"understanding exposure" distance setting...

[mrklaw]

i'm re-reading this, and finding it a lot more understandable now i have a little more experience.

However, i'm still confused by his comments on focusing for maximum depth of field.

He talks about simply setting the correct distance on the distance marker. But he doesn't seem to mention anywhere *what* distance to use. He just mentions that it will depend on your focal length.
Also, i can't see a distance marker on my lenses. Some have a focus window, is that what he means?

I have heard many talking about focusing 1/3 into a scene. Forgive my ignorance, but is that 1/3 actual distance between the closest and furthest subjects, or the point that appears 1/3 up in your viewfinder?

[tdodd]

You achieve maximum depth of field by focusing at the "Hyperfocal Distance". When you focus at that distance your depth of field will extend all the way to infinity, and will also give you sharp focus as close as possible (while keeping infinity sharp) for your given choice of camera body, focal length and aperture.

To be honest, the distance scales on the lenses are not very precise, and the precision rapidly disappears as soon as the focus distance is beyond a few feet. The best thing, in my opinion, is to work out the focus distance you need, from tables or a hyperfocal distance calculator, and then using your judgement just aim the camera at an object, even just the ground, that is at roughly the right distance away, and focus the camera at that point. Then recompose and take your shot.

You can workout the hyperfocal distance by using the online calculator here....

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

There's a thread here - http://www.talkphotography.co.uk/forums/showthread.php?t=81863 - which might help, or might confuse, but it goes into some depth about picking suitable apertures to achieve the DOF you want, taking into account sensor size and the concerns of diffraction at small aperture sizes. As you will note from the discussion in that thread, the idea of focusing 1/3 into the scene is nonsense, at least for landscapes. Maybe it works for a specific combination of sensor size, focal length and aperture, but in most cases you would focus far closer than 1/3 into the scene and still achieve a sharp horizon.

Of course, you also want to be able to control DOF more precisely, perhaps to do a group shot of three rows of people - say a football team or wedding party. You need to make sure you have everyone sharp from front to back but you hardly need the horizon to be sharp, probably better if it isn't. For thos situations you can forget the hyperfocal distance as it is not what you want to use. Instead you need to find a combination of body/focal length/aperture and subject distance that covers the range you need. Again you can use the online DOF calculator to help you figure out some numbers.

e.g. if you wanted to shoot a group of people three rows deep - say requiring a DOF of 3', and you were 10' away from the people in the front row with an 18mm lens on a 40D you could happily use an aperture of f/2.8 with no problem and still have ample DOF. If you focused on the middle row, say at 11', your DOF would extend from 7.08' to 17.5', which is more than enough. If you had a 50mm lens you'd actually need an aperture of around f/5.6, focused at 11', to give you a DOF from 9.63' to 12.8', which would be about perfect. Perhaps focusing at 11' 2" would be even better, but basically, focusing on the faces/eyes of the second row should be good enough.

[mrklaw]

I do understand that, but the author's point I think was that you didn't need all these tables, and it was all simple.

Doesn't seem to be then with modern lenses?

[tdodd]

When everything was 35mm film (which is when the book was written, despite its pretentions to cover digital) it was common for lenses - primes especially and sometimes zooms - to have a DOF scale marked so you had a rough idea of what would be in focus at different focal lengths and focus distances. I have a 50mm prime, a wide zoom and a long zoom dating back to the 1980s for my old Pentax Super A. The prime and the long zoom both have very rough markers indicating the DOF range for any given aperture, focal length and subject distance. The short zoom has no such markings. The markings for the long zoom are only for f/11 and f/22 so you'd only have the faintest clue what was going on at f/5.6, f/8 or anything else.

Here is an example of an old zoom lens that does have some lines marked on it showing how the DOF varies at different focal lengths for some given apertures. It's not my photo so I can only post a link to the web page, but I have an old lens quite similar to this...

http://litpixel.com/ee/photo.php?photo=757&exhibition=9&ee_lang=eng&u=1241,31

Nowadays, most/many lenses can be fitted to either a crop body or a full frame 35mm body, and that makes it next to impossible to mark the information required. Also, modern cameras in the XXD range have the A-Dep mode to allow the camera to select a suitable aperture for you and the 1D series allows you to spot focus chosen points and allow the camera to pick the required aperture to cover that focus range. Basically the technical environment and the technology has changed.

On top of that, Bryan seems to like to avoid the real nitty-gritty of some things and glosses over the details that can so often be important. It is a book for beginners, paying only lip-service to digital gear, and, I think, none at all to crop body cameras as a special case, and keeps things simple for its own convenience, even if that does nothing to aid the longer term education of the reader. That's just my opinion. Plenty of people love the book but I find it very disappointing.

If you point me to the page(s) in the book where you've been reading this stuff I'll have a look at my copy and see if I can make sense of what he says.

