Anyone familiar with tilt-shift?

oROSSo

oROSSo

Messages
640
Name
Ross
Edit My Images
No
Recently got a tilt shift so I can shift the lens to create pano's.

the setup I have is a collar that grips the lens, which then mounts the tripod so the camera slides left and right and the lens stays still (the avoid parallax issues)

I've been reading about tilting downwards recently for DOF but all of the articles I've read are written in a way that assume the lens will be tilting, where as I can't find an article that discusses what will happen if the camera were to be the component that tilts.

Anyone know of such an article?
 
oROSSo said:
Recently got a tilt shift so I can shift the lens to create pano's.

the setup I have is a collar that grips the lens, which then mounts the tripod so the camera slides left and right and the lens stays still (the avoid parallax issues)

I've been reading about tilting downwards recently for DOF but all of the articles I've read are written in a way that assume the lens will be tilting, where as I can't find an article that discusses what will happen if the camera were to be the component that tilts.

Anyone know of such an article?
Click to expand...

No.
 
oROSSo said:
"UaeExile has quoted your post"

you just know it's gonna be jibberish at best.
Click to expand...

I rarely have anything substantive to say when you're talking about expensive kit or shooting locations in the middle of bloody nowhere.
 
oROSSo said:
Recently got a tilt shift so I can shift the lens to create pano's.

the setup I have is a collar that grips the lens, which then mounts the tripod so the camera slides left and right and the lens stays still (the avoid parallax issues)

I've been reading about tilting downwards recently for DOF but all of the articles I've read are written in a way that assume the lens will be tilting, where as I can't find an article that discusses what will happen if the camera were to be the component that tilts.

Anyone know of such an article?
Click to expand...
No, but is it necessary? What matters in image making terms is the angle of tilt between lens and camera sensor plane. Whether that angle was achieved by tilting the lens or by tilting the camera doesn't matter. The only difference that makes is what the initial pre-tilt image is. The post tilt image is the same. It's all relative. I think :)

(No, I don't have a tilt-shift lens, but I do a of perspective shifting in software, and have read a lot about the optical geometry of tilt-shift lenses.)
 
chris malcolm said:
No, but is it necessary? What matters in image making terms is the angle of tilt between lens and camera sensor plane. Whether that angle was achieved by tilting the lens or by tilting the camera doesn't matter. The only difference that makes is what the initial pre-tilt image is. The post tilt image is the same. It's all relative. I think :)

(No, I don't have a tilt-shift lens, but I do a of perspective shifting in software, and have read a lot about the optical geometry of tilt-shift lenses.)
Click to expand...

Hm, not sure. If you tilt the camera so that the back is no longer parallel to an object (say a church tower), you'll get exaggerated perspective effects, whereas if the back stays parallel but the lens tilts, other effects will come into play. AFAIK.

Maybe look for some info on large format photography (LF), translate "front standard movement" to lens tilt and "rear standard movement" to camera tilt? Those LF guys move everything every which way!
 
Can't understand why you need a tilt/shift for pano's ?

Maybe you can elucidate ?
 
What you want to search for is rear standards tilt. In reality you'll find it easier to experiment and see for yourself, but you can do some forced perspective making the fore or background slightly more pronounced, and of course manipulate your plane of focus as well. The key with TS on small formats is to use very little tilt and see what it does, you'll be focusing by eye as the autofocus goes squiggly with movements, and so trial and error in a controlled environment first is recommended
 
Abbandon, I think you have that the wrong way round. For geometry, ie stopping leaning buildings etc you use the shift of a Tilt-Shift lens. The Tilt is to do with getting everything in focus, from a few inches in front of the camera to horizon (or whatever you want in focus). You focus the far end and then apply tilt until the front comes into focus as well.

At least that is what happened when I had a play with a real TS lens.

Do you have a photo of your set up Ross.
 
Last edited:
ChrisR said:
Hm, not sure. If you tilt the camera so that the back is no longer parallel to an object (say a church tower), you'll get exaggerated perspective effects, whereas if the back stays parallel but the lens tilts, other effects will come into play. AFAIK.
Click to expand...

Not if you end up with the camera pointing at the same thing, i.e., the same image in both viewfinders, whether you got there by tilting camera or tilting lens. When you adjust the tilt screws on the lens, they do the same ting to the sensor plane alignment with the lens, regardless of whether you happen to holding the body or the lens while you adjust the screw.
 
