Focus shift on D850, Z6 etc

Whatever.

 

To achieve a single shot like Ilkka's ice shots by aperture alone is impossible.

 

The only way to do that is camera movements, and to maximise their possibilities, utilising the Principle is the way to go.

 

The former has insufficient DoF.

 

The latter does have sufficient DoF.

 

At the same aperture.

It would be even better if a camera was designed so the sensor can tilt,

this is an error, Scheimpflug applies only to the lense, on a LF camera you see a clear difference between tilting front standard or rear, rear changes perspective^^

Scheimpflug determined, with Carpentier, the geometric relationship between the 3 planes. The Subject Plane, the Lens Plane and the Image Plain.

 

Assuming the Subject Plane is fixed, be it a building or the landscape, the other 2 can be rotated/moved* in relation to eachother to manipulate the Plane of Focus. In the typical landscape shot this is often done to align the Subject Plane to the Plane of Focus to achieve near to far sharpness that is not possible by use of a small aperture alone. This also allows the use of the 'best' aperture, regarding IQ, so there is no need to use a very small aperture that will induce diffraction limiting.

 

If you rotate the Image Plane the perspective does indeed change, but Scheimpflug still applies.

 

Assuming the camera can determine a depth-map of the subject, it should be pretty simple to shift & rotate the sensor to get everything just so. Tilting the lens may be more mechanically interesting!

 

* This is where Shift is important. It doesn't change the overall geometry, but it does move the sensor within the image circle that is most often displaced by lens rotation.

this is an error, Scheimpflug applies only to the lense, on a LF camera you see a clear difference between tilting front standard or rear, rear changes perspective^^

I don't see the point of this argument. With any camera the back can be tilted or swivelled. It just involves tilting/swivelling the whole camera when it has a rigid body. And then the lens needs to be tilted/swivelled and shifted to acheive whatever geometry is needed. So with sufficient front (lens) movement available, back movement becomes irrelevant. You just use the tripod head instead.

 

Conversely, the camera angle can be used to compensate for any back movement. Thus making image-plane (sensor) tilt/swivel equivalent to lens tilt/swivel.

 

Practically the full extent of both front and back movements have been applied to my DeVere monorail here, but a simple change of angle has left it being a straightforward 'box' camera. With back vertical.

 

It's a bit like those daft S-shaped hifi tone arms of the 1970s. All they did was alter the angle of the pickup cartridge WRT the arm pivot. Whereas a straight tube with an angled head would give exactly the same geometry.

Edited March 18, 2021 by rodeo_joe|1

It's a bit like those daft S-shaped hifi tone arms of the 1970s. All they did was alter the angle of the pickup cartridge WRT the arm pivot.

I remember seeing one of those and being very impressed, as a shape, but never knew why they existed. Nice to know after, umm, all this time!

Important to note, still, that movements do not maximize DoF.

Indeed, I often use tilt to photograph ice in sea coast and creeks; without it even at f/11 the background and extreme foreground tend to be less sharp. I use it especially when I want to emphasize foreground and place the camera at a relatively low level. I enjoy the movements and I nowadays can't even properly think about landscape without them. However, there is a learning curve to using them and I don't necessarily recommend this technique as a lot of people are not willing to spend 15 min setting up a shot. I enjoy the process and feel that it gives an edge, a kind of calmness to the images. I also use focus stacking but it's not a good choice for situations where the subject is in motion, such as some of the ice sheets can be, or the flowing water itself. One can run into difficulties creating a consistent result from multiple images in such cases, and masking may be needed. I generally prefer the lens movements but in the case of a subject that does not move, I use focus stacking, and sometimes both. I would say both techniques have a learning curve, as when I started out I was happy with maybe one in ten of the images I tried stacking with, the others looked somehow broken. Similarly at first my tilted plane was slightly off more often than not and even with shift I often overdid it on buildings. It's better to avoid extreme shifts as it can look like the building is leaning over. In case of a really tall building I nowdays sometimes purposefully leave a little keystoning (but not the full amount from an unshifted lens) and prefer the results that way. Anyway, both techniques require extra work but can give results that differentiate from the common normal.

OK, so i want to photograph a meadow full of flowers. Pop on a 35mm lens at f2, focus on the midground and go click. The midground is sharp but the near and far areas are soft......too shallow a DoF. Shot A.

 

Rack it round to f32 and all is kinda sharp, with enough DoF, but diffraction has killed it. Shot B

 

Rack it back to f2 and apply the correct camera movements and the shot comes out just fine. Shot C.

 

Aperture hasn't given me a shot with enough apparent DoF, Movements have.

 

Now maybe DoF isn't quite the term to use here, what is?.. but the result is obvious.

 

If I show Shot A, and Shot C to a photographer, they will happily say Shot A doesn't have enough DoF, whereas Shot C does.

 

Same aperture.

In case of a really tall building I nowdays sometimes purposefully leave a little keystoning (but not the full amount from an unshifted lens) and prefer the results that way.

Corrected but unattractive! Been there with film. Can look very fake.

