Reading mode: Light Dark

Why are straight verticals preferred? (TS_E lens theory)

Go to last post
jimcreative said:
So, when we view an image of a tall building taken with a standard lens there is a tendency for the mind to interpret the converging verticals as if the building is falling backwards and not straight. We sort of default to decoding the convergence as due to HORIZONTAL and not vertical distance. Funny, nobody looks at an image of a very WIDE building that has horizontal convergence and has an issue with it but the tall buildings look to be toppling backwards. This failure of our minds to rectify the two dimensional vertical data in a seamless manner (as it CAN do with horizontal convergence) has led to the development of first, the view camera system which allows for shifts and rises to maintain parallel verticals and, a bit later, the development of shift lenses for smaller formats.
Click to expand...
sometimes but when you are down below and not looking at a building from far off or centered the top is a LOT farther away from you than the bottom so it also makes it simply look farther away moreso than falling back, which it is. A very wide shot, taken from very low and very close and corrected makes the top look way large since you expect that it should cover a smaller portion of your vision since it is farther away.

anyway i think it depends a bit on teh details which looks the most 'normal' in a photograph
 
tko said:
Remember, it looks unatural because it is. Maybe a trained architect is used to it, but that doesn't make it natural. Tall objects, seen from the bottom, get smaller near the top.
Click to expand...
And that is the crux of your argumentation. What in the photo tells you it is seen from the bottom? Think hard, how do you know it has been taken from below and not from a hill some distance away?

Imaging taking the image from a hill some distance away, so that you are pretty level with the center of the building. You will have naturally straight lines since your distance to the building will be a multiple of the height of the building. And if you put such an image next to the one presented in the earlier thread, the outline of the building will be exactly the same. Two images with straight lines.

The problem really is that in the 17 mm T/S image you see elements of the facade of the building which clearly indicate that you look at the image from below, yet the outline of the building looks as if the image had been taken from some distance away on a hill (or rather as human eyes would see the outline when said human eyes were located some distance away on a hill). This mismatch, seeing elements from below, yet seeing an outline as seen from a distance, causes this reaction of things looking unnatural.
 
jimcreative said:
Otherwise, painters would by nature include converging verticals in their work... yet they habitually do not.
Click to expand...
One thing to keep in mind is that when a painter wants to paint a scene with tall buildings, it's rare that he gets so close that he needs to strain his neck to see the top. He looks from a greater distance, or at least imagines what the scene would look like from a greater distance. Ignoring converging verticals won't lead to "unnatural" effects in this case.

If we photographers could choose to use a telephoto lens whenever we wanted, we wouldn't have a problem with converging verticals, either.
 
Kuivaamo said:
One thing to keep in mind is that when a painter wants to paint a scene with tall buildings, it's rare that he gets so close that he needs to strain his neck to see the top. He looks from a greater distance, or at least imagines what the scene would look like from a greater distance. Ignoring converging verticals won't lead to "unnatural" effects in this case.

If we photographers could choose to use a telephoto lens whenever we wanted, we wouldn't have a problem with converging verticals, either.
Click to expand...
Look at this older photo of the Flat Iron Building, it is taken with roughly a normal lens from an elevated point of view (notice the roofs of the streetcars):



and compare it to this 'wideangle' view:

 
The old photo looks much more natural if you ask me. I can tell easily where is the point of view and distance the picture was taken. The perspective "corrected" version op showed has lost all the information and look like it's coming out of a PIXAR movie. Not that it is not pleasing but I think it is actually the more unnatural one.
noirdesir said:
Kuivaamo said:
One thing to keep in mind is that when a painter wants to paint a scene with tall buildings, it's rare that he gets so close that he needs to strain his neck to see the top. He looks from a greater distance, or at least imagines what the scene would look like from a greater distance. Ignoring converging verticals won't lead to "unnatural" effects in this case.

If we photographers could choose to use a telephoto lens whenever we wanted, we wouldn't have a problem with converging verticals, either.
Click to expand...
Look at this older photo of the Flat Iron Building, it is taken with roughly a normal lens from an elevated point of view (notice the roofs of the streetcars):



and compare it to this 'wideangle' view:

Click to expand...
 
The responses I am reading here tell me that I am talking to the wrong crowd.

Sorry. I have been shooting buildings and ONLY buildings for over a quarter of a century. I have worked with hundreds of architectural firms and photographed thousands of buildings. Without question and without exception, straight verticals are a strict requirement among designers and serious architectural photographers. Anyone who argues that straight verticals are more unnatural than buildings that lean back from convergence is simply out of touch with the real world of shooting structures.

I suspect that much of the resistance here is coming from those who do not own TS-E lenses and are trying to justify their own lens choices.

Just curious. Is there anyone here who actually shoots architecture for a living who will claim that I should have used a 16-35 II and let the verticals converge? If the answer is 'yes' then point my to your portfolio of professional architectural photography that is full of converging verticals and at least I will know that you are practicing what you preach.

