Asymmetric movements - gadget or gimmick?

[Apple]

Paul,

Thanks for your comments - my image above isn't that complicated to set up (but don't tell the rest or they'll think anyone can do it! )

It's a case of getting everything squared up on the ground glass using your eyes and spirit levels - making sure you're in the middle of the scene so that everything is symmetrical around the vertical centre of the frame and that the ground glass is parallel to the subject. This eliminates perspective distortion. Rising front is used to get the height of the ceiling in and cut out more of the chairs in the foreground. This rising front is the only movement that is in addition to normal 35mm cameras with fixed lenses.

The rest comes down to exposure and printing / scanning...

If you want a read on focusing techniques that are more useful out in the field, have a look at these pages written by Harold Merklinger. The concepts might be a bit mind-blowing at first but the animations help. There's also downloads of the articles available for bedtime reading...

Andrew

[Apple]

Hello Paul,
I won't take that personally...

Hello Tim,
Nice to see you taking part on this one as you have much to contribute. I'll grant you the ease of use thing, but the money you save is a lot more. Prices new are almost at a ratio of 1:3 and if you consider the ready availability of second-hand Toyos vs 45SU, the ratio could be a lot steeper, say 1:8. So not inconsiderable for he or she on a budget.
Now, in your estimation, would it better to have the asym movements on the front standard for architectural work (we are agreed that landscape does not present a challenge)?

To Everyone else,
Has anyone noticed any of the trapezoidal distortion that Emmanuel has reported, when using tele lenses? If I understand rightly the effect was first noticed on a wide-angle lens (with a 35mm SLR), I am not sure how much effect one would get at longer focal lengths, say 360-500mm, with a tele on a LF camera.

Thank you.
Charles

Charles, correct me if I'm wrong, but you can't have asymmetric movements on the front standard as you don't focus the image there - by which I mean there's no ground glass at the front standard as it's the lens. You can only focus on ground glass or any other nominal thing you can put in its place to allow the creation of a sharp image. This means that the asymmetric movements are only available on the non-lens standard. It is perfectly possible to transfer these movements to the front standard (lens end) and then either re-zero the rear (to return to natural perspective) or to re-adjust the rear to create combined movements. I.e. the creation of asymmetric movements are only possible on the non-lens standard.

The Sinar P/P2 system is geared and has scales marked on so that accurate transfer of the rear settings is possible rather than iterative guesswork as used by Ebonyites etc. The front standard can mimic the rear but the asymmtric movements are technically on the rear only.

IMHO (disclaimer )

For the 'distortion' effect with longer lenses, is there a difference between tele and non-tele lenses? I don't know as the longest I have is 210mm which I don't use much, especially for architecture.

Andrew

[Apple]

Emmanuel, Charles et al.,

Looking through a Sinar camera system book, I came across the following couple of paragraphs:

"Deliberate under-correction

When we look at pictures taken from a very high or very low viewpoint we tend to compensate visually for converging verticals.

For instance, we can with direct or indirect shifts render a tall office block absolutely straight with truly parallel verticals. Yet in the picture, the book may look uncomfortably top-heavy.

It is an optical illusion - as you can verify on the print with a ruler. The same effect - but with apparent downward divergence - can arise in a view of a cubical object taken from above with an extreme downward shift.

Though an illusion, this effect must still be allowed for by taking perspective correction only far enough to convey a natural subjective impression. Such correction is thus slightly short of true geometric compensation.

"A rule of thumb

Use this rule for deliberate under-correction of parallelism:

Verticals within your usual field of view with the head held straight (i.e. a viewing angle of about 15-20° up or down) should always appear parallel in the image.

However, verticals that you can only take in by tilting the head up or down should not be fully corrected. Incline the image plane according to the extent to which you have to look up or down - and stop before the subject looks over-corrected."

That explains my top-heavy Heptonstall shot as I was at the extremes of the rising front to get the top of the steeple in without tilting the camera.

[Charles Twist]

Hello Andrew,
(i) I mean focussing using movements on the front standard while looking at the effect on a gg at the rear. Or is that utter nonsense? I agree that would be a serious change to the camera's design.
(ii) Useful info from Sinar, but I wonder how usable it is given that the image is upside down (and back to front), therefore making it difficult to analyse for illusions. Other than that, looking at a couple trannies here, what's interesting is that the illusion is only really obvious when there is a foreground leading into the towering subject. Even with movements to the edge of the image circle, a picture of, say, just a facade does not present problems. I am guessing the mind is fooled by the 3D to 2D effect. I have also observed a similar illusion with considerable use of shift: your mind doesn't know if you're looking at the subject side- or straight-on. It's like a portrait the eyes of which follow you around the room, I guess.
Regards,
Charles

[timparkin]

Charles, correct me if I'm wrong, but you can't have asymmetric movements on the front standard as you don't focus the image there - by which I mean there's no ground glass at the front standard as it's the lens. You can only focus on ground glass or any other nominal thing you can put in its place to allow the creation of a sharp image. This means that the asymmetric movements are only available on the non-lens standard. It is perfectly possible to transfer these movements to the front standard (lens end) and then either re-zero the rear (to return to natural perspective) or to re-adjust the rear to create combined movements. I.e. the creation of asymmetric movements are only possible on the non-lens standard.

