Talk:Perspective control lens

Requested photos

The diagram is nice, but having actual images for both tilt, shift, and tilt/shift (vs. uncorrected). Cburnett 02:49, 20 October 2007 (UTC)

I agree that images showing the effects of shift might help, but at least so far, this article hasn't really been about tilts or swings. I think illustration of tilt might be more appropriately handled by expanding the article on the Scheimpflug principle. If is decided to include that topic, the title should be changed to tilt/shift lens with a redirect from the current title.

I'd still hesitate to get into combined tilts and shifts, because doing so would more likely confuse than educate. With many tilt/shift (PC) lenses (e.g., Canon and Nikon), the tilt and shift movement are at 90° to each other, and many of the examples I've seen of combined movements are rather contrived. The Canon (and perhaps also the Nikon) lenses can be modified to have both movements in the same direction, but illustrating the effects of modified lenses might be going too far. JeffConrad (talk) 22:40, 1 June 2008 (UTC)

Digital Photography

It would be good if someone could relate which (if any) digital SLR systems have compatibility with shift lenses, given how few there are. Or if none, simply state that fact. 82.10.108.49 (talk) 21:53, 26 February 2008 (UTC)

Probably hard to summarize - generally the viewfinder overhangs the mount, so mechanical collisions prevent use of the full tilt/shift ranges - not a simple yes/no. Too specific for an encyclopedia - see lens reviews ? --195.137.93.171 (talk) 00:46, 4 April 2011 (UTC)

I agree that it would be mighty complex (see the manual for the PC-E Nikkor 24 mm f/3.5; that manual, of course, mentions only bodies that were available at the time of its publication). And it’s not so much an issue of digital as it is one of bodies with built-in flashes; some film SLRs with built-in flashes don’t allow the full range of movements. For example, the Canon EOS 5/A2/A2E allowed full shift, but at full shift, the shift direction could not be rotated without the lens hitting the pentaprism/flash housing. In most cases, the tilt function isn’t restricted. JeffConrad (talk) 08:18, 4 April 2011 (UTC)

Available lenses

Perhaps I’m being too fussy, but the description of the new Nikkor lenses was reading more like a Nikon brochure than an encyclopedia article. The buzzwords like “electromagnetic diaphragm” and “CMOS” apply equally to Canon, but they’re irrelevant to this article (as probably is “FX-format”, as well), serving only to clutter things up. I also question the statement that only the full-frame formats use the full capabilities of the lenses. It’s true that smaller formats (such as Canon’s 1.6 factor and Nikon’s 1.5 factor) don’t use the full image circles, but they afford some advantages as well (e.g., the shift is a greater fraction of the format size). Arguably, the greatest difference is the narrower ange of view, which sometimes is a problem and other times is an advantage. NPOV would seem to require comparable treatment of Canon and Nikon (and any other) lenses, and throwing in all the buzzwords forces the reader to wade through all sorts of material that has little real bearing on the substance of the article. JeffConrad (talk) 04:46, 4 July 2008 (UTC)

I suppose there could be some point in mentioning the electromagnetic diaphragm on the PC-E Nikkor lenses if their preset operation on cameras other than the D3, D300, and D700 were contrasted with the mechanical preset operation of the PC Micro-Nikkor f/2.8D (non-E). I’ve added this to the article. If this information seems superfluous, get rid of it. I still recommend eliminating the last sentence in the paragraph; this article is about lenses, not camera bodies. JeffConrad (talk) 07:41, 4 July 2008 (UTC)

I've moved the discussion of aperture control to a new section to lessen the clutter in describing available lenses. I'm not sure this material ultimately belongs here, although I'm also not sure where it does belong—perhaps in its own article. I'm a bit surprised that there is almost no mention of this in WP, but I suppose that, aside from PC/TS lenses, it's little more than a historical curiosity. JeffConrad (talk) 04:26, 7 July 2008 (UTC)

Article title and subject

The vast majority of lenses/adapters mentioned in this article include tilt functions as well as shift, so the current title is somewhat misleading. Should the page be moved and expanded accordingly? Currently, Tilt-shift lens redirects to Tilt-shift photography, which doesn't make sense (a lens is not a technique). JeffConrad (talk) 02:18, 8 July 2008 (UTC)

Shouldn’t the term used in the gallery match the article title? Using “perspective correction” one place and “perspective control” in another just confuses the reader. Either term is workable, but the former may be preferable because it is more supplier neutral. JeffConrad (talk) 21:49, 8 July 2008 (UTC)