[izzy35]

I am so glad to have read this thread...In the book, i was confused and have read, re-read AND re-read, this section again and am clueless..I believe the page is top of 39..it begins on 38.it says, "distance settings....they allow u to preset the DOF b4 u take your shot.." I think this is what the OP is talking about, bc i have practically memorized these 2 pages hoping 1 day the light bulb would go off in my head..LOL!!

[mrklaw]

page 39, with further repeats of similar info later on. also the caption on page 39 is a good example that he uses often. "hey, I just put 2ft on the distance marker and voila!". voila what? how do you know to put 2ft?

[tdodd]

OK, in my copy I think we're talking about pages 38-39 and the specific point of concern is at the top of page 39, like you say. First the good news....

- On page 38 Bryan does remark about the auto-DOF capabilities of some Canon cameras, although I don't think his explanation of how it works is exactly correct, but it's something;

- Bryan does comment on the modern day bias towards zoom lenses, which seldom (read "never") have a DOF scale marked, so he has that point covered.

From there it all rapidly goes downhill. I think he has discombobulated his words when he gets to page 39.

He says you have "distance settings", which is true (but only for some lenses) - they are the focus distance settings. He goes on to say that these distance settings allow you to preset DOF. That is not really 100% true. They allow you to preset your focus distance, roughly, as the distance scale is not very detailed or precise. However, contrary to what he says, the DOF is not yet determined until you also set your aperture. It is the combination of focus distance and aperture together which determine your DOF.

He then goes on to say that for your storytelling composition you set aperture to f/22. Well that's not necessarily true either. A lot of lenses will stop down to f/32 or even f/45 so why not use those apertures in preference? No explanation is given. Further, there is an effect that occurs with smaller apertures (larger f/numbers) called diffraction softening. This is visibly evident at f/16 on crop body cameras like the 450D and 40D, making f/22 a potentially unwise choice for these cameras. Your DOF may increase at f/22 but actually the whole picture, including the sharpest focused point will appear slightly softer. f/22 might be fine on a 35mm full frame camera, but should be used with caution on crop bodies. Indeed, with very wide angle lenses, like the 10-22 on a 40D you can get a massive DOF at modest apertures like f/8. The advantage with stopping down only as far as you need, and no more, is that you can keep higher shutter speeds and/or reduce your ISO, which are both plus points.

Then to round off things there's this glib throwaway remark that you choose your focus distance based on the focal length of your lens. WTF? OK so I have a 10mm lens - where should I set my focus? Bryan gives no clue. What about a 35mm lens? Again, stoney silence on the subject. In the breakout text for the picture of the yellow flowers he says he set the focus distance to 2', but there is no explanation why he picked 2'. Is it experience, guesswork, photographic memory of all DOF/hyperfocal tables for all focal lengths, apertures, camera formats etc.? Perhaps he doesn't really care, because as far as he's concerned it's only newbies reading the book so who cares if he is not very clear or accurate? They won't know any better. Just let them swallow it up like the fools they are.

Once you get some experience you begin to realise how lacking the book is.

Short story, as I said, he totally glosses over the nitty-gritty of details required to do the job properly. He explains some things badly, some things wrongly, and some things not at all. Of course, he gets some things right, but I am entirely unsurprised that people are confused by much of what he says.

EDIT : and another bugbear of mine with this book - despite the revision to cater for digital, there is no mention made of using the histogram to check exposure and adjust if necessary. Nor is there any mention of exposing to the right for raw shooting. I know the book is for newbies, and can't cover everything, but to exclude continual reminders to check the histogram (and to understand how to interpret it) is, quite frankly, ridiculous in a book about "Understanding Exposure".

[izzy35]

focus distance settings that you get on any lens (apart from the Nifty Fifty, which has none) Ok as a TOTAL newbie, i don't even know what this is??? what are the focus distance settings anyway??

[mrklaw]

exactly tdodd!


I do like his explanations for the three main types of shot, and how you can group the apertures - yes you can use others but its a starting point. I also like his comments about the actual exposure - eg metering on different things to help you get the right setting.

But it does have some holes like that distance setting thing.


perhaps if old cameras used to have these tools, there is space for digital to include them in some other way via software?

[tdodd]

Ok as a TOTAL newbie, i don't even know what this is??? what are the focus distance settings anyway??
If you have a DSLR then many lenses will have a distance scale marked on them, in feet and meters, to indicate at what distance the focus of the lens is set. Five out of six of my lenses have a focus distance scale. The only lens I have that doesn't is the Canon EF 50mm f/1.8 II lens - the Nifty Fifty. I now understand that several entry level lenses, like the 18-55 and 55-250 kit lenses lack these distance scales as well.