My photos can be found at http://www.jimroofcreative.net

Feel free to visit the site and tell me what a bad job I am doing.

--

'Truth is stranger than fiction, for we have fashioned fiction to suite ourselves.' G.K. Chesterton

http://www.jimroofcreative.net
http://www.jimroofcreative.com
http://www.healthdesignphotography.com
 
jimcreative said:
The responses I am reading here tell me that I am talking to the wrong crowd.

Sorry. I have been shooting buildings and ONLY buildings for over a quarter of a century. I have worked with hundreds of architectural firms and photographed thousands of buildings. Without question and without exception, straight verticals are a strict requirement among designers and serious architectural photographers. Anyone who argues that straight verticals are more unnatural than buildings that lean back from convergence is simply out of touch with the real world of shooting structures.
Click to expand...
most of us probably are not architectural shooters
i've mostly done landscapes, sports and some other PJ-type stuff
jimcreative said:
I suspect that much of the resistance here is coming from those who do not own TS-E lenses and are trying to justify their own lens choices.
Click to expand...
not necessarily as I may ultimately go for the 24mm TS-E MkII
jimcreative said:
Just curious. Is there anyone here who actually shoots architecture for a living who will claim that I should have used a 16-35 II and let the verticals converge? If the answer is 'yes' then point my to your portfolio of professional architectural photography that is full of converging verticals and at least I will know that you are practicing what you preach.

My photos can be found at http://www.jimroofcreative.net

Feel free to visit the site and tell me what a bad job I am doing.
Click to expand...
i don't think anyone says you do a bad job at all

or that that look is not what is preferred for that job

but all the same in certain ways things can still look a little odd to others outside that realm
 
Kuivaamo said:
One thing to keep in mind is that when a painter wants to paint a scene with tall buildings, it's rare that he gets so close that he needs to strain his neck to see the top. He looks from a greater distance, or at least imagines what the scene would look like from a greater distance. Ignoring converging verticals won't lead to "unnatural" effects in this case.
Click to expand...
 
noirdesir said:
noirdesir said:
Remember, it looks unatural because it is. Maybe a trained architect is used to it, but that doesn't make it natural. Tall objects, seen from the bottom, get smaller near the top.
Click to expand...
And that is the crux of your argumentation. What in the photo tells you it is seen from the bottom? Think hard, how do you know it has been taken from below and not from a hill some distance away?

Imaging taking the image from a hill some distance away, so that you are pretty level with the center of the building. You will have naturally straight lines since your distance to the building will be a multiple of the height of the building. And if you put such an image next to the one presented in the earlier thread, the outline of the building will be exactly the same. Two images with straight lines.

The problem really is that in the 17 mm T/S image you see elements of the facade of the building which clearly indicate that you look at the image from below, yet the outline of the building looks as if the image had been taken from some distance away on a hill (or rather as human eyes would see the outline when said human eyes were located some distance away on a hill). This mismatch, seeing elements from below, yet seeing an outline as seen from a distance, causes this reaction of things looking unnatural.
Click to expand...
exactly, that is it and it also is why it looks like it is only 1 office wide, flat as a pancake and almost impossibly thin, when in real-life you see it is wedge shaped

not that the photo is not cool, but it does have a weird feel to it and not that it is not what the job requires
 
I've just been through a few of my books of reproductions in which architecture is depicted in paintings from Raphael through to Canaletto and Tiepolo.

No converging verticals in drawings and paintings meant to be viewed on a wall or a desk. But in ceiling frescoes, you bet!

Canaletto may have used a camera obscura for early drafts. If so he evidently placed the ground-glass screen vertical and shifted it relative to the lens/pin hole as in a view camera.

Perhaps:

Someone brought up in the great western tradition of architectural drawing sees converging verticals as distortion, but someone from the world of landscape photography sees vertical verticals as distortion.

If I ever get a T/S lens, say a 24 mm for use on full frame, I'll certainly experiment with the shift to give the perspective of a 50 mm, but not so far as to make all the verticals vertical. Maybe that will look best to me.
 
I am an architect by profession so I fully understand what you are talking about and I agree with you 1000%.

May be its easier for non architectural photographers to understand the strict requirment of vertically corrected images... its all because the building is the main subject of the photo, showing the true beauty of the building in its stable and static form, sometimes how it sits in the environment; No architect in this world would like his/her designed building looks like collapsing backwards...as simple as that!

Imagine if you are shooting a full portrait of a 6' tall beautiful model for cosmo or elle or whatever, would you shoot her from an angle lying on the floor, camera right next to her toe and pointing up!

jimcreative's 2 'handheld' pictures of Flat Iron and Chrysler Building is absolutely amazing, I wonder if I could ever have taken such perspective corrected pictures with a dslr in the urban context of Manhattan.

Exceptionally well done and thanks for sharing such eye openers with the 17TSE jimcreative.

Now I want one for myself, lol!