"Perspective control" is better. These lenses are used control perspective, not always to correct it. --Stybn (talk) 05:14, 10 July 2008 (UTC)

“Perspective control” gets a few more hits on from a Google search, and is arguably the more descriptive term. On the down side, only Nikon and Schneider currently use it, so it’s not the most neutral. An argument also could be made for “shift lens”, which accurately describes what the lens does, without much subjectivity. In any event, the terminology should be consistent, or perhaps two galleries are needed—I've never heard anyone refer to a Canon TS lens as a “perspective correction” lens or as a “perspective control” lens, so the current grouping is a bit confusing. We’d then need to think about which group would be appropriate for the most recent Nikkors ... Eventually, I think the article will need to be moved to “tilt-and-shift lens” or something similar, because the majority of currently available lenses provide both movements. Although it’s possible to create a separate page for TS lenses, the inherent duplication would be hard to justify, and there would almost certainly be a quick request for merger. JeffConrad (talk) 06:07, 10 July 2008 (UTC)

I agree with Stybn. These lenses all control perspective. "Shift" and "tilt" are how the lenses do that. I would focus on the end, not the means. I think the article should be renamed "Perspective control lenses." Perhaps the "Perspective correction" article should be similarly retitled. Split this one into two articles? I do not think the subject is expansive enough or of enough general interest to warrant that. How many "hits" does this article get? I doubt it is one of Wikipedia's stars. Until Jeff showed up, it was virtually ignored. And with DSLRs and imaging software it is of even less interest to small-format photographers. Finally, "neutral" seems a political term to me and not particularly relevant. Motorrad-67 (talk) 12:20, 10 July 2008 (UTC)

Statistics: [1] -- 2718 page views in May 2008: not a small amount. As for the article title, I don't see what's wrong with it as it currently stands; there are for either "correction" or "control" and perspective control lens already redirects here, so anyone searching for the that is still going to get to this article. I also don't think "tilt-and-shift lens" is a good title because not all of the lenses tilt, even if most (or all) modern ones do now. howcheng {chat} 16:39, 10 July 2008 (UTC)

Perspective is actually controlled by the camera back; shifting a lens simply allows the line of sight to be changed so that the back can remain parallel to the subject (or tilted further away from it). Opposite shifts are also sometimes used to create panoramas; I wouldn't call that controlling perspective. In this sense, “shift lens” is most descriptive of lenses that provide only shift functions and “tilt-and-shift lens” the most descriptive of lenses that provide both functions.

Howcheng makes a couple of good points: the current title is about as common in the photographic literature (e.g., Sidney Ray's Applied Photographic Optics) as is “perspective control” (Ray seems to use both interchangeably). Google hits aren't quite as evenly distributed: “tilt-shift lens” gave about 92,000 hits, “perspective control lens” gave about 19,000, “tilt-and-shift lens” gave about 8,000, and “perspective correction” brought up the rear at just under 3,000. I wouldn't make too much of the results, though, because they change daily and all show this article at or near the top, so any changes to it or its redirects probably have a strong influence. In any event, there is something to be said for consistency with the Perspective correction article; I can't really see changing that title because “correction” is by far the more common term when the technique is applied after the time of exposure. The point about naming is also valid; the problem, I think, is that there really isn't a name that applies to both types of lenses—I've never heard anyone refer to a lens that provides only shift as a “tilt-shift lens”, and I've never heard anyone refer to a Canon TS lens as a “perspective control” lens. The jury is still out on the new PC-E Nikkors; the PC designation is strictly historical, and the full lineup is still too new to see what people will call them. For the time being, perhaps it's best to leave the article title as it is. I'll look to adding some material on tilt and we can decide what to do at that time; we'll need to decide whether the primary thrust of the article is historical or current. I suspect that splitting the Available lenses section into two parts will be necessary; I'd probably classify the current Nikkors as “tilt-and-shift lenses” lenses, but it's probably not worth arguing about. For the time being, I'd change the gallery title back to match the title; eventually, we may need two galleries (and some additional images).

