Landscapes, Tilt, Drop, And Zones Of Focus. What's All This Then?

[68lbs]

I've been doing quite a bit of background reading around my chosen hobby over the winter months. In JC's First Light (my latest read), I keep on coming across phrases like "applied a little tilt", "deep lens drop" and "allowing the photographer to focus on elements at different distances simultaneously, providing they intersect the zone of focus".

What's this all about?

I understand the equipment is different, and that the tilt and drop thing is something to do with correcting perspective and converging verticals. But how do you have the ability to focus on more than one point? And how does that work with DoF and aperture?

The comments about 'tilt' remind me of the Canon TSE Tilt Shift lenses. Is this the same thing? And if it is, why does nobody really talk about tilt shift lenses on here? Why are they not more widely used? I have found myself starting to correct lens distortion in CS3 lately, but you do end up losing part of the image as you re-crop, which can b****r up the composition.

 

[melevo]

you have got me there????? intresting thread.....waiting to see answers

 

[raathistle]

Its refering to primrily large format equipment. Essentially you have to change the orrientation of the film plane commpared to the lens to get focus which is not parallel

Scheimpflug rule

Figure 2 shows you the basics of it. It means you can have a subject that moves into the picture in focus without the need for really tiny apertures

 

[Martyn...]

Same effect is possible with the specialist tilt and shift lenses available for DSLR's (at a specialist price)

 

[photon]

The comments about 'tilt' remind me of the Canon TSE Tilt Shift lenses. Is this the same thing? And if it is, why does nobody really talk about tilt shift lenses on here? Why are they not more widely used? I have found myself starting to correct lens distortion in CS3 lately, but you do end up losing part of the image as you re-crop, which can b****r up the composition.

Because people are ignorant (even to the extent of thinking that everything has to come from Canikon to be worth using), and we have our eyes "educated" by TV, film, photos and even computer games.

Look at the "then & now" comparisons in your local paper, or old postcards on eBay, and ask yourself what advantage, other than colour, the contemporary photo has? Sure, it's far easier to take a snap now, but 99.9% of them would be rejected by the Images of England project and any future archive about our built environment.

 

[Blapto]

By multiple areas in focus I think he means using tilt to move the focus plane such that it intersects some areas, but not others. Think of the plane as a wedge that you can make wider (stop down), narrower (open up), rotate (swing) and angle (tilt). You can then use rise and fall to change perspective.

 

[Les McLean]

This is the camera that JC uses (Ebony SW45), the arrows indicate the horizontal & vertical shift functions

 

[happy haggis]

Tilt and Shift lenses are a niche market, which is why they warrant little comment on photography forums. Given that landscapes, wildlife, portraits and sport are the most popular genres, there is little call for this type of lens. Their primary use is for architectural work and although they can add an extra dimension to some landscapes, it comes at a cost. The new Tilt and Shift lenses just announced by Canon have a SRP of around £2000+.
As for the benefits of modern cameras, I would respectfully suggest that they enable far more people to go out and enjoy the hobby. Whether or not they are snaps or quality images, the important thing is that we get pleasure from capturing the moment.

 

[Blapto]

I'd argue that the need for tilt/shift is as great in landscapes as it is in architecture. Perspective control is probably more needed in architecture though, or rather needed more often.

 

[jerry12953]

Never having used large format, I can't really imagine what tilt and shift actually DO...

Please enlighten me someone, but if you can get everything in focus using a narrow aperture on 35mm (or equivalent) what is the point of T/S?

 

[joxby]

Same effect is possible with the specialist tilt and shift lenses available for DSLR's (at a specialist price)

You can't do this with a tilt/shift lens, they don't have the same degree and range of movements as large format.

I've seen a scape like this shot with a Shen Hao 5x4, a fence runs down the middle, sharp from camera to horizon, but oof either side....

 

[joxby]

Never having used large format, I can't really imagine what tilt and shift actually DO...