Anyway, close up - like a foot or two away, the scale (if you have it) might be marked every foot or couple of feet. Further away, the distance scale gets less and less precise. If you shoot in manual focus mode you can adjust the focus by hand to a specific point on the focusing scale. It is hard (impossible) to be accurate, but you can get the lens somewhere in the right ballpark. If you want to set the lens focus to a specific distance but can't see it on the scale, or can't tell where exactly it should be, or don't have a scale at all, then you can guesstimate a subject or point at roughly the right distance away and focus (manually or automatically) at that point. e.g. if for some reason you need to focus 12' away then imagine the length of a small living room, or a couple of grown men lying down end to end, and then focus at that distance.

If you have a point and shoot then you will have no distance scale on the lens. If you are lucky (a bit) then you may have a facility to set focus manually for that camera. I have a point and shoot that lets me set focus through the menu system at 0.5m, 1m, 3m, 7m, 20m, infinity. Alternatively, you can use the technique above to aim and focus at something that is at about the right distance away and then hold that focus setting while you recompose and shoot. This may give you issues with your exposure but that's a limitation with compacts that don't let you separate the functions of focusing and metering your scene.

Here is an example of a distance scale on a lens, marked in meters and feet. Right now this lens is set to about 20m, or a bit over 50' (actually, quite a bit over 50', since 20m = 65.6', which illustrates how imprecise the scale markings are). The red markings are related to infra-red photography, and you can ignore them completely...

[Lowner]

"When everything was 35mm film (which is when the book was written, despite its pretentions to cover digital) it was common for lenses - primes especially and sometimes zooms - to have a DOF scale marked so you had a rough idea of what would be in focus at different focal lengths and focus distances".

I have never understood the logic behind the removal of those so simple, but so effective few engraved lines. It cannot be cost!

[tdodd]

Those engraved lines are fine on a push-pull zoom, but what about zooms with internal focus and twist zoom mechanisms, like the 70-200? The lines also only cover a handful of apertures - 2 or 3, not the full range - so they are of limited use. What would you propose as a solution to cover an EF lens that might be used on both full frame and crop cameras? What about lenses that do not even have a distance/focus scale?

Whether or not cost is a factor, there are some practical challenges to including the information on many lens designs and for some it would simply be impossible. The easiest solution is simply to drop the feature and keep everything consistent, at least for zooms.

It's true that primes could have markings - maybe they do - but my only prime is the Nifty and that has no markings at all. I don't know about the others. I'm sure that apart from the practical limitations Canon would answer that you have A-Dep as the modern day solution.

[Lowner]

Sounds like a partial reason. So why are they not on my 100-400? Crop camera users (I'm one myself) would get used to adapting the guides so thats a non issue. And I always found the "handful of apertures" enough.

I think Canon (and others) believed their own daft hype about automated depth of Focus systems when they were introduced back in the last days of film and are unwilling to reintroduce them for whatever reason.

I've used A-dep once only, just to say I had! I'd always assumed it was the same system as the one on my Eos3 and avoided it like the plague for that reason. I was actually pleasantly surprised at how simple to was to use on my 30D. No replacement for the even simpler engraved lines, but not too shabby.

[krb]

Sounds like a partial reason. So why are they not on my 100-400? Push pull on the AF lenses like the 100-400 or even the old EF 70-210 isn't the same as it was on the manual push pulls because the EF lenses have a separate focus ring. The manual one-touch lenses were truly one-touch where the out sleeve was moved in and out to zoom and the whole thing was twisted to focus.

[tdodd]

Here is a picture of a current Canon prime lens that does have DOF markings against the focus distance scale (again, not my picture, so a link only, I'm afraid)....

http://photography-on-the.net/forum/showpost.php?p=5819809&postcount=1

You can see how the focus covers a larger range as you go from f/4 to f/8 to f/11 to f/16.

[tonylong]

DOF markings would only be found on fixed focusing lenses, because they are for a set focal length. They are on 3 of my 4 primes.

They are guidelines, rather than pinpoint accurate, since the distance settings have gaps between and are increasingly compressed with more distance, so that the DOF will scale with the distance. We'd need an LCD continuous readout to get accuracy (which would work for zooms as well).

Anecdotally, I recently was shooting waterfalls with my 5D and 16-35 so was taking in pretty wide scenes with running water and foreground as well as background elements. I pushed things to an extreme because I wanted to ensure foreground-to-background sharpness as well as a slow shutter speed, so ISO 50 and f/22 were my chosen settings for some of the shots to give a 2" shutter speed. I was pretty pleased with the results, but I didn't have the time to really test out various aperture settings and hyperfocal lengths and whatnot.

Here's one of my shots, with the above settings @ 27mm: I used center point focus so the focal plane is in the center of the scene. Below is a crop from the foreground. When I print at, say, 12x18, with a light touch of output sharpening, the foreground comes out very crisp

http://www.pbase.com/tonylong/image/101785097/original.jpg

Crop from foreground:
http://www.pbase.com/tonylong/image/101785095/original.jpg