Using lens tilt on your digital SLR
Using lens tilt on your digital SLR
How to focus when you tilt a lens
Keith has used the Canon 90mm and 24 mm tilt/shift lenses for some time and we’ve just got a new lens for the collection, a 17mm tilt/shift lens.
How do you know what tilt angle to set?
If you are completely new to tilt/shift lenses you might first want to read the article about using tilt/shift lenses that Keith wrote after getting the 24mm lens.
This current article assumes a passing knowledge of such lenses. You may see such lenses described in a number of ways, such as a ‘variable plane lens’ or the ‘miniature world effect’ – basically lenses where the front of the lens ‘bends’ in some way.
I’ve also written another article showing an iterative focus technique for tilted lenses, which may be of more use for close-up use of such lenses.
I’ve noticed that whilst people using shifted lenses seem quite comfortable with how they work, there is considerable uncertainty when it comes to using tilt.
I’ve seen many examples of people just fiddling with adjustments in the hope of getting the image to look right.
I’m going to outline some of the principles here, and include some references to more detailed articles that I’ve found particularly illuminating.
Tilt or swing?
If you’ve experience of using large format cameras then you may well differentiate between rotating the lens about a vertical axis (swing – or side to side) or a horizontal axis (tilt – or up and down)
I’m aiming this article more at readers from a 35mm or digital SLR background, so I’ll just refer to any movement of this sort as tilt.
The key concept to understand is that of the ‘plane of focus’.
If I am square on to a wall and focus on the point directly in front of me, then I expect the whole wall to be in focus.
Note – for most of these examples I’m going to ignore the effects of depth of field.
This is where the zone of sharpness is not a thin plane, but extends a distance in front and behind it. I’ll come back to it later, but for the moment forget about it.
If I’m outside and set my lens to focus at 3 metres away, then any object that is directly 3 metres from the front of the lens (i.e. in the middle of the frame) should be sharp.
Think of that object on an imaginary wall or sheet of glass.
Objects on that sheet of glass that are not in the middle of my frame are also sharply focused – even though they are more than 3 metres from my camera.
Let’s say I’m taking a photo of two butterflies. One is directly in front of my camera (perhaps on a branch) whilst the other is on the ground.
I focus my camera on the one in front of me (the yellow one) and note that it is 3 metres away. The one on the ground (the red one) is also in focus, although, as you can see, it’s slightly further away from the camera than the yellow one.
When you have a lens that can be tilted relative to the camera body, the plane of focus shifts when you move the lens away from the normal orientation.
The photo below shows the tilting of a Canon 24mm lens on my old Canon 1D.
Move your mouse pointer over the lens to see the lens tilted upwards.
As you tilt the lens, the focal plane tilts over.
The diagram below shows the effect of tilting the lens downwards. This is a side view – so think of that dotted line as a plane that slices through the scene in front of you. Myself, I find the ‘sheet of glass’ analogy quite helpful.
As you can see, the plane of focus has shifted, so now both butterflies are not in focus.
Many people who’ve tried using lenses with tilt, find this effect quite unpredictable and just fiddle around with focus and tilt settings until they find that the plane of focus matches up with what they want.
There is however another way of looking at the effect of tilting a lens, and for this I’m eternally grateful to Harold Merklinger and his book about ‘focusing the view camera’.
I’ll include links to his on-line work at the end of the article. If you want to take your study of camera movements (as tilt and shift are sometimes known) further, then you don’t need to understand the maths completely, if you can visualise the concept of planes of focus.
It turns out that there are some interesting connections between the geometry of the plane of focus, the amount of tilt and the focus setting of the lens.
I’m going to include some short animations, courtesy of Harold Merklinger’s work (items are © Harold Merklinger).
In these, the camera is shown as a ‘view camera’ with bellows providing the focus adjustment and the whole front lens tilting.
First of all, the effect of tilting the lens.
The important thing to notice is the red line, which is the plane of focus.
For a detailed discussion of the other lines, check out the links to ‘Focusing the View camera’ at the end of this article.
