[MrA]

Staff note: This post may contain affiliate links, which means Pentax Forums may earn a small commission if a visitor clicks through and makes a purchase. If you would like to support the forum directly, you may also make a donation here.

---

Note: I resurrected this old thread of mine since my new post sort of follows on from it. New posts start about 8 posts down.

Here's a question I can't work out the answer to.

There exist tilt and shift adaptors for medium format lenses (say pentacon six), to 35mm mounts such as pentax-k. This means you can use MF lenses on full-frame, as it were, sensors, or film equivalent. Obviously there is a slight issue of the lens length - a 30mm lens at MF is a fisheye, whereas a 30mm lens designed for a full-frame camera is not.

You can also, of course, then use the same lenses with adaptors on aps-c sensors, but because they are two size-formats smaller, as it were, the effect of the lens lengths / crop factors is greater.

Why, then, can you not use lenses designed for 35mm cameras, with suitable adaptors, as tilt/shift lenses on aps-c cameras - why must you use lenses which are so much larger in format. Clearly when I use an old pentax lens on my dslr I'm cropping in on it by x1.6, meaning there is a whole periphery of glass, and of decent image, which I'm wasting.

Is there a reason an adaptor can't be made to allow "normal" lenses to shift or tilt on digital bodies?

---

Last edited by MrA; 09-08-2008 at 09:44 AM. Reason: Because I brought thread back from the dead.

[Zewrak]

Is there a reason an adaptor can't be made to allow "normal" lenses to shift or tilt on digital bodies?

I assume its because when you tilt the lens, you need bigger glass since you would otherwise see the edges of the lens? Speculating though.

[OniFactor]

because the adapter would increase register distance, and change the circle of confusion, focus, and throw everything out of alignment. 35mm lenses and 35mm cameras are engineered to be sharp, with a certain distance between the rear mount of the lens, and the film plane. when you change this distance without adding in extra optics (like a teleconverter), you change the size of the image projection. think about macro tubes, for instance. you make a larger image, by pushing the lens further away from the camera. but you can no longer focus to infinity, and you lose LOTS of light. make sense, now?

[ryan s]

If you decide to buy that adapter, let me know how it works out for you, since one is sort-of on my list

[MrA]

I assume its because when you tilt the lens, you need bigger glass since you would otherwise see the edges of the lens? Speculating though.

That was my asumption also - but the image circle of "normal" lenses is bigger than that of a DA / APS-C lens - that's what the whole "DA" stands for.

The "register distance" point, however, that sounds much more likely. Not something I've yet entirely got my head around, in truth (learning....!)

If you decide to buy that adapter, let me know how it works out for you, since one is sort-of on my list

I'm toying with it. If I can pick up an inexpensive MF lens I will certainly, but the very very wide lenses are neither common nor cheap. (student on budget here, I'm afraid!) Hence my (mild) frustration with crop factors and the reason for this thread in general.


Alas I see from a google-supplied list of all the difference "register distances" that the larger distances are all MF. With the possible exception of the T2 mount (tamron, right?)...


[edit]... for instance a T2 35mm f3.5 lens went for £4.99 on ebay recently. Maybe there's a DIY project there...

[nakey]

are you just looking for tilt/shift on pentax K, or is that you have a MF Tilt shift mechanism and want to use it on your K mount?

if it's the former, rugift has pentax K mount Tilt/shift lenses

[ryan s]

There is a K mount adapter on ebay to use Pentacon Six lenses...it's less than 100 bucks I think.

MrA: It has both to do with the register distance (MF is longer, and the adapter moves the lens away from the body), and the amount of surface the lens can cover (since the lens is moved further away from the center).

[MrA]

I'm resurrecting my old thread, because I've been pondering this some more (blame that other shift-lens thread in the General section), and this sort-of follows on.

MrA: It has both to do with the register distance (MF is longer, and the adapter moves the lens away from the body), and the amount of surface the lens can cover (since the lens is moved further away from the center).

I saw this image at Bob Atkins' webside. The outer circle is the standard image circle for a 35mm format lens. (He's talking about panoramic sensors but ignore that for now....)



