The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Focusing and other musings.

Dogs857

New member
Isn't it always the way.
I finally get my MF gear together again after a long absence since I had to sell off my Arca and IQ180 at the start of the year, and the wet season has hit in force. I've had nothing but pouring rain and strong winds for the last week. Now there is a couple of cyclones brewing as well. Life is great in FNQ.

Anyway I have been doing some reading about hyper focal focusing since this used to be my favourite way to do things. I had all the calculators and tried so many different settings but always felt a bit let down. I could never seem to really nail the right settings.

A very helpful post by Wayne Fox in another thread (sorry no idea how to link to it at the moment) had me wanting to try another way. Basically just ditch the calculators and take many test shots at different apertures and note the setting where acceptable sharpness is lost. That then becomes your hyper focal distance.

Brilliant. But then while trawling the Internet I came across this article. http://http://www.trenholm.org/hmmerk/DOFR.html This hypothesis is that at infinity it is easy to calculate what should and should not be resolved in your photo.

This article makes a lot of sense to me and I am just wondering what others think of it. I understand it was written in the days of film but it should surely translate to digital.

Now I just need the rain to stop so I can try it out.

March is looking good :(
 
Last edited:

torger

Active member
My own view on hyperfocal focusing and other aspects is at
How to focus a landscape scene

Anyway, it's related to the concept of depth of field which is always debated over and over again. Anyone knows that for large panoramic landscape prints which you can step close the standard recommendations falls apart as the viewer is virtually nosing the print. In that case you want to make the picture as sharp as needed where it counts.

In many landscape scenes the inifinty has finer details than the foreground, so sharpness is more important there. On the other hand, sharpening software can make a good job so maybe you don't need maximum sharpness there and can give a bit more to the foreground.

A tradeoff I like, which is actually resolution independent, is to set the CoC to the airy disk diameter, ie dependent on aperture, and aperture is usually f/11 or f/16, probably f/11 on IQ180. I think MF folks often shoot at a too large apertures which results in aliasing and thus false colors, I prefer to have a little bit more diffraction and less aliasing and get more depth of field as a bonus.

Anyway the rationale with aperture-dependent CoC is that with smaller aperture diffraction limits resolution and thus the CoC should be related to that. With my CoC=Airy it will be sharper in the plane of focus, but the edge of this DoF will respond well to sharpening and in a print be extremely hard to differ from the focus plane, even when nosing.

edit: 2x radius is 1x diameter... :)
 
Last edited:

Paul2660

Well-known member
I never really gave much attention to hyper focal distances until I purchased a tech solution. I quickly determined the range for each wide I have and niw never go back to the Arca cards unless I have the 90 or 120mm out as they such a narrow DOF.

To determine each approximate hyper focal I used a tape measure and a target that I could focus on.

The process is tedious but well worth it as it saves a ton of time in the field.

Paul
 

Dogs857

New member
Thank you both.

Torger that is a great article and I may need to read it a few more times to fully digest everything in it. I was hoping you would chime in. I love detail, and you provide that in spades.

Paul

I agree entirely, test test test. I have a Cambo now so the focusing is a little less precise but the theory would still be the same. I think I will give both methods a try and see which works better for me.
 

miska

Member
I'm a bit of a tech cam newbie, but a strategy that has been working well for me, is... not to focus :)
So at the beginning, I have calibrated my infinity focus (and on my Factum, it is pretty close to the zero setting for each lens). Now, it's basically set there, and if I need a bit more detail on a an area closer, I just close the aperture a bit more.
I just need to do the same with ~1 degree of forward tilt, and I think I'll get even better results (but the optimal setting will not be at infinity anymore)

I do have a Disto 5 to really focus on something else than infinity, but for the moment, I haven't used it. I guess my subject matters have been until now "grand landscapes", where the mountains in the background need the most detail.
 

Dogs857

New member
That is what this article I found states.

Keep the lens at infinity and stop down for your depth of field, then you can calculate what will and won't be in focus. He also gives a decent calculation for diffraction so you can judge how your infinity may resolve if you stop down too far.