Hi Andrew,

You could have asymmetric movements on the front standard if you want, you still focus on the gg though. It would be wierd that when you applied rise/fall of any sort, the line of constant focus would move around the screen. (i.e. if your tilt on the front was below the lens, the line that would stay in fixed focus would be above the ground glass. When you moved the ground glass up and down, the line would stay put.).

Also, asymettric tilt on the front would have the unfortunate affect of changing the focus too. This would also have the affect of moving the line of constant focus as you tilted.. far too complicated to get your head around which is why I imagine it wasn't included..

Tim

p.s. QUick question to Charles... Can you give me an argument for using your Toyo instead of buying a second hand ShenHao (which I've seen go for about £100)? I imagine the ShenHao is capable of getting the same result as your Toyo in most situations so why did you spend more on the Toyo?

[Charles Twist]

Hello Tim,
Thanks for the extra on the 45SU - I'll have to have a play with one to fully understand, I think.

Quick reply to Tim:
Because when I bought the Toyo, Shen Hao's didn't exist yet. Yes, indeed, why not buy a Shen Hao rather than a 45SU? Factors such as build quality come into this one, I guess. I wonder what would happen if Shen Hao decided to copy the asym feature as well; surely it can't cost that much to implement... The first disposable, whiz-bang LF camera I say this in jest as I believe quality has improved - couldn't have worsened, mind.

All the best,
Charles

[timparkin]

Hello Tim,
Thanks for the extra on the 45SU - I'll have to have a play with one to fully understand, I think.

Quick reply to Tim:
Because when I bought the Toyo, Shen Hao's didn't exist yet. Yes, indeed, why not buy a Shen Hao rather than a 45SU? Factors such as build quality come into this one, I guess. I wonder what would happen if Shen Hao decided to copy the asym feature as well; surely it can't cost that much to implement... The first disposable, whiz-bang LF camera I say this in jest as I believe quality has improved - couldn't have worsened, mind.

All the best,
Charles

I think you touch on something that is a big factor in a lot of peoples choices.. If I want a camera 10% more rigid I have to pay twice as much if I want another 10% more rigid I have to pay twice as much again.. At some point, depending on budget, we hit diminishing returns.. .. but .. the point where this happens is strongly affected by available budget and hence people with bigger budgets make the optimum choice higher in the camera price hierarchy.. The Ebony may only be 10% better (if you can quantify it) than the Toyo but some people are willing (and more importantly, able) to pay for the 10%.. I don't think the choice is made on asymmetric movements though (although I would honestly say that I would definitely miss them if they weren't there)..

The assymetric tilts offer a wizzy extra. That wizzy extra has a downside of about 3/4 lb though...

Tim

[Emmanuel Bigler]

Many thanks to Andrew-Apple for
#1/ presenting a superb image where something similar to Charles' image apperas, i.e. the shape of the vault in the chrurch generates an illusion that the pillars are diverging at the top, whereas they do not since the camera is perfectly well placed and adjusted.
I knew the effect for tall rectrangular buildings, I had never noticed before that something similar is to be expected inside churches. I mean : Gothic or Roman churches, our "Industry-standard" European model, not special items like Le Corbusier's famous Ronchamp Chapel (France)
http://cache.virtualtourist.com/2030073 ... nchamp.jpg

And thanks #2 for finding a reference in Sinar's textbooks, this is exactly what I have in mind and I appreciate the practical rule stating that if you have to raise (painfully) you head above a certain point so see the whole of the building, you'll probably have to refine the view camera setting with some "under-correction" of the classical rule for verticals (eventually this question of verticals is related to a question of cervicals ? )

To Charles : there might be two simple reasons why the effect I am reporting for wide-angle retrofocus lenses is hard to see with telephotos

- available LF telephotos have a pupillar magnification ratio which is closer to unity than the 24 mm TSE Canon
One of the most asymmetric vintage LF telephoto was the Voigtlander Telomar with a pupillar magnification ratio of .5 (1:2), but modern telephotos are not that extreme. This ratio is denoted by beta-prime_P in Schneider's technical brochures. For the recent apo-tele xenar, the factor is 0.75.
http://www.schneiderkreuznach.com/foto/ ... 55_1AD.PDF

- with a telephoto since you have to go quite far from the buiding in order to capture a reasonable portion of your subject , the angles you eventually have to apply to your optical axis are smaller than for a wide-angle retrofocus lens, placed at the foot of the buiding instead of far away.

