[stevebrot]

es, you go right on ahead and keep thinking that there is ~85' of dof in that shot

You need a new calculator.*

I would suggest the Wikipedia article on Depth of Field, but that is pretty heavy on math. There are multiple tutorials on the subject and many good books. I would suggest Ansel Adams, "The Camera", but he talks about film cameras and the concepts are obviously obsolete. Here is something to chew on...

When all variables are accounted for and those that can be are cancelled out are, DoF hinges on three considerations: final magnification, viewing distance, and absolute aperture (not f-number)

I know...but, but, but...


Steve

* Yours is off by an order of magnitude

[osv]

we agree in that the *perception* of dof can(depending on fl) change as you change the distance.

that's what the dof calculators give us; the data is ~300ft., f/7.1, ff, 420mm... even Online Depth of Field Calculator says that photo has 67 feet of dof

blame it on the dof calculator? see if you can find a better one

my point is, there is only one hard-coded actual dof in a photo, you can see it at 100% viewing size, it can't be altered, and you can take advantage of that when focusing the camera, using magnification in liveview, to set an exact focus.

[stevebrot]

that's what the dof calculators give us; the data is ~300ft., f/7.1, ff, 420mm... even Online Depth of Field Calculator says that photo has 67 feet of dof

Yep, that is what mine says too. However, the image shared in your previous post has been downsampled and/or cropped. The crop may or may not affect DOF, but downsample or upsample always does. Assuming the horizontal dimension had only been resized, that equates to almost a 5-fold decrease in magnification and a proportional increase in DOF.

You are correct that 100% on the rear LCD or EVF represents a set DOF that cannot be narrower, but is specific for a given size display (130mm for the A7R) and magnification (0.71x for the A7R)*. What's more that set DOF will typically be the same or less (usually a lot less) than a print made from the capture viewed at a reasonable distance, depending on magnification.

Just so there is no confusion, DOF means that within that range, the viewer perceives the subject as being "in focus". Move back a few centimeters and what was once blurry become sharp and DOF increases. In other words, DOF is always perceptual and is related to the viewer's visual acuity.

Are there doubts? If so, that is OK...continue on...

Going back to the example of 100% on an EVF, lets consider 100% on an external monitor with 7360 x 4912 pixel display resolution (36 Mpx 3:2, same as the A7R) and 36" x 24" physical screen size (43" diagonal). Yes, this is pixel-to-pixel mapping of the sensor to the monitor and when viewed should represent the native DOF for a given subject distance, focal length, and relative aperture (f-number). Sound good? BTW, horizontal dot pitch for this monitor is a little over 200 dpi if that helps.

Now that we have a nice big real world representation of "real" DOF, lets scoot our chair forward until we can just barely see the individual dots and then back up just a smidgen. Wow! We are really close and have trouble viewing the screen at a single glance. The DOF at that viewing distance is the narrowest for this setup with the individual dots occupying a smidgen less than 1 minute of arc (general resolution limit of the human eye).

Question? What happens if we scoot the chair back and double the distance? Will more or less of the frame appear in focus or will it stay the same? Why?


Steve

* A corollary is that an increased pixel count translated to the same EVF will show even less DOF at 100% than a lower count.

[RGlasel]

Are there doubts? If so, that is OK...continue on...

Sorry, but unless you are getting some personal satisfaction from this, there is no value in beating this dead horse. You are correct in every way, but some people seem to forget the idea behind photography is taking pictures to be viewed later and all the logic, reason and common sense you contribute isn't going to change that. I don't want to put you on ignore or unsubscribe from the thread, but for me, this is a very one-sided discussion at the moment.

[mee]

thank you, that's pretty helpful

A little different but this one is fun too -- click on the 'advanced settings' link to get your hands dirty.

Understanding Your Camera’s Hyperfocal Distance

[Digitalis]

if you are on crop, the first rule is to never stop down further than f/8 unless absolutely necessary, because that's the beginning of the range of visible diffraction.