Ploddles said:
Abbandon, I think you have that the wrong way round. For geometry, ie stopping leaning buildings etc you use the shift of a Tilt-Shift lens. The Tilt is to do with getting everything in focus, from a few inches in front of the camera to horizon (or whatever you want in focus). You focus the far end and then apply tilt until the front comes into focus as well.

At least that is what happened when I had a play with a real TS lens.

Do you have a photo of your set up Ross.
Click to expand...


Tilt

If you click the flickr links I posted the first uses front tilt to attempt to get everything in focus from the leaf in the foreground to the tops of the trees with out exaggerating perspective whereas in the rear tilt example the water lilies at the front are some what emphasized.

Actually the example of converging verticals and how not to get them would be an example of front rise,.
 
Ploddles said:
Abbandon, I think you have that the wrong way round. For geometry, ie stopping leaning buildings etc you use the shift of a Tilt-Shift lens. The Tilt is to do with getting everything in focus, from a few inches in front of the camera to horizon (or whatever you want in focus). You focus the far end and then apply tilt until the front comes into focus as well.

At least that is what happened when I had a play with a real TS lens.

Do you have a photo of your set up Ross.
Click to expand...
Here you go:





 
john-oh said:
Can't understand why you need a tilt/shift for pano's ?

Maybe you can elucidate ?
Click to expand...
You don't need one, I've done panos with a pano head but if you slide the camera instead of the lens left and right you have no parallax issues.
 
From your initial post I though it was a "home made" TS lens, clearly not.

Using a TS for a pano won't give you a very wide image though if all you are moving is the camera but leaving the lens static, maybe the equivalent of 2 full frames, if that, stitched together.

I can't see any benefit to this setup as opposed to a pano head whereby you could do anything up to a full 360 degree shot.

I also fail to see how parallax comes into it but that may just be me being thick and not fully understanding your problem.
 
Ploddles said:
From your initial post I though it was a "home made" TS lens, clearly not.

Using a TS for a pano won't give you a very wide image though if all you are moving is the camera but leaving the lens static, maybe the equivalent of 2 full frames, if that, stitched together.

I can't see any benefit to this setup as opposed to a pano head whereby you could do anything up to a full 360 degree shot.

I also fail to see how parallax comes into it but that may just be me being thick and not fully understanding your problem.
Click to expand...
No no, not homemade. Gives an aspect ratio that I prefer, around 2.5:1

With this set up it acts like a MF sliding back, zero distortion not to mention weight saving

Anyway this wasn't about the shift capabilities, I'm more than happy with that, was more to do with the relation of the "wedge of focus" if the sensor was to be tilted instead of the lens
 
Last edited:
chris malcolm said:
Not if you end up with the camera pointing at the same thing, i.e., the same image in both viewfinders, whether you got there by tilting camera or tilting lens. When you adjust the tilt screws on the lens, they do the same ting to the sensor plane alignment with the lens, regardless of whether you happen to holding the body or the lens while you adjust the screw.
Click to expand...
+1. To get the same image using the lens mount one would have to tilt the body and then shift the tripod head...the end result is the same.

If you hold the lens steady and move the camera you are moving the camera's smaller FOV around inside the lens' larger FOV. The effect it has on the scene w/in the FOV (shift) is much greater if you allow the camera to move instead of the lens (because the camera is looking in a different direction it changes perspective). The wider the TS lens is, and the greater the tilt angle is, the more significant this becomes.

Basically, you have two FOV's and two Focal Planes. You have the lens' larger FOV and FP (Image circle and DOF) out front and you have the camera's FOV and FP inside the camera (the sensor). When you tilt the lens you shift the FP at the front and you use a different part of the larger FOV to project the same area/scene onto the sensor. When you tilt the body you shift the FP at the back and the camera is looking thru a different part of the lens' FOV at a different area.
 
Last edited:
sk66 said:
A TS Pano should almost be called something different than a pano. It's more akin to the Brenizer method than a standard pano.
Click to expand...
Yer, it's like a flat image rather than a sweep.
 
  • Like
Reactions: ST4
You must log in or register to reply here.
Back
Top