Please enlighten me someone, but if you can get everything in focus using a narrow aperture on 35mm (or equivalent) what is the point of T/S?

You can get everything in focus at f2.8 with a tilt shift lens, depending on the subject.

Luminous Landscape T/S

 

[Martyn...]

You can't do this with a tilt/shift lens, they don't have the same degree and range of movements as large format.


I've seen a scape like this shot with a Shen Hao 5x4, a fence runs down the middle, sharp from camera to horizon, but oof either side....

That is the advantage of the large format, over DSLR / 35mm.

A strange image to view, the way the fence changes focus each side as it recedes from the camera.

There will of course be software that can attempt to copy the effect.

 

[joxby]

That is the advantage of the large format, over DSLR / 35mm.

A strange image to view, the way the fence changes focus each side as it recedes from the camera.

There will of course be software that can attempt to copy the effect.

yep, this shot is sharp in a straight line, just as the railings bend sharp carries straight on to the rocks.
I forgot to mention in the scape, the fence was "S" shaped into the distance, sharp was "S" shaped........very weird, can't imagine the movements for that..:shrug:

 

[jerry12953]

So you've got all these movements on your Fuji Pro Disposable Mk 2, then.....

Seriously though, why would anyone want to get sharpness from foreground to infinity at f2.8? Isn't sharpness - full stop - the objective for most people?

 

[Blapto]

No. The very best landscape photographers know how to use depth of field effectively.

Focus can be used to lead the eye more effectively. Also, the human eye is used to seeing things at a distance as being slightly blurry, and some camera lenses out resolve the eye. In order to make a scene look more natural for given print sizes some photographers will adjust focus so that it looks more like what the eye would see if you looked at a scene. It's all rather difficult and something that comes with years and years of experience. Us mere mortals have to get much more practice.

 

[jerry12953]

No. The very best landscape photographers know how to use depth of field effectively.

Focus can be used to lead the eye more effectively. Also, the human eye is used to seeing things at a distance as being slightly blurry, and some camera lenses out resolve the eye. In order to make a scene look more natural for given print sizes some photographers will adjust focus so that it looks more like what the eye would see if you looked at a scene. It's all rather difficult and something that comes with years and years of experience. Us mere mortals have to get much more practice.

I may be a bit of a cynic, but apart from the quality, which is undeniable, I wonder if there's just a little bit of elitism among those who use large format? And do editors see a big trannie and think, well this must be good then, its on large format!

Joe Cornish produces some stunning results with his, but there's another who uses LF, and gets his stuff published, whose pictures seem to me to be little more than snaps....the guy seems to have no idea about composition, or lighting, or filtration...

 

[Blapto]

Probably, but this isn't really a thread for that. There's huge elitism in every aspect of art.

 

[Steve Smith]

Front rise is mainly used in architecture to keep vertical lines of buildings parallel. If you point a normal camera up towards the middle of a building (to get it all in frame) the sides will converge and the building will appear smaller at the top than the bottom. You don't need a camera to see this, it's how all buildings appear when you look at them from ground level.

With a view camera you set it up so that the film plane is vertically level. That way the vertical sides of buildings will also appear level. Without any other ajustments though, the camera will be looking straight ahead and the middle of the image will be the area of building at the same height as the camera. In order to make the camera 'look upwards' the lens is moved up (Front Rise) so the film sees a different part of the image that the lens is projecting.
For this to work, the lens needs to project a greater diameter of image than just the film area. This is known as the image circle.

Front tilt is mainly used in landscape and works on the relationship between the position of a lens relative to the film and the area which this brings into focus.
If a lens is set in its normal position at its focal length (or infinity position) then generally everything from about 50' onwards is in focus. If you want to focus on something closer e.g. 8' then the lens is moved further away from the film to get the image in focus.
If you have a composition with a mountain in the distance occupying the top of the frame and a plant in the foreground occupying the lower part of the frame then it is possible to get them both in focus using front tilt.

If the lens is tilted forwards then the distance from the lens to the lower part of the film is shorter than that from the lens to the upper part of the film.

