Tilt / Shift movements for macro photography: a question

<p>Hello!<br>

My experience with tilt shift is modest. I believe I understand, more or less, the basics: shift when you want to correct the perspective, tilt when you want to change the plane of focus in order to increase or decrease the depth of focus.<br>

With a small Sony NEX-5 camera and a Lensbaby Tilt Composer adapter, I can use Nikkor lenses, tilted. I know how to use this in order to photograph flat subjects from an angle and obtain full depth of focus. For example, the photo of the cans of sardines that I am showing here.<br>

With my simple equipment I cannot do SHIFT movements.<br>

I would like your help to understand this issue: <strong>are shift movements useful for macro/close-up photography?</strong> If I understand things correctly, they are not. However, I am not certain at all and in fact I am asking.<br>

My curiosity arises from seeing the famous Contax bellows that allow complex movements. I cannot find these movements in other bellows. I would be curious to know (I have an adapter to use Contax lenses onto my Sony) if such a bellows would allow me photography that I cannot now afford with simple extension tubes and Lensbaby Tilt composer.<br>

My first guess with my present understanding is that such fantastic bellows, aside allowing me probably higher magnification and better lenses, would not substantially add to what I can do now - however I am not sure of this.<br>

Thanks!<br>

Marco</p><div></div>

<p>Not only Contax made elaborate bellows.</p>

<p>If you can find one (and I have been looking for years now) the originally inexpensive Spiratone Bellowsmat is another option. Not only are they hard to find, but they go for pretty fair prices these days unless you get awfully lucky. There are a few other manufacturers who offered something similar.</p>

<p>Here's the 1964 Spiratone Catalog description. If you read closely, it also explains what you can do with shift and tilt with it.</p><div></div>

<p>Hello JDM,<br>

I did not know this interesting model! Thanks.<br>

However, let me say that my ignorance persists even after reading this description. In fact, they mention the removal of the distortion (shift movement) in architectural work, however I am interested in close-up applications. I believe that in architecture you cannot easily move your camera up 10 floors (for example) but in macro you can shift your camera 10cm to remove distortion. Once done that, you use tilt to change the focus.<br>

Am I right? If so, I still do not know whether such bellows could offer me any advantage over extension tubes and Lensbaby Tilt.</p>

<p>It's not simple, no matter how you explain it.</p>

<p>What is involved here is something called the Scheimpflug principle ( http://en.wikipedia.org/wiki/Scheimpflug_principle ). What makes it work for product photography, for example, will also work for macro, or perhaps more properly close focus, work in general. Architecture needs shift more than tilt, just as a general rule, while product photography may need tilt more than shift, but both adjustments are needed to achieve the best results.</p>

<p>The simplest, but not necessarily completely accurate, way to put it is that shift is to correct a <em>tilted</em> camera (really film plane) effect on perspective, while tilt is to bring the more-or-less plane of focus into alignment with the objects being photographed.</p>

<p>Find a good large-format textbook (Stroebel's <em>View Camera Technique</em> is sort of the classic one) and it will explain in more detail what is for doing what.</p>

<p>I was lucky enough to pick up a used Nikon PB-4 Macro Bellows attachement last year, but unfortunately have not had a chance to use it much.<br />My understanding of this bellows attachement is that it allow you to do <strong>Tilts,</strong> <strong>Swings</strong> and <strong>Shifts</strong> which are the up and down movements used in perspective control.<br /><br />I'm not an expert but usually Shifts are used to keep tall objects parallel to the plane of focus such as when you are photographing tall buidings and you don't want to tilt the camera up which would cause perspective problems. I'm not sure how the up and down shifts would be helpful in Macro work, but it might have it's use in close-up photography.<br /><br />With the PB-4 you can only perform Front vertical(up and down) Tilts and Shifts when the Bellows attachement is in the vertical position on the tripod, but not Rear Tilt/Shifts such as on a View camera. In the horizontal position, you can only perform Front left-to-right Swings and Shifts.<br>

<br />These operations can become time consuming and very annoying if you don't have a good ball head and tripod since the bellows weighs a ton by itself.<br /><br />It's not as flexible as a real view camera, but for Macro and Close-up the movements may just be enough.</p><div></div>

<p>Thanks for all the answers but...<br>

...I still cannot imagine an example of product photography where shift could improve the result. If you can figure one out, please describe it, or show it. I am really at loss at imagining one.<br>

The only case I can maybe think of is if you want to reproduce a sheet of paper but cannot place the camera orthogonal to it because of an improper support / tripod, and you then solve the thing with shift. However this would be obviously solved by a different tripod rather than by optical means.</p>

