The Scheimpflug Way

It's easy to understand how allure of working with big, beautiful 4"x5" negatives and the motivation they provide photographers to step up to large format. Since returning to 4x5 photography for a significant portion of my own work the difference this makes has not gone unnoticed. Leaving aside the oft touted benefits in terms of grain and sharpness, images from these negatives seem to exhibit a crispness and presence that seems to come through even when viewing scanned image on a not particularly high resolution screen where you wouldn't expect to see any significant differences when compared to similar images from smaller negatives. 

To be frank however, if this were the only advantage I'm not sure I'd bother. I'm still a believer in the balance of quality, compactness and versatility of medium format for a lot of work. While I enjoy the sort of slow photography approach that large format demands there are situations that simply don't allow for that and I'm sure a good number of my more treasured images wouldn't have come to be if I had to capture them with a large format camera. Oh, and let's not forget every time the shutter opens to expose another sheet of film it costs me $2 before any sort of print is made, and it's only that cheap because I'm using black and white film I process myself. If I shot colour transparency film and had it processed by a lab that figure quickly quadruples. 

As I mentioned in a previous episode what really cinches the case for large format, at least for much of my photography, is the readily at hand ability to use camera movements. When those with even a passing familiarity with the idea think of camera movements the phrase that often comes to mind is "perspective control". Perspective control is nice and now and then I may take advantage of that capacity since it's there anyway, but for the kind of work I like to do a phrase that excites me more is "plane of focus control". While I mentioned this in the aforementioned article, I wasn't able at that time to offer much by way of example other than to present an example of a medium format photograph that, though reasonably successful as is, could have been better if I had the additional sort of control large format photographers take for granted. For those not familiar, I think a bit more explanation of where this control comes from may also be in order.

While you may not have heard the term plane of focus control before you may (or may not) have heard of the Scheimpflug principle that describes the idea in more concrete terms. To understand how photographers can benefit from the ability to use consider first how ordinary hard bodied cameras are focused. The lens barrels of most 35mm, DSLR and medium format cameras will have a focus ring marked with a distance scale. Imagine standing square on to a brick wall exactly 3 metres away. If you set the focus distance on the lens to 3 metres and point then take a picture of the wall it will all be in sharp focus. If we ignore subtleties like optical distortions or the fact that the spot on the wall directly in front of you at eye level is a little closer to you than other parts we can say the flat surface of the wall (Did I say the wall was flat? It's flat.) occupies the plane of focus. Replace the wall with an object at the same distance and it will be in sharp focus to. Objects at different distances, either nearer or farther, will appear less well focused in proportion to how much nearer or farther they are. The difference between the sharp focus distance and the actual subject distance that can be tolerated without this lack of sharpness being noticeable is what the term depth of field means, but no matter how far we stop the lens down it doesn't change how we define the plane of sharp focus.

I don't know many photographers, any actually, who have achieved success photographing flat vertical surfaces like this, but a portrait photographer who typically has a single subject to focus on should not have any difficulty working this way. And if you imagine you're a landscape photographer (not too difficult for some of us) shooting a scene with a winding river about 75 metres distant leading off towards mountains a few kilometres away, having the focal plane at a fixed distance similarly presents no challenge because the nearest subject is distant enough to be sharp at infinity focus just like the mountains in the background. But let's say that you move a little farther on and find an interesting detail along the river bank you want to photograph from only a few metres away while still keeping the distant mountains sharp in the background. Now it's not so easy. The keen among you might consider the possibility of stopping down in hopes you can get them both within the hyperfocal distance zone. You might be able to do it, but then again they might not. Even if it's possible it might require a longer shutter speed than you'd prefer to use and introduces compromises such as the fact lenses don't perform as well when stopped down to minimum aperture.

But what if, rather than being fixed parallel to the camera (or perpendicular to the lens's line of site if you prefer), that imaginary flat surface we call the plane of focus could be angled any which way? Well it turns out that if you aren't bound the way most cameras are to keeping the centre of the lens fixed parallel to that other flat plane, the surface of your film or digital sensor, you indeed can be free to angle the plane of focus any which way. Well, within the limits set by your lens' capabilities at least.

Enter the Scheimpflug principle, which I will attempt to explain in as plane a language as I am capable of (possibly not one of my great strengths I'm afraid). To begin with picture the flat surface of the film as just a small part of an imaginary plane, the film plane, that, like our plane of sharp focus, can extend out to any size we need, and that there is a similarly a lens plane roughly defined by the front of the lens. In an ordinary camera these three surfaces are always parallel to each other. Imagine tilting the front lens forward so that its extended plane is no longer parallel to that of the film. They meet at some point (actually some line before any of you mathematicians are paying attention.) The Schempflug principle says that doing so causes the plane of focus to shift also in such a way that its angle causes it to connect with the other two planes at the same place where they meet. Got it? Didn't think so, but at least I avoided terms such as "oblique tangent" like the Wikipedia page on the subject does. Perhaps a diagram would help.

