Tilt Shift Lenses on Digital Cameras

[g2imaging]

Here is the dilema... think it out.

 

Will a tilt shift EOS mount Canon lens work the same way on a

Digital (1d or 10d) as it will on a 1:1 ratio film body like the 1v?

 

My logic says no since the 1.6x1 or 1.3x1 ratios of the digital

cameras act like cropping a full frame image, thus using the center

area of the glass.

 

If the above is true, then by tilting and shifitnng a TS lens, you

must tilt and shift the lens more for the same effect. I like to

use a TS lens for selective focus and don't think it will work the

same as it would on a bellows camera like the Fuji GX680 2.25 film

camera.

 

Any experience or expertise?

[bobatkins]

Good question.

 

I presume the shift feature will be unaffected. If you shift to give non converging verticals, you'll need the same amount of shift whether or not you crop the image. Your field of view will be different of course.

 

Tilting will be a bit complicated since you have a different DOF for a cropped image, though if you tilt to get any particular plane in focus, the degree of tilt will again be the same whether of not you crop the image.

[g2imaging]

Bob,

 

Remember that the nature of the smaller digital chip is in and of itself a cropped image. With a non 1n camera, the light and image is focused on an image plane which is larger than the chip, thus the perceived ratio and zoom effect of a lens. The camera viewfinder compensates for this.

 

I'm still not sure the effect of a TS. I don't use it to correct lines but to selectively throw focus. Worth figuring out.

 

Thanks

[bobatkins]

Like I said, there's no difference in the amount of shift required to "deconverge" verticals and there's no difference in the amount of tilt required to convert a tilted object plane into the sensor/film image plane. Neither of these depends on film or sensor size, full frame or cropped.

<p>

There will be angle of view and depth of field differences. See my article <a href="http://www.photo.net/learn/optics/dofdigital/">Depth of Field and Digital</a>.

<p>

Shouldn't be too hard to do the calculations for DOF of a tilted lens, but I haven't tried to do them.

<p>

Off the top of my head, since you get less DOF on a 10D with a given lens than you would on a film body (see above referenced article), your tilt angles will be different from film depending on whether you're trying in contract or expand the DOF range about the focal point. You'd need less tilt to decrease it, more tilt to increase it. That's just a guess, I haven't run the numbers. I could be wrong, I haven't had my evening coffee yet...

[victoriaperelet]

I used Sinar for quite awhile. For last few years I used Canon 24mm TS lens (mostly for fashion:) on digital bodies.

After LF, you have to get used to TS. Because of telephoto lens design of 24mm lens nodal point in TS lens is shifted

from lens optical center, and standard LF focusing techniques fail. Actually because of that tilt also causes visible distortions

not only DOF. 45 focusing points on d bodies help a lot.

 

After you get used to it 24mm TS is very nice lens.

 

Victoria.

[g2imaging]

Thanks Bob... I guess actually hands on for me may be the best way to figure it out. Geometry was never my strong point. If anybody actually puts one on a digital body, please let me know how it turned out. I will try to get my hands on one as well.

[peter_phan]

I have the 24mm TS-E, 45mm TS-E, and a 10D. I haven't used the TS lenses in a while, but I didn't noticed that it operated any differently on the 10D than it did on my Elan IIe. Just that the field of view is narrower. I haven't used them side-by-side (Elan IIe and 10D), so I really can't comment on how much difference there may be, if any. Maybe when I have some free time, I'll give it a shot.

 

Keep in mind that just because the frame size is smaller doesn't necessarily mean you would have to tilt more. Tilting sets the angle of the plane of focus, and focusing moves that specific plane of focus closer or farther away. A smaller frame size doesn't change that. So if you've set your plane of focus at a 45 degree tilt from the film plane (mind you, this is the angle of the plane of focus, not the angle of tilt applied on the tilt lens), focusing would move that 45 degree plane closer or farther away. And if your film frame is cropped to 1.6x or 1.3x, it wouldn't make any difference other than you might have to set focus at a slightly different position.

 

Not sure how that would work for shift. I'll have to try it out to see the difference. Frankly, when using the TS's, usually eye-ball it, so I never paid much attention to the exact amount of shift or tilt applied on the lens anyways. I used to use a calculator to figure out how much tilt to use, but that was no fun, and didn't seem to give me any better results.

[Troll]

Isn't it important to have the light rays strike a digital sensor at 90 degrees, while light from the tilt/shift lens would be at a more acute angle?

[gary_ferguson1]

"Isn't it important to have the light rays strike a digital sensor at 90 degrees, while light from the tilt/shift lens would be at a more acute angle?"

 

I posted some shots a while back showing the results from a shifted 24mm T&S on a full-frame 1Ds. Bottom line? No problems at all.

[oskar_ojala]

My guess is that the lenses will work pretty much like they do on the film cameras, but the main problem is whether the narrower angle of view caused by the cropped image will suit your work (at least this is the issue for me.)