Notice how, as the lens tilts, the position of the plane of focus is given by the points where the film (sensor) plane, and two other planes meet.
I’m not going to go into details here as to why this happens but it’s an important aspect of making the position of the plane of focus go where we want it to rather than trusting to luck and fiddling with the tilt and focus.
The next important aspect is the effect of changing the focus of the lens.
In normal use this just moves the plane of focus towards or away from us.
When the lens is tilted, the effect is quite different.
Look at the one point that remains stationary – the ‘Hinge line’.
It’s called this since you can think of our imaginary sheet of glass being hinged along a particular line.
You may note mention of the Scheimpflug line – I’ll leave discussion of this to the more detailed articles in the links. It’s not important in the current discussion.
So, we have two effects shown in the two animations.
- Tilting the lens tilts the focal plane about the Hinge line.
- Altering the focus setting of the lens also tilts the focal plane, but the hinge line does not move.
These two effects allow us to place the plane just where we want it.
Watch the focus and the tilt move back and forth a few times to get a feel for what is happening – it’s that red plane of focus and the point it shifts around are the things to note. Come back to these animations once you’ve seen some real examples of how tilting and focusing the lens moves the plane of sharp focus.
In reality the camera setup will still need some tweaking, but at least we can predict whether a particular tilt lens can produce the effect we are looking for.
How do I transfer this knowledge to a form I can use to help me take photos?
Without going into the maths, I can use the distance between the centre of the lens and the hinge line, to specify where the plane of focus is.
This distance is often known as ‘J’ and I’ll include some handy tables and a spreadsheet later, to make it easy to use.
I have a printed ’tilt table’ in my camera bag that I use as a guide for when I need to use tilt in my architectural and interior photography.
Look at this scene in my kitchen.
Initial test setup.
The camera (Canon 1Ds Mk3) has a TS-E 90mm lens on it.
I’ve chosen this lens, since at f/2.8, it has a narrow depth of field that will show up the effects of tilt in images at the size I’m using here on the web.
Notice how the top surfaces of all the bars, all intersect along a line on the top of the work surface, under the camera.
As you might have guessed, this is our hinge line.
The vertical distance between the lens and the work surface is what I called ‘J’
Using the lens normally (i.e. without tilt) , I’ve focused it on one of the squares on the rightmost piece of wood.
As you’d expect, parts of the scene closer or further away are out of focus.
Our plane of focus is vertical here – look for other areas of the scene that are in focus, they are all on the vertical plane of focus, just like the butterflies right at the start of the article.
Let’s see what happens when the lens is tilted downwards.
This lens has an adjustment knob that allows you to tilt it by an amount up to about 8 degrees.
The adjustment scale is quite fine, but even with care I’ve never been able to set it to an accuracy much better than about a third of a degree.
The lens centre is set some 65cm above the work surface, thus our distance ‘J’ is 65cm for this example
Note that the charts later work just as well in feet and inches. I’ve set the lens tilt in this instance to 8 degrees, which is about right for a ‘J’ distance of 65cm.
The camera is also square on to the wall at the end of the room.
The centre of the tilting part of the lens is directly above our ‘hinge line’.
Note – there are some practical difficulties with working out the exact centre of the lens for this purpose, but I’ll leave discussion of this until later.
Look what happens to the image as I subsequently adjust just the focus of the lens.
I’ve taken two shots at different focus settings (and not touched the 8 degree tilt setting).
The 8 degrees of tilt initially sets the focal plane to run through the hinge line, the line where all three bars point to below the camera.
Move your mouse over the image to see the effect of altering just the focus setting alone – if need be, go back to the 2nd animation above to see how our plane of focus moves when you adjust focus and leave the tilt unchanged.
By setting the correct tilt, I’ve managed to place the plane of focus through either the metal shelf bracket, or the far piece of wood.
In this next example, I’ve swapped the lens for the TS-E 24mm f/3.5
For the same height (‘J’) above the work top (65cm) the appropriate tilt with this longer focal length lens is lower. My tables give a value of just over 2 degrees.