I find all this quite interesting as I have always had SLR / DLSR's up until this point. They were all autofocus and I never really gave much thought to how things work. For Landscapes I used to do the focus into 1/3 of the scene thing and was happy with results.

It was only after I started taking things seriously that I started paying much closer attention to how things were actually being taken, and how they looked when you printed them out.

This is why I loved having a tech camera, it made me think about my photography so much more than I was.
 

Jamgolf

Member
Two pdf books "The INs and OUTs of FOCUS" and "FOCUSING the VIEW CAMERA" authored by Harold M. Merklinger, also the author of article that 'Dogs' referenced can be downloaded here:

Download The INs and OUTs of FOCUS

I've read The INs and OUTs of FOCUS and found it to be informative. I need to read it again though to really absorb the material.
 
Last edited:

Stuart Richardson

Active member
This is probably a bit against the grain, and also not always the case, but I think it is far more important to focus on the most important part of the image...and I mean that both literally and metaphorically. Not every landscape image needs perfect infinity to foreground sharpness, and understanding to what you are trying to draw attention is more important. Hyperfocal focusing can allow one to ignore this choice, which is not good! With medium format digital and good lenses, you are only going to have one area in perfect focus, and it should be the primary subject of the photo.

Of course, there are going to be situations where it is better to focus slightly off the main subject so that you can extend the area of acceptable sharpness, but personally I think it is detrimental to be too formulaic about it... I say this from both an artistic and a technical standpoint. I am a printer and a landscape photographer. I use the S series and print large, and the areas of maximum sharpness are clearly visible in large prints, even when using wider lenses at medium to small apertures...when the system is good enough and the prints large enough, you will see the transition, and it can be distracting to have the main conceptual focal point be different from the main optical focal point.
 

miska

Member
With medium format digital and good lenses, you are only going to have one area in perfect focus, and it should be the primary subject of the photo.
I think this is quite interesting. So take the example of a (smallish) barn in front of a grand mountainous landscape. The barn is not big in the frame - the mountains are.
Without the barn, the picture would look dull - it's just a mountain after all.
So in my mind, the "subject" is the barn, it makes the picture. One would have the tendency to focus on that. It is the subject after all.
But then most of the frame is out of focus - since the barn occupies only a small area.

So I would focus on the mountain (at infinity), and try to keep the barn "pretty close to focus", but not in focus.

It's a tough choice. I don't want to start focus stacking. Tilt could be a solution to have "everything" in focus...
 

Stuart Richardson

Active member
Miska -- in this case, I would choose the barn. That is the little detail which will draw the eye, and more likely to be scrutinized. The mountains themselves will not likely have a ton of detail, or at least not that would not be adequately served by photographing at a medium aperture...their outline will likely still be crisp and natural looking. Keep in mind that our eye is also a lens and does not have infinite depth of field...when you are focusing on something the objects in the periphery are not fully in focus! It is not a pinhole...

I think it is actually kind of rare to have a composition where if you focus on the main subject, the rest of the composition is not well served just by using the right lens and aperture...
 

torger

Active member
It depends a bit on style, I'm more of the f/64 type I guess. I use tilt a lot, and sometimes even swing, but not as much as 8x10" folks do.

f/64 can sound like a really small aperture but if the reference was 8x10" (?) , that's only about f/11 on a medium format digital camera. The extended and optimized DoF with tilt and swing.

I agree though that it is quite rare that you actually need hyperfocal, focusing on a main subject works well most of the time, and that's good for me because I focus on ground glass and have no hyperfocal presets. With tilt in the arsenal, hyperfocal becomes even more rare (although you can have some convenient presets involving tilt and hyperfocal if you happen to have a high precision focusing ring).