<p>Scheimpflug has got nothing to do with shifting.</p>

<p>Shifting is most commonly used for keeping true perspective. An example of small product photography could be a matchbox, say you want to show the top of the box and the front with the label, but with no distortion? Set the camera slightly above and level to the box, then shift the lens down, this achieves your objectives of keeping the box square, the label face on and also being able to see the top of the box.</p>

<p>Shift is very commonly used in product photography, things like watches, cameras, anything with a box etc etc.</p>

<p><a href="http://www.bhphotovideo.com/indepth/photography/hands-reviews/canon-series-l-tilt-shift-lenses">This video</a>, with the images of the G11, might help illustrate the point.</p>

<p> </p>

<p>Thanks Scott,<br>

I begin to see it!</p>

<p><em>"...I still cannot imagine an example of product photography where shift could improve the result"</em></p>

<p>In most cases you probably would not use it, but if you were trying to create special effects such as exagerating the "keystone effect"(vertical parallel lines) in a certain subject, it might come in handy.</p>

<p>Unfortunately in the linked video at 1:53 he incorrectly states "very easily corrected by simply tilting the front element of the lens" the lens was shifted to achieve that result!</p>

<p>Shift can be very useful in product photography to remove the reflection of the camera in a flat reflecting surface when you can't tolerate any keystone distortion.</p>

<p>Specifically, you move the camera parallel to the plane of the object, but keep the plane of the film/sensor parallel to the plane of the object. You then shift the lens to bring the center of the image of the object back to the center of the sensor. Since your camera is now looking at the object at an angle, not normal to the surface, you don't have to worry about your camera appearing as a reflection in your image.</p>

<p>BTW, moderate amounts of "shift" is one of the easiest adjustments to make in post processing: Compose the shot looser, and with a considerably wider field of view than you actually need. Then, put the image of the center of the object off-center. Crop around the off-center object after the fact. If you have enough pixels / lens resolution, this works beautifully.</p>

<p>Tom M</p>

Front tilt and swing will move the plane of sharpness (as per Scheimpflug) while keeping 'perspective'/image geometry the same.<br>But it will shift (!) the entire image, and you thus need shift to center it in the frame again.<br><br>Rear tilt and swing will also move the plane of sharpness (as per Scheimpflug), but will additionally change 'perspective'/image geometry.<br>Rear tilts and swings keep the frame inside the image circle, so no additional shift is needed to put it back in again.<br><br>So if you use front movements to keep image geometry unchanged while playing with the plane of focus, you will also need shifts, no matter whether you are taking pictures in the macro range or not.<br><br>Shifts alone could be considered as (because they are) a more complicated thing: a combination of both front and rear tilt plus rear shift.<br>If you want to photograph something not straight in front of you, you need to turn the camera towards it. That is: tilt both front and rear, which moves the plane of focus and (rear tilt) changes the 'perspective'/image geometry, and shift the image across the frame. You correct both (plane of sharpness and image geometry) by tilting both front and rear back towards the position from where you started. If you do, you end up with a shift of the front relative to the rear.<br><br>This could also be usefull in photomacrography, yes. For instance (as mentioned by Tom above) when you do not want to see a reflection of your camera in your subject (text book example, that).<br><br>So if you tilt or swing the front, or need to avoid shooting straight on, you will need shift, also in macro.

<p>Shift is a convenience, not a necessity. All shift does is create the illusion that the camera is in a different location. In macro work, having the ability to shift would simplify the process of framing the subject the way you want it framed. Most people get by without shift - they simply move the camera or subject as necessary. But because the amount of movement may be very small, its a trial and error process and is tedious. Ok for the studio, less fun in the field (eg, nature macr0 - insects, flowers, etc) where there can be natural movement.</p>

<p>Someone mentioned the Spiratone bellows. Another product that I recall from the Spiratone catalog was a micro-positioning stage. This was basically an adapter that fit between the camera and the tripod that allowed the camera position to be adjusted on two axes using knob-style verniers. I never had one of those, but I do have a focusing stage that provides vernier adjustment along the optical axis by shifting the camera in and out - I can use it to emulate shift by simply turning it 90 degrees so that the adjustment is side to side. I found mine at a photo flea market and paid something like $5 for it..</p>

As mentioned, shift is indeed a necessity if you swing or tilt the front.<br>And that - tilting or swinging the front - is a necessity if you want to move the plane of focus and not distort the shape of the subject.<br><br>Shift does not create the illusion that the camera is in a different location either. Perspective (in the true sense of the word) depends solely on the position of the lens. If you do not change that, you do not change the sense of where the camera is. And if you do by shifting the lens, you do indeed change the sense of where the camera is, but then it's not an illusion.<br><br>Shift by itself (i.e. without swings/tilts) can be quite usefull to change framing, yes. And it can be a necessity to avoid reflections.<br><br>You can't emulate rear shift by moving the entire camera, because when you do, perrspective will change (which it will not if you only shift the rear, not the lens).