The camera is in the same position, but will obviously capture a much wider view.
I’ve put a strip of masking tape on the worktop to show up a bit more clearly than the black stone.
As before, I’ve taken two photos, both with the lens set at slightly over 2 degrees of tilt.
Remember that it’s difficult to precisely set tilt values on the lens so for this test ‘slightly over 2’ will suffice.
The first is with the camera set to a fairly close focal distance (just under a metre).
The second (move your mouse over the image to see it) shows the effect with the lens set to infinity.
Remember that the tilt has not been altered between the two photos.
This is an important result, since it shows that with the lens set to infinity, the plane of focus runs parallel to the ground (assuming the camera is level), but at a distance below the camera equal to the distance (J) to the hinge line.
This example also shows how the field of view of a lens can change with a combination of focus adjustment and tilt.
A practical use for this.
Say I’m taking a photograph of a landscape and my camera is on a tripod pointing at the horizon.
By setting an amount of tilt appropriate for the lens I’m using and the height above the ground it’s at, I know that with the lens focused at infinity, the plane of focus runs along the ground.
If the ground slopes away or upwards, I just alter the focus to line the plane up with the ground (look at the second animation again if you are not sure about this).
If I want to alter the framing of the image and my lens can shift up or down, I can move the horizon (level ground) up or down in my frame by shifting the lens. This is independent of the tilt.
If I tilt the whole camera forwards or backwards then the distance to the hinge line increases and its position moves backwards or forwards of the camera.
The diagram below shows that when I tilt the camera upwards. The new distance ‘J2’ is larger than just the height above the ground we used before for our ‘J’ value.
In this case, to make the plane of focus follow the ground, I need to have the lens focused beyond the infinity mark (the focus ring on my new 17mm lens goes past the infinity mark).
Another use for this example
Now think of the diagram above as a view from above.
In this case the line represented by the ‘Ground’ could be a wall I want to photograph.
As before, our new value ‘J3’ is a bit bigger than the simple distance of the camera from the wall. I keep a roll up steel tape in my camera bag that I can use to measure this distance.
When you look at the tables below you’ll get a feeling for the accuracy required for all this. When first trying this stuff out, don’t get too hung up on precise measurements. A few experiments should give you a feel for what the numbers mean.
In the diagram above, I’m tilting the lens sideways (swing) to make the plane of focus run along a wall, whilst aiming the camera to get more of the wall in shot.
Here’s an example from a real job where the partition installers wanted a picture showing the glass doors.
The photo was taken with a 1Ds and the 24mm tilt/shift lens.
I’ve used an aperture of f/5.6 and a degree of vertical shift (downwards) to get the composition I wanted.
The tilt function is not something I use very often in my work with the 24mm lens, but when it does work, it works very well.
Hopefully you’ll now appreciate that with a knowledge of the distance ‘J’ you can place the plane of focus fairly close to what you want for an image.
Of course, it helps if you can visualise the plane of focus you want.
However, I know that some people find this less intuitive, so I’d suggest some experiments.
If you have the chance, try and duplicate the kitchen example from above, or try taking your own wall photo such as that of the glass partition doors.
The picture below, shows a shifted and tilted (swing) view of the kitchen, taken with the new TS-E 17mm.
I’ve set the plane of focus along the kitchen units (the ruler is on this plane of focus, and is sharp.
A detail of the image (taken at f/4).
Here is the table of distance ‘J’ against tilt angle and lens focal length that I use.
It was produced from a spreadsheet I put together, that was inspired by one created by Harold Merklinger.
The small one is printed and laminated, for keeping in the camera bag
I also have this larger version, which includes imperial units and other focal lengths
The mix of metric and imperial heights are because I’m old enough to use both interchangeably when estimating distances.
You can change the focal lengths as needed, and by judicious adjustment of the metric values (only 2 decimal places shown) you could produce a table in whole feet and inches, or whatever you wanted.
Note how small the tilt values are for J distances of over a metre for the shortest lens focal lengths.
