Hyperfocal distance - tutorial misleading?

[Downrigger]

AAAARRRRHHHHHGGGGG!!!!!!!!!!!!!

Go out and take some pictures.

I love a bit of Tech Talk as much as the next (wo)man but sometimes I just want to grab people and shout at them. Surely getting out a bit more and trying techniques by taking photographs - kind of the whole point of a camera - will produce more real world knowledge than frantically Googling it.

Please don't take this the wrong way - I'm just sitting here (not taking my advise by the way) and screaming at the iMac in exasperation and had to post something

Oh, for sure. No question that's the most important thing. I started this - really just to get some comments on how the experienced folks use/disuse DOF.

[Downrigger]

I'm at work - so I'm allowed to be wasting time!!!



Just a quick question, from someone who still doesn't fully understand. How can you calculate this circle of confusion, if you don't have all the information? I think I might be missing something - I got f22, I got 23mm, and I got FF camera. But we're still missing one variable, aren't we?

I like John's point that there's really only one plane of best focus... and the other thing is that the circle of confusion by which these DOF calculations are possible is meant to apply to an image expanded to 8"x10" - so that inserts some relativity and unknown in the whole business since one may not know in advance whether they're shooting for a wall size print or a 5x7.

[ajohnw]

I like John's point that there's really only one plane of best focus... and the other thing is that the circle of confusion by which these DOF calculations are possible is meant to apply to an image expanded to 8"x10" - so that inserts some relativity and unknown in the whole business since one may not know in advance whether they're shooting for a wall size print or a 5x7.

There can be a bit of a problem with that. If rhe circle of confusion is bigger than the detail that you want to show final size doesn't come into it so much. On the other hand given some level of detail a certain minimum final viewing size is needed to even show it. As a a for instance say a shot is taken of something with a tree in the background and on the full sized image detail can be seen on the leaves. On a PC screen that size might represent a 50in image or even more. When the image is reduced to say 10in the fact that there was detail on the leaves doesn't really matter as it can't be seen any more. The important thing there was that when the shot was taken the circle of confusion needed to be small enough not to wreck the fact that it is a tree with leaves on it at the final size - unless that is what some one wanted to do to some extent or the other. Maybe this illustrates the effect. Playing with a camera just after I bought it and didn't notice it was in manual focus. None of this image is in focus including the grass.

Reduced

Can't upload at the moment.

On the other hand you should be able to see that most of the time people do not worry about it too much and just shoot as there are far too many things to think about. Unless blur is wanted the best option is to over kill the depth of field aspect but not too over the top and guesstimate a 1/3 in focus point if you can. Geting a feel for that involves going out and taking some photo's. If concerned take several at different F stops.

John
-

[jcuknz]

Nothing to do with thread apart from it seems to have good DoF "What will dawn bring"
You should be able to see a brilliant example of photographic art but Tinypix is playing silly Bs

[Colin Southern]

I like John's point that there's really only one plane of best focus... and the other thing is that the circle of confusion by which these DOF calculations are possible is meant to apply to an image expanded to 8"x10" - so that inserts some relativity and unknown in the whole business since one may not know in advance whether they're shooting for a wall size print or a 5x7.

My strong suggestion is to simply forget about circles of confusion, 8 x 10" images, diffraction etc. In the real world 99.9% of non-photographers stand back and view an image as a whole -- so the bigger the image, the further back they stand (so questions like "how big an image can I print with an "X" megapixel camera become moot"). Same with diffraction; the image degradation that occurs due to an insufficient shutterspeed or too wide an aperture are much worse than the alternative of small high-ISO noise. Only with photographers is the minimum viewing distance limited only by the length of their noses!

In the real-world I'm always saying to people "Don't ruin the image trying to save the pixels".

[Downrigger]

Thanks - I've heard that before, Colin, from other site seniors - don't be afraid of high ISO speed noise or diffraction at high f's if they get you what you need. Not many see the noise that left after you clean up, and no one much cares about it.

[Colin Southern]

Thanks - I've heard that before, Colin, from other site seniors - don't be afraid of high ISO speed noise or diffraction at high f's if they get you what you need. Not many see the noise that left after you clean up, and no one much cares about it.