<p>Great answers: Louie , QG, and Tom ! </p>

<p>Thanks Harry...........</p>

<p>And Thanks Scott sorry you were on the previous page...</p>

<p>So it seems that after two pages of useful comments, my initial doubts have disappeared and I know now something more. Thanks.<br>

Eventually I understand that for the type of things I like to photograph I don't need a bellows nor shift movements (I like tilt and moderate close-ups); at least now I am sure of what I am saying.</p>

<p>Shift becomes tilt and tilt becomes shift depending on which way the camera is mounted on the tripod ? At least that is how I worked with basic bellows cameras with limited movements.</p>

<p>JC,</p>

<p>That is so wrong I don't know where to start!<br>

Shift and tilt are entirely unconnected and completely different movements. Shift has become a ubiquitous term meaning any parallel movement, including shifting, rising and falling. Tilt has come to describe the non parallel movement swing as well as tilt.</p>

<p>Tilt is a change in vertical angle and becomes swing when it is a horizontal movement. It never becomes shift.</p>

<p>Shift is a horizontal movement that becomes rise and fall when it is vertical.</p>

<p>JC - Read this, or at least look at the drawings that illustrate the movements: http://en.wikipedia.org/wiki/View_camera#Shift</p>

<p>Tom M</p>

<p><em>JC,</em><br /> <em>That is so wrong I don't know where to start!</em><br /> Perhaps not as wrong as you think Scott :-)<br /> It does depend on what sort of camera you are using and my experience is limited to Thornton Pickards and Speed Graphic Monorail. With the later obviously I had the movements you are used to but with the TP the setting up from folded camera gave me a limited degree of swing effect which could be used with the camera in the 'normal' mode or if the tripod was suitable tilting the camera 90 degrees sideways. The adjustment for the back in conjunction with limited movement of the front gave me upwards or downwards shift movement.<br /> So it does depend on if you know the potential of your gear.<br /> So don't jump in so quickly please Scott ... this is not the first time.</p>

<p>Thanks Tom but I am aware of the potential movements.</p>

<p>No JC, as my answer explained, and Tom's link adequately visually demonstrates,Tilt and Shift are not the same thing and one never turns into the other.</p>

<p>As I explained, and you are confirming, tilt becomes SWING, but <strong>not</strong> SHIFT.</p>

<p>Scott: "<em><strong>Tilt</strong> is a change in vertical angle and becomes <strong>swing</strong> when it is a horizontal movement</em>"</p>

<p>JC: <em>"a limited degree of <strong>swing</strong> effect which could be used with the camera in the 'normal' mode or if the tripod was suitable <strong>tilt</strong>ing the camera 90 degrees sideways</em>"</p>

<p>It doesn't matter what gear you are using the naming convention stays the same, even for you. Your Thornton Pickards were not special, when you set them up if you didn't fully extent the front standard you had a degree of upward tilt, if you put the camera sideways that turned into swing, if you mounted it to the left you had left swing, if you mounted it to the right you had right swing, if you mounted the camera upside down you could have got the most useful for a plate camera, downward tilt. But that lack of fully extending the front standard never gave you shift, or rise and fall.</p>

<p>"<em>Thanks Tom but I am aware of the potential movements</em>."</p>

<p>Clearly not, that is why I stepped in.</p>

<p>If you were aware of how older cameras worked you would appreciate what I am talking about instead of nit picking. Your explanation of how the TP works brings a smile to my face. I guess I will have to tolerate your foolish snipeing at me.</p>

<p>JC,</p>

<p>Why not take a picture of your camera, or link to one off the internet if you no longer have it, that illustrates how your camera is different, how the rules of terminology are altered for the use of a Thornton Pickard. I can assure you I am very familiar with how many older cameras work, I do fully appreciate what you are saying and that is why I am saying you are wrong.</p>

<p>Either the lens board is parallel to the film plane, in which case any movement is a shift, rise or fall (or a combination), or the lens is not parallel with the film plane, in which case the movement is tilt or swing (or a combination). Rotating the camera does not change the angle of the lens board to the film/sensor plane, so rotating a camera does not mean <em>"Shift becomes tilt and tilt becomes shift".</em></p>

<p>Please understand, I am not following this up to be pedantic, nit pick, nor indulge in foolish sniping, I am doing this in the public forum because it is searchable and I would hate future readers to be confused or misguided by your comments. You know full well I am happy to point out the factual errors in your statements in private.</p>