 

I even de-corrected a corrected shot once by tilting the enlarger head until it looked sensible again.

OK, so i want to photograph a meadow full of flowers. Pop on a 35mm lens at f2, focus on the midground and go click. The midground is sharp but the near and far areas are soft......too shallow a DoF. Shot A.

 

Rack it round to f32 and all is kinda sharp, with enough DoF, but diffraction has killed it. Shot B

 

Rack it back to f2 and apply the correct camera movements and the shot comes out just fine. Shot C.

 

Aperture hasn't given me a shot with enough apparent DoF, Movements have.

 

Now maybe DoF isn't quite the term to use here, what is?.. but the result is obvious.

 

If I show Shot A, and Shot C to a photographer, they will happily say Shot A doesn't have enough DoF, whereas Shot C does.

 

Same aperture.

Mike, you keep confusing being in focus with DoF.

 

Could you find other people who could do the same (the basis of your argument)? Perhaps. They then would also need educating.

I won't bother. No need.

They then would also need educating.

 

Anyone who has a Tilt and Shift Lens or a camera with movements is just fine thanks.

 

If they have a T&S lens and are unsure or got here via a websearch, I found this to be handy.

 

Using Tilt-Shift Lenses to Control Depth of Field

I once tried to 'map' or graph what happens to depth-of-field with a tilted plane. Tough one!

 

We know that D-o-F increases almost exponentially with subject distance. So it seems logical that the tilted D-o-F should also do so. Resulting in a depth that has a curved shape in profile.

 

However, the calculation of the vector of D-o-F in the vertical or sideways axis isn't straightforward, and I never did crack it properly. But it definitely varies with distance from the camera. Unlike a 'normal' D-o-F that has fixed linear planes where the C-o-C reaches a specified cutoff diameter.

 

Not accounting for lens curvature of field or other imponderables!:eek::confused:

Trying what used to be called 'anti-Scheimpflug' to mimimise the 'sharp' bits is a fun game, all-be-it briefly.

 

Making 1/4 of a line of text sharp on a page of newspaper or just 2 eyelashes for a 3/4 facial portrait is bizarre, but gets old very quickly!

 

It looks very much like a severe version of the 'Miniature' effect available in some cameras and apps.

Trying what used to be called 'anti-Scheimpflug' to mimimise the 'sharp' bits is a fun game, all-be-it briefly.

 

Making 1/4 of a line of text sharp on a page of newspaper or just 2 eyelashes for a 3/4 facial portrait is bizarre, but gets old very quickly!

 

It looks very much like a severe version of the 'Miniature' effect available in some cameras and apps.

That's because it is.

It looks very much like a severe version of the 'Miniature' effect available in some cameras and apps.

 

That's because it is.

 

...and all without an aperture or movements too....:cool:

...and all without an aperture or movements too....:cool:

Well, no. With movements...

With movements...

Your Smart Phone has movements? Wow!

Your Smart Phone has movements? Wow!

Who said anything about smart phones, Mike?

OK, so what has movements?

 

I said its in some cameras and as an App. The later work on Smart Phones.

The Miniature Effect works with cameras with no movements as its purely a Digital filter.

Guess the same with smart phones, and most of them dont have an aperture either.

So they produce the visual impression of a very narrow DoF.

So nothing to do with actual optical geometry.

OK, so what has movements?

 

I said its in some cameras and as an App. The later work on Smart Phones.

The Miniature Effect works with cameras with no movements as its purely a Digital filter.

Guess the same with smart phones, and most of them dont have an aperture either.

So they produce the visual impression of a very narrow DoF.

So nothing to do with actual optical geometry.

O.k. apps. Still irrelevant and wrong. Visual effects created by apps also do nothing to DoF, unless those apps know what in a given area of an image is at what relative z-axis distance to what else in that area. Which these apps do not. No deoth information available nor used. So again, or still, it is a matter of being sharp or not, not of DoF.

No deoth information available nor used

Whatever.

 

...and you might mean depth?

 

Many new 'High End' Smart Phones can now produce a depth map by using 2 camera separated by a few cms.

 

This allows software to blur what it now knows to be the background.... and keep the portrait, or whatever, sharp.

 

Clever...:cool:

Edited March 19, 2021 by mike_halliwell

Many new 'High End' Smart Phones can now produce a depth map by using 2 camera separated by a few cms.

IPhone Pro Max 11 (the one I have) has 3 cameras/lenses that communicate with each other: 13mm f/2.4, 26mm f/1.8, 52mm f/2.0. One can use the focus-wheel to select focal length, and it can also macro. One can also select the intensity of light , even at daytime, such as lighting up only a select area and let the outer area gradually darken and blur. Normally I only use the obvious features, I am sure it can do much more. I believe IPhone 12 is more enhanced - though it also has 3 cameras.

52mm f/2.0

You gotta wonder how you can get a 50mm lens into a phone body just 8.1mm thick......:)

It's not really 52 mm but something like 6 mm as far I could find out by googling.

Oh, it means 50mm equivalent!

 

So, a Normal or Standard Lens?