Imaging
jimcreative said:
The responses I am reading here tell me that I am talking to the wrong crowd.

Sorry. I have been shooting buildings and ONLY buildings for over a quarter of a century. I have worked with hundreds of architectural firms and photographed thousands of buildings. Without question and without exception, straight verticals are a strict requirement among designers and serious architectural photographers. Anyone who argues that straight verticals are more unnatural than buildings that lean back from convergence is simply out of touch with the real world of shooting structures.

I suspect that much of the resistance here is coming from those who do not own TS-E lenses and are trying to justify their own lens choices.

Just curious. Is there anyone here who actually shoots architecture for a living who will claim that I should have used a 16-35 II and let the verticals converge? If the answer is 'yes' then point my to your portfolio of professional architectural photography that is full of converging verticals and at least I will know that you are practicing what you preach.

My photos can be found at http://www.jimroofcreative.net

Feel free to visit the site and tell me what a bad job I am doing.

--

'Truth is stranger than fiction, for we have fashioned fiction to suite ourselves.' G.K. Chesterton

http://www.jimroofcreative.net
http://www.jimroofcreative.com
http://www.healthdesignphotography.com
Click to expand...
 
I'm with you on this one Jim, both as an architectural and landscape photographer.

Your two NYC shots are excellent and exactly as I would have chosen to represent the views - almost wish we had some really tall buildings in Leicester ;-)

I wrote an article several years ago on using T/S lenses and I regularly got people saying that it was overkill and you could do it all in Photoshop. As you'd suspect not usually people who actually produced shots for architects ;-)

The one below was taken from across the street from the Mariner's shop in Seattle (handheld TS-E24 mk1) No way I'd have been able to frame it with a normal wide angle and then 'fix' the verticals. I have used PS correction, but find the almost random loss of coverage once you crop the trapezoidal 'fixed' image back to rectangular is an element of unpredictability I'd not be happy with on a paying job



--
bye for now

Keith Cooper
 
Tizzio said:
Imagine if you are shooting a full portrait of a 6' tall beautiful model for cosmo or elle or whatever, would you shoot her from an angle lying on the floor, camera right next to her toe and pointing up!
Click to expand...
Well, if you shot from the same angle with a 17mm T/S lens on full shift, she'd look like Tarzan with tits.
 
And what were you expecting? Full agreement? There will always be people who will counter even the most reasoned argument.

I shoot architecture (but not for a living). Fully corrected is usually a requirement - but not all the time (certainly not for non-critical applications).

As has been pointed out, sometimes it is worthwhile to add a very slight amount of convergence to make a picture look more natural. Fully corrected can sometimes look over-corrected to the casual viewer. Towers, for instance, never look nice if they look "top heavy". There are two ways to get around that - stand back further so the correction needed is less or apply slightly less correction to create (or counter) the optical illusion. Architects do the latter often; look up "entasis" as an example.
 
Your portfolio is very impressive.

The only pic that looks unnatural to me is the flatiron. I'm guessing that that is mainly because the proportion of the facade that is 3D is a lot greater than other buildings and it gives the game away a bit!

I completely understand the desire for parallel verticals though!
 
Hmmmm....we call those balconies, lol.

Cheers!
**** England said:
Tizzio said:
Imagine if you are shooting a full portrait of a 6' tall beautiful model for cosmo or elle or whatever, would you shoot her from an angle lying on the floor, camera right next to her toe and pointing up!
Click to expand...
Well, if you shot from the same angle with a 17mm T/S lens on full shift, she'd look like Tarzan with tits.
Click to expand...
 
mainly on the choice of the word 'natural'.

there are the factual errors in your posts--for instance, very many paintings of paris buildings show converging verticals, if the perspective is from street level--and there are also errors of logic--it is illogical to cite your extensive training in a highly specialized niche genre as evidence of what ordinary people 'naturally' see--but the main problem, i think, is that you insist on saying that your highly artificial (ie, nothing ever encountered in nature) renderings are 'natural'. what does or does not appear natural is always subjective, and if something looks 'unnatural' to somebody, then you can't win by insisting that it really does look natural. if you have to explain why something looks natural, well, you've already pretty much admitted that it doesn't. you posit in your first post that ordinary viewers will find your shots with parallel verticals more natural, and then when many people disagree and report that in fact, they do not find them natural, you tell them they are wrong. that just doesn't make any sense.

look, your work is obviously excellent, your technical skills solid, and you are completely correct to study what your clientele desire and how to deliver it to them. no doubt to many architects, renderings absent vertical perspective really do look natural--to them. but that is a learned perception, not a 'natural' (ie innate or universal) one.

as some others seem to have already suggested, when your viewers are not architects, you might experiment with keeping a slight amount of convergence in pictures such as your flatiron. i suspect you will find that on balance, such versions won't raise as many questions about how 'natural' the shot looks-even though there is nothing natural about it, that is, even though they are strongly, but not perfectly, corrected for converging lines.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top