Exactly!

[ajohnw]

My strong suggestion is to simply forget about circles of confusion, 8 x 10" images, diffraction etc. In the real world 99.9% of non-photographers stand back and view an image as a whole -- so the bigger the image, the further back they stand (so questions like "how big an image can I print with an "X" megapixel camera become moot"). Same with diffraction; the image degradation that occurs due to an insufficient shutterspeed or too wide an aperture are much worse than the alternative of small high-ISO noise. Only with photographers is the minimum viewing distance limited only by the length of their noses!

In the real-world I'm always saying to people "Don't ruin the image trying to save the pixels".

I went that way Colin because it is a better approach to understand what is actually going on and the complications - also of use if some one wants to get the tape measure out. That really is the only situation where acceptable depth of field is critical and is a lot more difficult to do than so called creamy bokeh.

It's curious how often comments such as saving pixels crop up when circles of confusion and diffraction are mentioned when in real terms there isn't any direct relationship other than noting the effects are there. So far having bought a number of lens over many years diffraction in particular is completely irrelevant other than under very extreme circumstances..

If some one needs the sharpest shot they can get for some reason it will all depend on the gear they are using. The only guidance on that really is that what ever the aperture of the lens best results will usually be obtained in the F5.6/F8 region on fast lenses, even F1.2's and 1.4.s and if it's a slow lens probably when it's stopped down 1 to 1 1/2 stops. Still a problem though. Say it worked out at F8 that doesn't mean that the results would very noticeably worse at F11 or F16 or even F22. Much depends on what is in front of the camera and more importantly a particular lens - and the size it's viewed at when the subject is a problem in that respect. Here really the notion of circles of confusion can be useful - results are no good need a better lens. Diffraction limits are very miss leading.To obtain good images based on that the resolving power needs to be 2 to 3 times what diffraction suggests. This touches on why technical tests on camera lenses use so called MTF50. It's way short of the diffraction limit and is when contrast out of a lens has dropped to 50% of what it actually is. Again of little real interest to most photographers but on the other hand ff some one wants to produce this image of a mosquito head it really does matter. The person who took this isn't entirely happy about it by the way. It is a good attempt though.



I sometimes feel I am accused of holding views that I just don't have. Yes I pixel peep - why so I can decide just how much I can crop out of the images from this and that lens. I also notice where AF tends to lock on from this and that camera. I'm probably a little more astute at that than many. What it probably means in practice is having to close the lens down more or checking manually.

John
-

[Colin Southern]

All I can really say John is that people are free to concentrate on whatever aspects of photography they choose. Where I feel the need to chip in is where I see "newbies" who appear to be "majoring in minor things", whilst completely over-looking things that have a far bigger impact on their photography. They won't shoot over ISO 400 because of the "noise" - won't stop-down past F11 because of the "diffraction" - and then post something for c&c that has a hopelessly inadequate DoF and or camera shake/motion blur. Or they'll avoid zooms and buy 3 primes because "primes are sharper" and then present work that's completely unsharpened.

I guess I kinda see it as my role to lead them past the reviews and the lab tests out into the real world. If they have a specialised areas where things like CoC make a real difference then that's fine -- it's just that most of the time it doesn't -- and they'd get more "bang for their buck" addressing other areas first.

[ajohnw]

All I can really say John is that people are free to concentrate on whatever aspects of photography they choose. Where I feel the need to chip in is where I see "newbies" who appear to be "majoring in minor things", whilst completely over-looking things that have a far bigger impact on their photography. They won't shoot over ISO 400 because of the "noise" - won't stop-down past F11 because of the "diffraction" - and then post something for c&c that has a hopelessly inadequate DoF and or camera shake/motion blur. Or they'll avoid zooms and buy 3 primes because "primes are sharper" and then present work that's completely unsharpened.

I guess I kinda see it as my role to lead them past the reviews and the lab tests out into the real world. If they have a specialised areas where things like CoC make a real difference then that's fine -- it's just that most of the time it doesn't -- and they'd get more "bang for their buck" addressing other areas first.