I'm an amateur and don't shoot a huge volume of images, I just scanned to find a hyperfocal example image from the last year and well I did not find a single one, but I understand if you have a preset on the camera you can use it more often. It does happen though that a scene of mine don't really have a main subject, it's more about composition and atmosphere than pointing at some specific. In intimate landscape scenes it quite often happens that I don't focus on anything particular, but on the other hand in those cases infinity is rarely visible so it's not hyperfocal anyway.

Attached one such example. I guess you could argue that the fallen tree closest in the foreground is the main subject, but my intention of the scene was more global. I don't really remember where I placed focus but I think it's between the fallen tree and the closest standing tree in center, and letting the background become slightly soft, shot at f/16.
 

Stuart Richardson

Active member
Trees are hard, it's true! Even then, I tend to just pick a main point and then stick to it, but I know what you mean. For this image for example, I just used my judgement and chose the second closest tree and shot at f16. It was the 70mm S lens, so depth of field is not huge even at f/16. This came out very well, but the closest tree is slightly out of focus when you look at the huge print. But in this case, the main point of interest for me was the path flowing in the woods, so I wanted the area in the middle in sharper focus. The background and foreground are both less sharp than the middle, but in this image that is fine. You could make the point that I am mentally focusing hyperfocally here, but that is one reason to include it! There are always exceptions to every rule!
 

jlm

Workshop Member
always makes the logical mind chuckle:
"There are always exceptions to every rule"
 

Dogs857

New member
f/64 can sound like a really small aperture but if the reference was 8x10" (?) , that's only about f/11 on a medium format digital camera. The extended and optimized DoF with tilt and swing.
Can you please expand on this for me?? I have heard this said before but I always thought aperture was the same regardless of format size.
 

MGrayson

Subscriber and Workshop Member
:lecture:
For a fixed f-ratio, DoF decreases as you make the system bigger. You wouldn't think so, because the bigger sensor/film needs less enlargement, and that allows a larger CoC, but that effect scales only linearly with size.

What is surprising is that the size of an out of focus point image grows *quadratically* with system size. You get one factor from the bigger lens opening - rays passing through opposite sides have a bigger angular separation, and another factor from the increased aperture to sensor distance - the rays have more time to diverge, making a bigger blob on the sensor.

So in the final print or viewed image, OOF point sources make disks that scale linearly with camera size. To get the same DoF from the bigger camera, you need a smaller aperture. 8x10 has a MUCH bigger sensor, so you need a really small aperture to get things in focus, hence f/64.

I love this calculation (yes, I am a Mathematician), but it is heresy to allude to it in certain fora. People are strange.

Matt
 
Last edited:

MGrayson

Subscriber and Workshop Member
Although I am a Mathematician, I don't like formulas. Or, at least, I find pictures more convincing than formulas, and don't really believe a result until I can see it. I have always wanted to write up this explanation that, given a distance to the subject, it is the physical size of the lens aperture alone that determines the size of an OOF disk relative to other elements in the picture - the size, if not the quality, of the bokeh.

Here's the setup as seen from above:



The image with infinite DoF looks like this:



Now suppose we want that OOF light to make a disk of a certain size:



Let's see what happens if we move the light a bit behind the subject



so that the OOF disk is just barely visible:



But to be visible at all, the rays from the light must still enter the lens



If the lens is smaller than that, no light gets in, so the OOF disk can't be visible. We didn't use anything about the focal length, sensor size, film vs. digital - all that matters is the size of the opening.

Yes, if I move the lens further from the subject, the required lens opening grows. So if you change focal lengths but keep the subject size the same, you have to change the opening as well, and THAT leads to a constant f-ratio. A 200mm at f/4 has the same OOF look (to first order) as a 100mm at f/4 if the 100mm lens is half the distance to the subject.

Phew - I feel better now.

--Matt
 

jlm

Workshop Member
the actual opening measures to be the focal length divided by the f-number.

so f/64 is a smaller opening than f/32, assuming f is the same.

as marc points out it is the actual opening size that is relevant to DOF, however, the reason f/64 is required for 8x10 is because typically the focal lengths used are longer as the format gets larger.
 
Top