Remember that the image is actually projected upside down on the film and you will see that the lower part of the film is focused at a greater length than the upper part of the film.

By balancing the amount of tilt and the general focus you can get both the distant mountain tops and the nearby plant in sharp focus.

You cannot always get everything in sharp focus just by stopping down. A lens stopped down too far can suffer from diffraction effects which actually reduce the sharpness of an image and the depth of field is inversly proportional to (or inversly proportional to the square of) the image size. The larger the format, the less depth of field you get. That is another reason why these rise, tilt, swing, etc. tricks are useful to the large format photographer.



Steve.

 

[jerry12953]

Steve,

Thanks for that explanation; Im sure the OP will find it helpful as well.

I'm interested in whether you can now get the type of results that LF photographers get using tilt/shift etc, on smaller formats, eg full frame digital.

I understand that converging verticals can be corrected using front rise; it can also be done in software now. Other movements, it seems from your post, can be used to get increased depth of field.

I'm not trying to make a point here, and I've never used LF myself, but would there be anything other than the sheer size of the transparency which can't now be replicated in digital?

Hope nobody thinks I've hijacked this, but it seems to lead on naturally from the OP.

 

[Steve Smith]

would there be anything other than the sheer size of the transparency which can't now be replicated in digital?

Eccenticity of the operator will never be repeatable in digital!


The advantages of the large format camera are: superior image quality due to the area of film and better dynamic range; much greater range of swing, tilt, shift, rise, etc. than can be obtained on any SLR tilt/shift lens and the fact that you can view the image as you are adjusting the settings so you can see exactly what you will be getting on the ground glass. This would be very difficult to evaluate on a tiny SLR viewfinder.


Steve.

 

[68lbs]

Thanks for that explanation; Im sure the OP will find it helpful as well.

Most helpful... and I'm very grateful for someone taking the time to post such detail. As for threadjacking - feel free to divert. It's all knowledge and I am feeling sponge-like.

 

[Blapto]

Correcting for perspective in software always results in cropping the image, so it's not really desirable.

As far as I know, 35mm Tilt/Shift lenses only work along one axis, so you can emulate tilt or swing, but not both.

 

[joxby]

I don't think software correction does a great job anyway, depends on how much correction is needed, if its just very a minor say line straighten, its just about acceptable, anything more and things start getting fuzzy.
Better to correct with a TS lens or movements at shoot time.

 

[Rodintee]

Front rise is mainly used in architecture to keep vertical lines of buildings parallel. If you point a normal camera up towards the middle of a building (to get it all in frame) the sides will converge and the building will appear smaller at the top than the bottom. You don't need a camera to see this, it's how all buildings appear when you look at them from ground level.

With a view camera you set it up so that the film plane is vertically level. That way the vertical sides of buildings will also appear level. Without any other ajustments though, the camera will be looking straight ahead and the middle of the image will be the area of building at the same height as the camera. In order to make the camera 'look upwards' the lens is moved up (Front Rise) so the film sees a different part of the image that the lens is projecting.
For this to work, the lens needs to project a greater diameter of image than just the film area. This is known as the image circle.

Front tilt is mainly used in landscape and works on the relationship between the position of a lens relative to the film and the area which this brings into focus.
If a lens is set in its normal position at its focal length (or infinity position) then generally everything from about 50' onwards is in focus. If you want to focus on something closer e.g. 8' then the lens is moved further away from the film to get the image in focus.
If you have a composition with a mountain in the distance occupying the top of the frame and a plant in the foreground occupying the lower part of the frame then it is possible to get them both in focus using front tilt.

If the lens is tilted forwards then the distance from the lens to the lower part of the film is shorter than that from the lens to the upper part of the film.

Remember that the image is actually projected upside down on the film and you will see that the lower part of the film is focused at a greater length than the upper part of the film.

By balancing the amount of tilt and the general focus you can get both the distant mountain tops and the nearby plant in sharp focus.