Might seem strange to some but I agree 100% with that Colin. All I often post on subjects like this really is ok you have read the tutorial and this is what it means in practice. Diffraction is the worst aspect on the web and in general books. People mention it and from what I often see don't really understand what it means in practice - under 8% of the contrast that went in to the lens coming out of the back - truly awful pictures of any detail level that needs it. In practice it would be rather fuzzy too - if a camera lens could ever get there which I doubt.. It's another red herring often mentioned that is really there but of dubious use in practice. All people can do is take their camera and lens out and use it and find out what it can do. Same applies to depth of field. shooting into back light, trying to correct exposures based on the scene and a host of other things including how many pixels needed for an image of this size. Tutorials help, may be misleading in some cases and there is probably some one about who reckons taking 1 sec hand held exposures is easier than ones taken at 1/1000. There are also those about who will fudge results to show what they are advocating. Maybe best to stick to this site as from what I have seen it is reliable but no substitute for using the camera.

John
-

[Colin Southern]

Shot this yesterday at F32 ... as you can see, the diffraction is somewhat under-whelming!

[revi]

Shot this yesterday at F32 ... as you can see, the diffraction is somewhat under-whelming!

Next you show us the 44'' canvas (to be observed from about 10 cm, of course)?

[ajohnw]

Next you show us the 44'' canvas (to be observed from about 10 cm, of course)?

What ever you do don't show it to any pre teens as some can focus well down to 100mm and lots down to 150mm or so.

Ain't I a mine of information.

John
-

[revi]

I'm not a pre-teen, but rather myopic, so 10 cm or so isn't a problem :P

[ajohnw]

I'm not a pre-teen, but rather myopic, so 10 cm or so isn't a problem :P

It's a curious subject really. I reckon if some one looked at a PICTURE from even 250mm it wouldn't be all that big and even smaller from 10cm. Considerably to take in the whole image. Personally I feel PC screens prove the point about how big things can go, pixels and viewing distances all tied together and if people scroll round a full sized unreduced image they can get a fair idea just how big an image can be but not from 10cm.

As a for instance I stuck this one on the web because some reckoned the original kit lens that came with an e-pl1 was no good. Hope to add it as a link rather than an image as it's rather big. Open it and click on it and get the full view. My OMD E-M5 will go even bigger. More pixels. Past a sensible limit really for a sensor of that size. JPG's however appear to improve things on most shots..

http://backup.cambridgeincolour.com/...98480/original

John
-

[Colin Southern]

Next you show us the 44'' canvas (to be observed from about 10 cm, of course)?

Not printed on canvas yet ... however ...

... if you'd like to buy a copy I'd be more than happy to print & spray one for you

[rambler4466]

Another over-rated problem is noise. How many times have you said, "wow, what a great shot, but look at all that noise in the shadows."?? Have you actually seen a great photo ruined by noise?

Here is another solution to focusing suggested by the professional, Jim Zuckerman, Pro Secrets, p. 132.

Divide the focal length you are using by 3. Example, 18mm divided by 3 = 6. Focus at 6 feet with the lens at the smallest aperture and you will have the most DOF. The catch,....that works with a full frame camera, but with a cropped sensor camera, you have to multiple by 1.5 or whatever. Example 12mm = 18mm on a full frame, divide by 3 =6!
Don't get bogged down in the math, but at least your guesses will be educated ones!

[Downrigger]

Many great comments - thank you all. Anyway, DOF, it's a great thing, isn't it? I got a good bit here at f13 (I'm at or beyond hyperfocal, apparently, since the horizon's pretty sharp). The interesting thing is that it seems our vision has nothing like this DOF, yet images with lots more of it than we have can engage us.

[jcuknz]

Assuming one has good eyesight one has as much DoF as any camera set-up and probably more becuase the eye never looks at the whole picture even when it is reduced in size on our monitor but is sampling small areas to assemble an impression with perifcial vision of the whole message... I'll repeat myself and suggest you only see the area of a single key of the keyboard perhaps ten inches in front/below your head. I gather the exception to this is the speed reader.