Setting a tilt of 0.6 degree accurately is not easy, and I’d suggest that 0.2 degrees would be almost impossible to set on any tilt/shift lens I’ve looked at.
If you want to produce your own version for your own personal use, then the spreadsheet is available for download below.
Please note that this is provided freely for non commercial use and cannot be included in any product that is sold.
In case you are wondering why the tables include 150mm and 180mm, they are for when I’m using my view camera adapter. I’ve written an article describing the construction and use of a DSLR view camera adapter camera.
If you’ve any comments or suggestions then please do let me know. I’m hoping that this article helps people use tilt/shift lenses more easily.
Those of you who know camera movements a bit better, may notice that I’ve simplified some details of using such lenses.
Buying a TS-E lens – We make a specific point of not selling hardware, but if you found the articles and information here of help, please consider buying a lens, or any other items at all, via our link with Amazon.
Amazon UK link / Amazon Fr / Amazon De
Amazon USA link / Amazon Canada link
It won’t cost any more (nor less we’re afraid) but will contribute towards the running costs of our site
or, if you’d prefer, you can look at
or B&H Photo
This was entirely to try and make the key principles more ‘digestable’, and I’d suggest that anyone’s next port of call would be the site with Harold Merklinger’s work on it.
I’ve not covered the effects of depth of field in the examples above. With an untilted lens, image sharpness drops off on either side of the plane of focus. This effectively gives a thickness to the focal plane, which is even from side to side and up and down.
With a tilted plane, this thickness varies with distance from the camera. In other words, it’s very thin near to the camera and gets thicker further away. The upshot of this is that if you have the plane of focus running near to the camera, the depth of field can be very small, even at smaller apertures.
At small values of ‘J’ it becomes more important to decide just where to measure it from.
I pick the rotational axis of the lens, but this depends on lens design. It’s similar to the problem of deciding where to rotate a lens for eliminating parallax effects when taking photographs for stitching. Once again, a few experiments should help.
As you change the focus setting of the lens, you are effectively moving it back and forth – with the camera fixed on a tripod, this moves the hinge line back and forth slightly.
Liveview is very useful for ensuring correct focus. If I’m trying to get a plane of focus spot on, I’ll set an approximate value of tilt and then rack the focus back and forth to see if the plane matches what I want to photograph.
If you don’t get a line up, then slightly increase/decrease the tilt and repeat the procedure. With a bit of practice it’s quite a quick process.
I’ve written up a second article describing this iterative focusing process for tilted lenses in more detail. I find that it works particularly well in close-up situations, particularly with extension tubes, where measuring the distance ‘J’ is not that easy.
I find that using the tilt tables to get an initial value for the tilt angle, gives much more repeatable and predictable results.
All our UK used photographic items come from MPB – check out their stock – it’s where I got our original TS-E24 and TS-E90 lenses from
Some Info on other sites
- Focusing the view camera – where to go next
- Keith’s tilt table spreadsheet (zipped file)
- Using tilt/shift – Canon article
- Perspective Control – clear overview
- Shift lenses – Very good intro with clear diagrams
- The View camera – Wikipedia article
- Using 35mm T/S lenses – includes modification procedures for Canon lenses
- Using MF lenses with a shift adapter
- Comprehensive collection of T/S info
In August 2017 Canon announced three new TS-E lenses
- The 50mm replaces the 45mm and offers ±8.5º of tilt
- The 90mm replaces the 90mm and offers ±10º of tilt
- The 135mm is a new lens and offers ±10º of tilt
As a result of this I’ve created two new versions of the tables – you can still download the spreadsheet and customise as you wish [click to enlarge]
All the latest articles and photo news items appear on Keith's Photo blog .
We've a whole section of the site devoted to Digital Black and White photography and printing. It covers all of Keith's specialist articles and reviews.
If you start your buying of anything whatsoever from Amazon (not just what's listed) via one of our links below, it helps myself and Karen to keep the site going.
We really do appreciate this - thanks [link for Amazon UK | Amazon US]