If I understand correctly, then, every 35mm lens has a large amount of image circle that isn't being exploited when you use it on a aps-c body. [Obviously, with a Medium Format lens, that circle is larger still].

Given that Pentax K mount has a registration distance of 45.46mm, and Tamron Adaptall II / T mount have registration distances of 55mm (before adaptors), what then stops say a Tamron 24mm f2.8 or 17mm f3.5 being bodged into a perfectly workable shift lens?

I mean, assuming 10mm is enough to insert a mechanism to translate the lens, is there an optical reason it wouldn't work. Something I don't know about.

My understanding is that the physical size of the rear lens element (its diameter) isn't important - its the 'image circle' on the sensor which is important - and presumably this is constant for all 35mm-format lenses (or they wouldn't be full frame lenses).

Comments?

[eurostar]

Well, I DID built a tilt shift lenses for my K10D using an Olympus 18/3,5mm wideangle and a Hama tilt and shift bellow. I had to saw the back portion of the lens barrel to let only the front lens and diaphragm ring to sit outside of the bellow, while the rest of the lens live inside it, and also had to modify the bellow a lot to have the back of the lens at register distance from the sensor.

What I have discovered? That the coverage of the FF lens is still not enough for a large amount of tilt, and that there is not a lot of gain in depth of field as closing to f/8 you have plently of DOF.

So, I spent a lot for a very good lens and in the end I have simply killed it. But if you are in the game for narrow DOF, as in the Toy City effect, then it could be worth. Just, use cheaper lenses that an Olympus OM...

[MrA]

What I have discovered? That the coverage of the FF lens is still not enough for a large amount of tilt, and that there is not a lot of gain in depth of field as closing to f/8 you have plently of DOF.

Interesting. How about shift? Similar story?

Just, use cheaper lenses that an Olympus OM...

Indeed. I was pleased to see that copies of the Adaptall 28mm f2.5 regularly go for under £20 - so is probably ideal to mess around with. (The 17mm seems to be somewhat rarer, so is better pondered if the idea works...).

[Cambo]

Here's a question I can't work out the answer to.

There exist tilt and shift adaptors for medium format lenses (say pentacon six), to 35mm mounts such as pentax-k. This means you can use MF lenses on full-frame, as it were, sensors, or film equivalent. Obviously there is a slight issue of the lens length - a 30mm lens at MF is a fisheye, whereas a 30mm lens designed for a full-frame camera is not.

You can also, of course, then use the same lenses with adaptors on aps-c sensors, but because they are two size-formats smaller, as it were, the effect of the lens lengths / crop factors is greater.

Why, then, can you not use lenses designed for 35mm cameras, with suitable adaptors, as tilt/shift lenses on aps-c cameras - why must you use lenses which are so much larger in format. Clearly when I use an old pentax lens on my dslr I'm cropping in on it by x1.6, meaning there is a whole periphery of glass, and of decent image, which I'm wasting.

Is there a reason an adaptor can't be made to allow "normal" lenses to shift or tilt on digital bodies?

I would've thought they were relegated to the dustbins of history with all the perspective controls in Photoshop and the like, but they seem to be making a comeback. Some people claim really bad artifacts from perspective correction in image editors, but I have never found this to be the case if you are careful, and start with an oversampled image that is too large for your needs. I would think that the chromatic abberations and distortions from a tilt-shift lens would be far worse than anything Photoshop could do to an image; then there's the expense of the things.

Is it just people are lazy and would rather do the adjustments 'in camera', or is there some other compelling argument for these contraptions?

I must say, I still want to get a 4x5 one day just to shoot some Velvia, if it is still around. Lots of tilting and shifting available in those things.

Cheers,
Cameron

[ryan s]

Is it just people are lazy and would rather do the adjustments 'in camera', or is there some other compelling argument for these contraptions?

Rather the opposite...I'm too lazy to sit in front of the comp for hours and edit when I could just take a couple extra shots when I'm using the camera to begin with.

To me, the PS corrected image looks...PSed. I think about it this way, for example: If I can make a shot in focus IN the camera, the less work it needs to look acceptable. I do very few adjustments in post, unless I want to experiment and make an effect like Velvia...