Tilt-shift. Why not?

[amolitor]

Are you familiar with the definition of "passive aggressive" as well? Perhaps you could tell us a little about that, Braineack.

 

[Braineack]

see: braineack's posts.

 

[veriwide]

No clue what you guys are talking about.... but I do know something about TS lenses as I own and use a Canon 35mm TS on my F1 - film camera. I've been photographing for well over 45 yrs and for the longest time I kept going to wider and wider lenses to try to get the tops of buildings and the like in my photos. Years ago I was all set to buy a 24mm Olympus TS lens but the salesman talked me out of it.. said to just use a high quality 24 or 21mm lens, shoot fine grain film and crop out the extra. Well, that kind of works but you end up with wasted film space and in any case the visual perspective is off compared to a normal view. Same thing happens if you shoot digital and 'fix' the tilt of the scene.. you end up cropping away a lot of pixels. Now with the TS all is fixed.. you shoot 'full frame' and you eliminate the 'extra' foreground if you wish. I use the TS as my standard lens all of the time. One more thing to consider. Lets say you are on a bridge and you want a lower angle view in your shot.. well, saved again, as you can also 'drop' the lens to get the view you want. In fact you can fine tune all of your photos to represent the actual angle of view that you want without moving.

The only fly in the ointment for me is that being a FD lens I have to use stop down metering. Worth it though.

Get one if you can. You won't be sorry.

 

[Rick58]

I have the Nikkor 35mm PC. I've never used it. One day maybe

 

[TCampbell]

I have a Canon TS-E 24mm f/3.5L II and I do use it.

I should mention a few comments that have not been made.

1) Every tilt-shift lens I have ever seen is manual focus only. There are probably two good reasons for this... #1 is that you aren't trying to focus a "point" like you are with a traditional lens.. you're trying to focus a "plane" -- so point focus would probably not be very useful (I was going to say "pointless" but I didn't want to hear the groaning.)

2) The "shift" part (which is what handles perspective correction) is fairly straight-forward ... except to say that the human brain seems to expect a tiny amount of narrowing when photographing tall things... so if you "perfectly" correct for perspective distortion on *some* subjects, it creates the optical illusion that the building is getting wider (even though if you were take a ruler to the image, you'd see that it's not). I read a long time ago that they recommend you correct the perspective distortion... and then just *barely* back off a little for a more natural look.

3) The "tilt" part is actually a little hard to master. There are a few ways to do it -- but do expect a learning curve.

I have read many different techniques. The one technique that I thought was closest to the best for manually working in the focus is to leave the tilt normal (0º tilt) and focus to something in the background (on the plane you want in focus) first. THEN... slowly adjust the tilt while keeping an eye on a foreground object that you *also* want focused. Once the foreground is focused, re-check the background... some adjustment may be necessary.

The tilt axis is actually marked out with little index lines to indicate the angle of tilt. Most lenses handle about 8º (or 8-1/2º) of tilt in either direction. Incidentally they also usually have a rotation axis which allows you to choose the axis of tilt (e.g. if you want to tilt up/down vs. left/right and yes you could tilt along a diagonal axis if you want and the same all applies to the shift.) I figured if they take the time to add index marks to the tilt axis to indicate the precise angle of tilt, there must be a reason you'd want to know that... there must be a way to mathematically find the tilt angle for any given situation. I did a lot of web searching... mostly came up empty, but EVENTUALLY I find a site that provided enough information that I was able to work out the math.

Here's the formula:

tilt angle = arcsine (lens focal length / distance from lens axis to focal plane)

In case it isn't obvious... since lens focal lengths are measured in millimeters, the distance from the lens axis to the focal plane must also be measured in millimeters to keep our units the same).

also I should point out that the distance from lens axis to focal plane is measured perpendicular to the lens axis and NOT parallel to the sensor plane. In other words as the lens tilts, that line also tilts.

Here's an example:

If I am using a 24mm tilt-shift lens and I am 12" above a table surface and I want the table surface to be perfectly in focus, then the numbers and math are:

lens focal length = 24mm
distance from lens axis to focal plane = 12" which is really about 305mm

So:

tilt angle = arcsine (24 / 305)

That works out to 4.5º.

But we did mention that if the lens is tilted 4.5º then the line from the lens to the table surface is ALSO angled at 4.5º... so we're not really measuring straight down. However... the difference is inconsequential. While we could try to very fractionally adjust the camera height, the math here is: cos = adjacent / hypotenuse. We know the hypotenuse is 305mm and the angle is 4.5. So cosine(4.5) = a / 305. Another way to write that is cosine(4.5) X 305 = a. When plugging in the numbers, distance 'a' is 304mm -- not really enough to worry about.

I have put this to the test... set up the camera over, say, a Persian rug with an intricate pattern... measured the distance from the camera to the rug, plugged in the numbers, got the tilt angle, dialed the tilt into the lens, focused and... the entire "plane" comes into focus with no fuss.

However... I think most photographers probably don't carry a calculator with their tilt-shift lens... they just focus the background and adjust tilt until they get the foreground in as well.