But the experiment is easy to do with any telephoto if you can tilt the optical axis upward and keep the film parallel to the subject plane.
To the best of my knowledge (since I learned something today about unexpected assymmetric swings, I have to be more cautious than ever ! )
I have never seen 35mm tilting telephoto lenses, but many companies offer tilt+swing attachments for 35mm systems (including, since the last 2008 photokina, Arca Swiss with their F-line M2) ; and in MF+LF except the special Hasselblad tillting extender, or a telephoto mounted on the Rollei Sl-66, you have to use a view or a field camera with movements.

[Apple]

Emmanuel,

Unfortunately, my image you saw previously is one that doesn't exhibit the illusion - it only looks a bit like it due to the arches at the edges.

This is a very quick scan of the picture I meant with the illusion - a bit more obvious, I think

I've also wondered how much the positions and curves of the trees affect the perception of the steeple...



I've left the top of the picture in to show the amount of vignetting given that I was using a 90mm SA XL which has quite an image circle, i.e. I was pushing the rising front a bit too far...

Andrew

[Charles Twist]

Hello Andrew,
I hate to say this but the tower is a bit wider at the top - not much agreed, but it's there: maybe that reinforces the illusion which is at work in the Wells picture.
All the best,
Charles

[Charles Twist]

Hello (part 2),
I hope Joanna has her answer and will allow me to continue on the subject of distortion and illusion.
Firstly, before I get launched, if I understand you rightly, Andrew, you’re suggesting that the vignetting at the top of your second picture is due to the image circle. It won’t be that but more likely a filter holder or something such (bellows?). The image circle would appear at the bottom if you shift the lens up and if your lens is anything like the 210mm ApoSymmar f/5.6 which I have clipped a couple times, the line would be quite a bit sharper. You would also notice a fall-off in light of around 1 stop as you get near the edge (not sure how fast it will fall off, but I use a ND0.3 hard grad on the 210mm, which is probably equivalent to a soft grad on 90mm – thanks to Tim P. for his article on ‘graddage’).
Now on to the matter of distortion and illusion. Firstly (again!), I would like to return to Emmanuel’s cone illusion. Here be a picture taken with the above 210mm which shows it is not just wide angles which lead to the problem. And I don’t see the Sinar manual advice being much use in this case, given the subject matter – the rendition has to be die strait (now they were sultans of swing, or so they said ).



Secondly: a thought which is fresh for me but maybe plain and simple for others, in which case I apologise for wasting your time. It is to do with applying perspective corrections in camera versus in silico. If I am mistaken, please let me know. When one applies correction via the film standard in order to conserve verticality and/or horizontality, the corrections will be most strongly felt for objects nearer the lens. Because of distance and the diminution in apparent size of the object (diminution in real size of the image), the amount of correction will be greater in the foreground, in terms of absolute movement of the image. Now, in silico, the computer is dealing with a flat surface and will apply the same amount of pixel translation, whether the object represented is near or far. It seems to me that this is a further reason, why mimicking LF with an SLR, is not fool-proof.
Must buy myself a monorail (not a Shen Hao as Tim suggested). Any suggestions of one that is solid enough, with plenty of movements and yet can be carried be around in a rucksack? Oh and it must have spirit levels on the rear standard. And be cheap.
Best regards,
Charles

[Emmanuel Bigler]

Here be a picture taken with the above 210mm which shows it is not just wide angles which lead to the problem.

Superb example, Charles of the illusion I was speaking about !
And without a wide angle lens !
Now to come back to the starting point of this discussion, I doubt that asymmetric tilts could have any influence about this effect

a monorail (not a Shen Hao as Tim suggested). Any suggestions of one that is solid enough, with plenty of movements and yet can be carried be around in a rucksack? Oh and it must have spirit levels on the rear standard. And be cheap...

I have some suggestions regarding this choice, but I prefer to discuss about it off-line

[jb7]

That is a good example of the illusion-

In this case, I think I can figure out why-
I think it's the dihedral lines in the centre of the picture, which give the overall picture a pinched appearance-

without those lines, I think the verticals might read more, well, vertical-

I don't think it's to do with the projection-

Perhaps we should borrow from the columns of the ancient Greeks,
and give our frames a slight barrel shape...

joseph

[dennis]

Fascinating thread this one, but surely 'perspective' proper only really changes with a change of viewpoint. What happens within the camera is just a distortion/correction of existing perspective? Dennis (happy with metal Toyo 45A)

[Emmanuel Bigler]

Hello all !

I take the liberty to make this discussion re-surface, since the last monthly delivery of http://www.galerie-photo.com has an interesting example of the "ice cream cone illusion", in the work of the Chinese large format photographer Xiao Chuan SUN

Ze article (hin frennche, sorri)
www.galerie-photo.com/xiao-chuan-chine-bouleversement.htm

Ze image
www.galerie-photo.com/images/xiao-chuan-sun-09ok.jpg

Baille ze ouais, zis meunsse, zère hize hanne article in hingliche, baille Daniel W. Fromm euh-baoutte Apo Nikkor laine-ssizz.
www.galerie-photo.com/apo-process-nikkors-en.html