sum of the parts. At the same time, since it was a roadside thing with limited time and movement options, I couldn't get down and positioned more dramatically behind the bush in the foreground. So I think I just focused on the bush from where I was, but wasn't happy with the overall end result--it didn't seem sharp throughout the entire picture (and the bush didn't seem so sharp either). If it were you, where would you have focused in this one?

[ATTACH=full]1300889[/ATTACH]

 

This looks good as-is unless a pixel peeper. You need to stack photos in PS to get 100% sharpness.

 

I don't do much landscapes. I'm into street work. For street work you don't use the formula so much. You try to estimate the point of interest to set focus and let the hyperfocal work for you making it better.

  • Like 1
Link to comment
Share on other sites

The DOF focus is fine but the camera appears to be hand held. Use a tripod for landscapes. Then worry about DOF.

 

Concur. Landscapers are the anal branch oh photogs, like studio photogs. A good landscaper scouts out the scene days ahead then gets there in plenty of time the day of the shoot to get the light. I could never do it.

  • Like 1
Link to comment
Share on other sites

The first thing to remember is that using the hyper-focal distance doesn't mean that "everything between this distance and that is in perfect focus". It means that everything is in "acceptable" focus for a particular magnification of the final image (print or screen). The only distance that is at "perfect" focus is the one set on the lens. From that point, in front and in back, the focus gets progressively worse. This is just a matter of physics.

 

So, if the table tells you that your hyper-focal distance is 2 feet and nothing in your image is 2 feet or closer, you're wasting your time by focusing at 2 feet. In fact, that will actually assure that EVERYTHING in the image is OUT of focus. You will see little difference between focusing at 10, 15, 30 or 50 feet in your image.

 

In the long run, it's best to decide what distances you want in sharpest focus and focus on that distance. Then, in an image like this, find the aperture that ensures that the closest and farthest distances that you care about will be in focus. With a 10mm lens on a DSLR, somewhere between F5.6 and f11 is going to work best. Going much smaller introduces diffraction which is a whole other story.

  • Like 1
Link to comment
Share on other sites

The first thing to remember is that using the hyper-focal distance doesn't mean that "everything between this distance and that is in perfect focus".

+1 to Ed's comments. For landscapes, when using hyperfocal distance, I typically focus on an object as far away as possible while still giving me the DoF I want. It's important to keep in mind that there are few instances when OOF foreground elements will be a positive component, but with some notable exceptions.

  • Like 3
Link to comment
Share on other sites

I got to thinking about the nuances of DoF when using a UWA lens. Going back to the OP and the issue of DoF and UWA lenses, my recent post in the NW forum reminds me of an example:

1376195336_CrabPot-7336a-sml.jpg.46fbdf5819677c2d14c224ad60793a0a.jpg

In this case the crab pots were all stacked fairly close together. I wanted to isolate some interesting details without completely losing the larger context. My 11-16mm Tokina allowed me to get right up close and personal with this stack (12 inches focal distance +/-), keeping the other stacks in view, and the f/5 aperture further isolated my main subject. Had I wanted to keep the background in focus I would have used a substantially smaller aperture, say f/16, but I would have lost the isolation effect provided by a modest DoF and introduced some diffraction effects. The lens's maximum aperture of f/2.8 would have been far too shallow for this subject, at about 6 inches, while this image uses a DoF of about 12 inches. What I hope this illustrates is the need, particularly with UWA lenses, to understand just how sensitive images are to selecting both the right aperture and correct focus point. Using a UWA lens for vast landscapes is fine, if that's what you want, and focus point is more flexible. Using a UWA lens for shots like this one, which is where I believe they really stand out, requires a detailed understanding and application of the principles governing DoF and composition. I hope this is helpful...

Link to comment
Share on other sites

There seems to be a misunderstanding the phrase 'hyperfocal distance' here. It's a specific instance of depth-of-field, not an alternative use of 'depth-of-field'.

 

For any given sensor size, lens focal length and aperture, there is only one hyperfocal distance. It's that focussing distance where depth of field reaches exactly to infinity - no less and no more. It's meant to maximise depth-of-field, specifically for landscapes. The hyperfocal distance tells you where to focus the lens for maximum depth-of-field at any given aperture, and with a particular lens. 'Twice the distance', and other such nonsense, don't.

 

Even the concept of depth-of-field is open to question in these days of easy huge magnification through pixel-peeping. There is only one plane of focus, and everything else is more-or-less out of focus. Keep magnifying the image and your depth-of-field gets narrower and narrower as you magnify.

 

So rules-of-thumb, and even conventional hyperfocal or depth-of-field calculations, are only good for an 'average' viewing size, whatever that may be.

 

There's an alternative way to look at maximising depth-of-field. Theoretically, the biggest image blur circle (circle of confusion) you can get, is equal to the physical lens aperture diameter multiplied by the magnification. So a lens of 10mm focal length, at an aperture of f/10, will have a 'hole' size of 1mm; and that's the largest amount of blur of any object in the real scene. For example, blades of grass will appear, at maximum, to have a one millimetre spread of blur around them in the real world, as will anything else that's completely out of focus. That's actually quite sharp-looking in a landscape image; meaning that with a 10mm lens @ f/11 you can pretty much focus anywhere and the image will be acceptably sharp. Try the same with a 300mm lens @ f/4, and its 75mm real-world maximum blur circle will quickly become very noticeable! Even at half or twice the focussed distance.

Edited by rodeo_joe|1
Link to comment
Share on other sites

I don't think that there is much of a misunderstanding here . . . Just some issues being left out of the discussion.

 

"Average viewing size . . . " is selected by whoever prints up the DOF charts. It takes into account the sensor (or film) size and some print size. If you are shooting 4x5 film and make 8x10 prints, you are only magnifying the circle of confusion by a factor of x2. Making the same 8x10 print from 35mm you are magnifying the circle by a factor about x9. Then, of course, you need to consider the viewing distance for the print . . .

 

So, in general . . . These conversations have no end . . .

Link to comment
Share on other sites

Rodeo-J, Thank you for clearing that up. I understand hyperfocal distance but was getting a bit confused by all the nonsense put in. There is only one hyperfocal distance for a given lens at a set aperture based on acceptable focus from the horizon or infinity. "Horizon/infinity" is the key word(s).

 

"Twice the distance" is misleading. It would work in my example above if one viewed an object 9 feet away and set the focus to 18 feet. That is the hyperfocal distance in my example. Thus, acceptable focus is 9 feet to infinity.

 

At f/22 on an 80mm lens, if one were to view an object 6 feet away and set the focus to 12 feet then acceptable focus would be from about 6 feet to 36 feet. The horizon is not in acceptable focus. That may not be as important with ultra wide angle lenses but let us not confuse things.

James G. Dainis
Link to comment
Share on other sites

....from the horizon or infinity.

 

Technically the horizon isn't infinity, but for landscape work it's as far distant as we usually care about.

 

While we can't mathematically put a number on 'infinity', it has a very definite and measurable optical meaning. It's that lens-to-image distance where in-focus rays become parallel, and can be found practically by a device called an auto-collimator. This stuff is done by the lens maker, and hopefully there'll be an infinity mark on the lens's focus scale to tell us where that happens..... hopefully.

 

On the other hand; depth-of-field tables are purely theoretical, and assume 1) That the lens's nominal or marked focal length is the same as its actual focal length and 2) That the lens behaves like a perfect 'thin lens model'. These are both very big assumptions! So depth-of-field tables and hyperfocal distance calculations should be taken with as large a pinch of salt as you find appropriate. Sometimes they're useful, and other times it's just a waste of time reading 'em!

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...