A possibly lethal amount of salt would be required to take on this advice, especially with wide angle lenses with near-far compositions you have no alternative but to stop down to f/11 and beyond. Diffraction isn't a wall, it is a gradual softening and there are sharpening techniques that can be used to counter its effects*. People should not have the fear of god put into them over issues such as diffraction, what do you have to say to the macrophotogapher who @ 3:1 magnification has to stop down to f/32? Or the Astro/landscape photographer who uses their lenses wide open at f/1.2? both will encounter different negative effects on image quality but in the end: it is the image that matters, who are you to pass judgement on them?


*To be honest, to an extent.

[Ian Stuart Forsyth]

we agree in that the *perception* of dof can(depending on fl) change as you change the distance.

that's what the dof calculators give us; the data is ~300ft., f/7.1, ff, 420mm... even Online Depth of Field Calculator says that photo has 67 feet of dof

blame it on the dof calculator? see if you can find a better one

my point is, there is only one hard-coded actual dof in a photo, you can see it at 100% viewing size, it can't be altered, and you can take advantage of that when focusing the camera, using magnification in liveview, to set an exact focus.

There is no hard-coded DOF in a photo there is just the point of focus and everything on either side is OOF. Not until we define what is an acceptable level of sharpness can we derive a usable DOF, that all depends on how we are going to view the image, size of viewing, distance and eyesight. What you are viewing when zooming at 100% has nothing to do with DOF , what you are seeing is blur from the resolution limit of the sensor, lens aberrations or diffraction, and all you are doing is hiding the true point of focus in that blur and that is not DOF.

---------- Post added 04-13-2016 at 10:03 PM ----------

A possibly lethal amount of salt would be required to take on this advice, especially with wide angle lenses with near-far compositions you have no alternative but to stop down to f/11 and beyond. Diffraction isn't a wall, it is a gradual softening and there are sharpening techniques that can be used to counter its effects*. People should not have the fear of god put into them over issues such as diffraction, what do you have to say to the macrophotogapher who @ 3:1 magnification has to stop down to f/32? Or the Astro/landscape photographer who uses their lenses wide open at f/1.2? both will encounter different negative effects on image quality but in the end: it is the image that matters, who are you to pass judgement on them?


*To be honest, to an extent.

Often times people that make a statement of not shooting anything smaller than F /8 on cropped do not fully understand that blur from not having the needed DOF has a far greater blurring effect on the image than diffraction would ever have.

---

Last edited by Ian Stuart Forsyth; 04-13-2016 at 10:12 PM.

[pacerr]

A looong time ago, before I'd ever heard of such things as hyperfocal-watchamacallit I understood two things.

First, it was my responsibility to ensure the point of sharpest focus was going to be where I wanted it to be when the shutter snapped. Made no difference what sort of view finder was available, one did one's best with the tools available. If I had to 'ambush' the action by pre-focusing on a point so be it. (Would that be pre-chimping?)

Second, that the 'window-of-really-sharp-stuff' was gonna be dependent upon the aperture I chose and that was gonna be absolutely dependent on all that other exposure stuff no matter how much DOF I wanted.

If the smallest aperture suited to the exposure didn't allow for as much 'really-sharp-stuff' as I wanted . . . well, that was just the best that could be done at that point in time. To my knowledge, that situation hasn't changed a bit in the last half century. Aperture rules!

I'd note there's a whole lot of difference between ASA 25 and ISO 3200 to help tweak DOF too. (Let's see about tryin' some a that fancy Tri-X next payday.)


And, yes, you can set hyperfocal distance with any lens, with or without a distance scale assuming you're familiar enough with the sensor/lens chart to know what you want it to be. Simply go manual and pre-focus on something, anything, at the proper distance. DOF will still be whatever the aperture allows it to be and a DOF chart for the sensor and lens FL will offer a clue as to what the results ought'a be.

[panonski]

a little different but this one is fun too -- click on the 'advanced settings' link to get your hands dirty.

understanding your camera’s hyperfocal distance

and that's exactly what i'm talking about

[stevebrot]

Sorry, but unless you are getting some personal satisfaction from this, there is no value in beating this dead horse. You are correct in every way, but some people seem to forget the idea behind photography is taking pictures to be viewed later and all the logic, reason and common sense you contribute isn't going to change that. I don't want to put you on ignore or unsubscribe from the thread, but for me, this is a very one-sided discussion at the moment.