I agree with Motorrad-67 that interest in lenses with only shift functions is waning for two reasons: 1), there aren't many still available, and 2), convergence can be fixed via software (albeit with some loss of quality). This is consistent with Nikon discontinuing their 28 mm and 35 mm PC lenses. But I don't agree about the lack of interest in lenses with tilt functions. Although the effect of tilting the plane of focus away from the subject can be emulated in software, orienting the PoF to coincide with the subject cannot be done in software. Focus stacking can sometimes be used, but the effect isn't quite the same, and it's usually limited to subjects that don't move. Arguably, then, some mention of “tilt” is by far the most descriptive. With the move away from view cameras to small- and medium-format digital capture, there actually seems to be increasing interest in lenses that provide movements—those accustomed to movements are often loathe to give them up. Nikon's introduction of three new lenses and Hasselblad's announcement of a tilt-and-shift adapter are hardly consistent with waning interest; though I doubt that any will become best sellers, I can't imagine either company making the rather substantial investment without expectation of some sales.

As for “neutrality”, it's absolutely relevant if an article title uses one manufacturer's product name. Perhaps this is better illustrated by practice in standards and regulations, where the use of trade names is scrupulously avoided. I once was involved with the drafting of a regulation that used my company's trade name in a requirement (it wasn't my doing ...); although it was flattering, our competitors obviously were not amused, and a couple threatened legal action. I worked to change a few words to make the requirement generic, and everyone lived happily ever after—I don't think we lost any sales, and I got a lot less grief at subsequent trade conventions. So yes, “neutrality” is relevant. In that sense, the current title is a good choice, even if it isn't the most descriptive. For the record, I've owned a shift lens, two PC lenses, and three TS lenses. JeffConrad (talk) 20:53, 10 July 2008 (UTC)

This article is about perspective control but this is only half the story for a tilt-shift lens so why is it re-directed here? There needs to be some description of the tilt effect which is mainly to control focus. A tilt-shift article is needed. Unfortunately the tilt-shift photography article has morphed into an article on the photoshop technique. Kwenchin (talk) 23:21, 28 February 2010 (UTC)

Absent a real article on tilt/shift lenses, the redirect is probably better than nothing. As I hinted above, an article solely about perspective-control lenses in 2010 is primarily for historical interest—I'm not sure any shift-only lenses other than the Schneider are still manufactured. As I mentioned, there would seem two reasonable ways to handle this:

1. Expand the article to properly cover tilt as well as shift. Because “perspective-control” is a subset of tilt/shift, it would seem reasonable to move the page to the latter and redirect from the former.
2. Create a separate article about tilt/shift lenses, incorporating some, but not all of the material here. This would certainly be a way to avoid arguments about the title and content of this article, but there would necessarily be a fair amount of overlap, and I'd be surprised if someone didn't quickly propose a merger. But whatever works.

As I have said many times, I wish I knew what “tilt-shift photography” really meant. Without some reliable sources, it's tough, as with any neologism. Though some folks do seem to use it to describe the use of camera movements on small- and medium-format cameras, the preponderance of usage seems to refer to “miniature faking”, and perhaps in most cases, faked with Photoshop, as Kwenchin has mentioned. The role of WP editors is to document rather than critique, so I've tried to have the article on Tilt-shift photography reflect what most people take it to mean. Perhaps the article on Tilt-shift photography can evolve if we find some reliable sources. But I'm not sure it's the place to cover tilt/shift lenses. Camera movements are probably best covered in the the View camera article, or perhaps in a format-independent article devoted solely to that topic. A proper article on tilt/shift lenses would necessarily need some mention of movements, but could do so more briefly, and concentrate more on the lenses than the techniques. JeffConrad (talk) 01:31, 1 March 2010 (UTC)

Lens Gallery

Perhaps once again I'm being too touchy, but this is starting to look more like a personal essay than an encyclopedia article. JeffConrad (talk) 21:53, 31 July 2008 (UTC)

A "lens gallery" could be a personal essay? I do not understand. Is it because I added one photo showing that a shift lens can now be used on a DSLR? Note this comment from above above:

It would be good if someone could relate which (if any) digital SLR systems have compatibility with shift lenses, given how few there are. Or if none, simply state that fact. 82.10.108.49 (talk) 21:53, 26 February 2008 (UTC)

Let's see. I have three photos in there -- a Minolta, a Pentax, and a Nikkor. I guess you are not talking about my personal essay. Motorrad-67 (talk) 22:06, 31 July 2008 (UTC)

There's certainly nothing wrong with describing system compatibility. I think common sense suggests that if the mount hasn't changed, the lens is compatible, but a mention in the text still might not hurt. I wasn't so much talking about the photos themselves as the most recent caption; it's more like an advertising puff piece than something for an encyclopedia. Moreover, the implications in the text and in the caption that there is lesser compatibility with less-than-full-frame sensors are just plain wrong—that's what I mean by “personal essay”. But again, maybe I'm just too fussy. JeffConrad (talk) 23:44, 31 July 2008 (UTC)