Here we can see that by tilting the lens forward the focal plane has been transformed from the imaginary vertical wall sitting in front of the camera at whatever distance has been focused on to a horizontal surface that runs along the ground. In this situation if an image were taken of the ground going off into the distance every point along the ground would remain tack sharp from foreground to horizon without any reliance on depth of field to get things "sharp enough". I should mention that a similar effect takes place if the lens plane is kept vertical and the rear film plane is tilted back. The same rule applies about the focus plane shifting so that all three planes meet at a common point, but tilting the film back also results in a perspective shift, exaggerating the relative size of nearer objects, which may or may not be desirable in a given situation.

While I've suggested ordinary hard bodied cameras keep the lens fixed parallel to the film you can achieve this kind of focus magic with SLR/DSLR cameras through the use of expensive and not particularly common tilt-shift lens. Lenses for 35mm size (film or full frame sensor) usually go for in the neighbourhood of $2000, and for medium format let's just say much higher and leave it at that. But the capability that smaller cameras can achieve only through the use of highly specialized optics are just par for the course for large format photographers. This is because instead of the rigid plastic and metal boxes that are the essence of the kind of camera bodies most of us are used to the "box" that forms the bodies of the vast majority of large format camera consists of a flexible bellows supported between two rigid plates, one in front that supports the lens, and one in the back where the film goes. With just a little simple engineering it's easy to design such a camera to allow the orientation of the lens and film planes to be adjusted in all sorts of useful ways. This kind of camera can be generally referred to as a view camera. Not all large format cameras are designed this way, the soon to hit the market Travelwide is a good example, and there are smaller view cameras that shoot medium format size negatives. By and large however large format photography is nearly synonymous with the control you get with all the lens and film plane movements view cameras make possible.

Let's stick to the one we've been talking about however, tilting the lens. That is because of all types front and rear movements possible, tilt (front and rear) is what excites me most. If I could just have that one movement on my medium format cameras I may not have ever gone back to 4x5. There is a medium format camera, the Rolleiflex SL66, that is capable of limited lens tilt and for a short while I was giving some thought into whether ditching all my other kit might leave me enough funds to get a basic Rolleiflex system started but abandoned the idea when my research began to indicate the tilt capability had a bit too many limitations for my tastes. Just as well though, the large format world is not a bad place to be.

So why all the fuss about tilt? I've been getting not too shabby results without it for years after all. Well I'm happy to say this time out I come bearing a good and proper example of exactly what this capability can do for an image. The photograph near the top of this weeks post is one of the first I made with the Wista 45SP I spoke about at length last time. Finding myself in a location I wouldn't have a chance to get back to anytime soon and still in the testing phase with the Wista I brought along the old Iskra to let me grab a few insurance shots. So it is I have a very similar shot done on medium format.

Now I didn't make these to shots with any sort of comparison in mind so I took no real measures to keep them as comparatively identical as possible. Both were shot on Ilford HP5+ and processed in PMK Pyro. They are shot from slightly different positions since I recomposed this second shot to give me something I thought was a little more pleasing with the square format. The lenses, a 150mm Symmar-S for the 4x5 and the Iskra's 75mm Industar are to a reasonable degree of approximation equivalent in terms of focal length. An early 60's Soviet Tessar vs. a more modern six element German optic might not seem to make for a fair comparison it's entirely capable of delivering impressive results and I hope to convince you that this and other differences such as the smaller format don't account for what I am about to show you. While I think both shots look fairly reasonable on screen at the resolution I present them at here, let's see what they look like at higher resolution with a pair of unsharpened negative scans from my Epson V500.

Of course the Iskra shoots a smaller format negative, but it's still a respectable medium format negative and looking closer the lack of sharpness is obvious long before grain becomes apparent. Yes the large format shot was taken at f/22 rather than f/11, but with double the focal length it needs to be stopped down further. The scans are a reasonably close match of the same are of foreground rock a few metres from the camera. When I was using the Iskra I focused on the island which was several tens of metres distant, relying on depth of field for whatever sharpness could be maintained in the foreground. The difference with the Wista image is that I didn't have to chose between the island and the foreground rock, I was able to focus on both at the same time, and I mean bang on critical focus, courtesy of the option I had to tilt the lens forward. As clear as the difference is here, if you open the comparison image up a full resolution you'll see the blades of grass keep showing more detail up to 100% magnification, not just the stalks but the tiny bits of detail on them. It just seems more impressive the closer you look while the comparison shot from the Iskra looks worse. But lest you think the image on the left looks bad because of some overall shortcomings in the image notice that beyond the foreground rock itself even the little bits floating in the water that are nearer to the distance at which the camera was focused don't look nearly so bad. But for the ability to get that foreground in focus without losing sharpness elsewhere there might not be so much to chose between the two shots.

There are other camera movements to speak of such as swing, the side to side equivalent of the fore and aft of tilt, and rise and fall movements to combat keystoning and other perspective effects that might be undesirable in many situations. And for every potential problem these movements have the potential to overcome, they also have the potential to creatively enhance. For example tilt and swing movements can be used to artificially induce a shallow depth of field effect that can often create the illusion that a real scene is a photo of a miniature model. 

All of these movements can come in handy. With the 4x5 cameras I've owned I'm sure I've found occasion to use all of them once or twice. Honestly though I could live without most of them though. All except tilt. To me having a large format camera that lacked that one capability would really just feel like a waste.