Thanks for the reminder.


Steve

[stevebrot]

A little different but this one is fun too -- click on the 'advanced settings' link to get your hands dirty.

Understanding Your Camera’s Hyperfocal Distance

That "Advanced Settings" for the DOF calculator is just too cool and like most of the stuff on "Cambridge in Color", very educating. Thanks for sharing.


Steve

[osv]

Question? What happens if we scoot the chair back and double the distance? Will more or less of the frame appear in focus or will it stay the same? Why?

you are just re-hashing what's already been said: "we agree in that the *perception* of dof can(depending on fl) change as you change the distance."

it's not germane to the point that was made, which was, the actual dof in the photo can be used to set far more accurate focus than is possible with any hyperfocal calculation.

---------- Post added 04-14-16 at 10:23 AM ----------

There is no hard-coded DOF in a photo there is just the point of focus and everything on either side is OOF.

if it was that easy, we could use one aperture for everything, all dof would look the same, f/1.4 would look like f/8... but you can stand at any distance, and see the difference between f/1.4 and f/8; because dof is hard-coded into the photo.

you can't change the point of focus regardless of how much you change the distance, which is the entire point of using magnification, you have complete control over where the point of focus is, and you can judge the limits of dof much more accurately when you can see where actual dof blur begins and ends.

i feel sorry for people who think that they are setting accurate focus without even looking into the ovf, just basing it on some sloppy hyperfocal guesswork calculation

---------- Post added 04-14-16 at 10:28 AM ----------

Often times people that make a statement of not shooting anything smaller than F /8 on cropped do not fully understand that blur from not having the needed DOF has a far greater blurring effect on the image than diffraction would ever have.

using the wrong dof is not relevant to diffraction blurring, it's an entirely different problem, which is easily solved with magnification, because you can see it better.

[pacerr]

i feel sorry for people who think that they are setting accurate focus without even looking into the ovf, just basing it on some sloppy hyperfocal guesswork calculation

The whole point of using HFD is that you DON'T KNOW exactly what pop-up opportunities will arise and/or you may not have time to refine focus so you're doin' the best you can to capture a fleeting moment in time within a pre-focused window.

You wanna see truly effective use of HFD? Look through some of the combat images from any 20th century conflict shot with the likes of a 4x5 Speed Graphic or a 6x6 TLR or an RF 35mm - or even an old Spotmatic. Auto-focusing back then meant habitual use of HFD!

If you're in the studio with oodles of lighting options and aperture is flexible, get out the micrometer and decide for yourself just how much DOF is enough. A DOF chart is based on the physics of optics but you do have to take into account its definition of 'sharp enough' - it's situational - and it's subjective.

The corollary to the "F/8 and be there." quote (attributed to Weegee) is, "You're a fool if you don't habitually pre-focus for the most likely scenario." (me) The HFD/DOF chart applicable to your gear is the best tool you've got to anticipate results and ensure success. (PPPPPP!)

Makes no difference how well you focus, or miss focus, when the shutter snaps DOF is dependent on aperture and your control of sharpness is gonna be primarily dependent upon having sufficient shutter speed. It is what it is. It's about pragmatism, not perfection; setting HFD's simply the best tool available under some circumstances.

Do you want a chance at that once in a life-time, money shot? Make it a habit to set an appropriate HFD for the expected scenario. Oh, yeah, an' then there's pre-setting Sunny Sixteen SWAGs for exposure too. Another "sloppy" habit that makes success more likely.

[stevebrot]

The corollary to the "F/8 and be there." quote (attributed to Weegee) is, "You're a fool if you don't habitually pre-focus for the most likely scenario..."

I could not have said it better...


Steve

[monochrome]

What an interesting read this thread has been.

Given that all precision is deterministic, "Intelligent estimation" wins the working concept award.