You cannot always get everything in sharp focus just by stopping down. A lens stopped down too far can suffer from diffraction effects which actually reduce the sharpness of an image and the depth of field is inversly proportional to (or inversly proportional to the square of) the image size. The larger the format, the less depth of field you get. That is another reason why these rise, tilt, swing, etc. tricks are useful to the large format photographer.



Steve.

Errrr OK.....

Great post !

 

[springtide]

Front rise is mainly used in architecture to keep vertical lines of buildings parallel. If you point a normal camera up towards the middle of a building (to get it all in frame) the sides will converge and the building will appear smaller at the top than the bottom. You don't need a camera to see this, it's how all buildings appear when you look at them from ground level.

With a view camera you set it up so that the film plane is vertically level. That way the vertical sides of buildings will also appear level. Without any other ajustments though, the camera will be looking straight ahead and the middle of the image will be the area of building at the same height as the camera. In order to make the camera 'look upwards' the lens is moved up (Front Rise) so the film sees a different part of the image that the lens is projecting.
For this to work, the lens needs to project a greater diameter of image than just the film area. This is known as the image circle.

Front tilt is mainly used in landscape and works on the relationship between the position of a lens relative to the film and the area which this brings into focus.
If a lens is set in its normal position at its focal length (or infinity position) then generally everything from about 50' onwards is in focus. If you want to focus on something closer e.g. 8' then the lens is moved further away from the film to get the image in focus.
If you have a composition with a mountain in the distance occupying the top of the frame and a plant in the foreground occupying the lower part of the frame then it is possible to get them both in focus using front tilt.

If the lens is tilted forwards then the distance from the lens to the lower part of the film is shorter than that from the lens to the upper part of the film.

Remember that the image is actually projected upside down on the film and you will see that the lower part of the film is focused at a greater length than the upper part of the film.

By balancing the amount of tilt and the general focus you can get both the distant mountain tops and the nearby plant in sharp focus.

You cannot always get everything in sharp focus just by stopping down. A lens stopped down too far can suffer from diffraction effects which actually reduce the sharpness of an image and the depth of field is inversly proportional to (or inversly proportional to the square of) the image size. The larger the format, the less depth of field you get. That is another reason why these rise, tilt, swing, etc. tricks are useful to the large format photographer.


Steve.

Great post. Very well explained.


I found this an interesting article, although I do still have a limited understanding of all the factors:

http://luminous-landscape.com/tutorials/resolution.shtml

The resolution falloff due to diffraction was a very interesting part.
Not sure how accurate the figures are (haven't seen any counter arguments to the article so far), but if you look at Table 3 (about ¾ of the way through the article) the table gives an estimated maximum resolution for each of the formats at various apertures.

The article goes into a lot more and the fact that this is not the whole picture. E.g. 'mostly' in digital, each pixel is only a single colour – meaning 'image processing' needs to occur across the ‘bayer mosaic pattern’ to achieve full colour per pixel (hence these are 'simplified calculations')

But going back to the article and the table in question(Table 3), you can see (using the centre column), the resolution falloff appears to be huge at small apertures.

E.g.
On FF, max resolution at: f11 = 16MP, f16 = 7MP, f22 = 4MP
On APS-C, max resolution at: f11 = 7MP, f16 = 3MP, f22 = 2MP

Looking at table 3 again (if I'm reading the data correctly), even increasing the format size doesn't seem to help with resolution when trying to achieve large DOF, as it appears the gains made with resolution increases from larger format sizes, are lost as you'll need a smaller aperture to get the same DOF.

I must admit, I don’t fully understand how this relates to ‘real life’ and at the point where diffraction is limiting resolution, how the quality of the lens comes into play. I did post this question on FF – but have got no replies so far.

The article implies that diffraction will be limiting resolution to some degree from f11 on 20+ MP high resolution DSLR's. And looking at the f22 figures, it really does put you off using such small apertures.

BTW - I'm open to guidance here as I'm far from being an expert - but have been trying to understand.