I was actually starting to think that the lens gallery is getting too large. For the most part, each lens looks pretty much the same. Maybe we could whittle this down to four that show the widest variety between the lenses and put the rest on Commons? howcheng {chat} 02:50, 1 August 2008 (UTC)

Point well taken. If we do trim the gallery, we should try to treat all manufacturers equally, in terms of what's included, as well as the section title: either “Perspective correction lenses” to match the article title, or “Perspective control, tilt, and tilt/shift lenses” to reflect what's actually included. If I had to pick four, I take one of the older PC-Nikkors, the Canon, the Pentax Shift 6×7, and the Olympus, because it's a reasonably representative list and the backgrounds are reasonably harmonious. If the most recent addition remains, the reference to full-frame should be removed, because it incorrectly implies that the lens isn't compatible with crop-format Nikon DSLRs. There actually is an issue with the tilt/shift Nikkors on some Nikon bodies—if the body has a built-in flash, the pentaprism housing interferes with rotation at maximum rise (Canon have the same problem)—but it has nothing to do with sensor size. I don't know if the problem occurs with the older PC-Nikkors. JeffConrad (talk) 08:07, 1 August 2008 (UTC)

There is no problem with older PC-Nikkors hitting the prism housing on any Nikon SLR or DSLR, with the exception of the very first f/3.5 35mm PC-Nikkor of 1961, before Nikon extended the lens base to clear the housing on the FT/FTN series.

Personally, I see no problem with showing the reader in the gallery the full range of shift lenses historically available from all manufacturers. Indeed, there are lenses not illustrated in the gallery that could be if photos of them were available. It adds to the encyclopedic nature of the article. I was not aware of a limit on gallery sizes by policy or esthetics. However, I find that my opinion is usually in the minority. Motorrad-67 (talk) 12:04, 1 August 2008 (UTC)

I usually counter this argument with Wikipedia is not an image repository. We don't need to have all of these images when the reader has a difficult time telling the difference between them. Galleries have their place in articles, but not in cases like this. Tortoiseshell cat makes good use of galleries to show the wide variety of tortoiseshell/calico patterns, and I just added a gallery to Casa de Estudillo to highlight various aspects of the property -- in both cases, there are still even more photos on Commons. howcheng {chat} 16:37, 1 August 2008 (UTC)

I tend to agree with howcheng. Although I'm not sure there are specific guidelines for the number of images, the general statement still governs. It's always a somewhat subjective call; we had a similar issue in the Depth of field article, which kept accumulating images that were largely gratuitous and served more to clutter and distract than inform. Using the same rationale as howcheng, we trimmed the gallery.

I did not mean to imply that there's any specific Wikipedia policy on aesthetics; what I suggested was my opinion, for whatever it may be worth. I, too, find it in the minority more often than than I would have it. JeffConrad (talk) 22:35, 1 August 2008 (UTC)

Zenitar lenses

Can editor 199.227.143.21 provide a source for the availability of K-mount Zenitar tilt/shift lenses? I see MC Arsat versions of these lenses, on both the Arsenal site and the Hartblei site, but can find nothing on the Zenitar site. Whatever the source, the other available mounts should also be mentioned. JeffConrad (talk) 07:44, 1 August 2008 (UTC)

Compatibility of Canon TS-E and Nikon PC-E lenses

There are serious compatibility issues, beyond those stated or implied in the article, with Nikon PC-E lenses and various Nikon camera bodies. Automatic aperture control is only available with the D300, D3, and D700 bodies; some other bodies support the electronic preset aperture control via the lens pushbutton, but apparently some do not. Ken Rockwell describes these problems in considerable detail; I've had only brief contact with a 24 mm PC-E lens, so I can't confirm his findings. On some camera bodies with built-in flash, the lenses cannot be rotated if maximum rise is employed; on others, it may not even be possible to set maximum rise without hitting the bottom of the prism housing.

Canon TS-E lenses provide automatic aperture control with any EF-mount bodies, but clearance of the prism housing is an issue on some DSLRs with built-in flashes.

To my knowledge, neither Canon nor Nikon have bothered to identify these incompatibilities. I'm not sure the responsibility falls to this article, but adding at least a general caveat might be a good idea. I'm lukewarm on including a link to Rockwell's article, because I've found his writings to be uneven in quality and even accuracy. Currently, though, I'm not sure what else is available, so perhaps a footnote mentioning an external link would point the reader to the article without the implication of endorsement. I know of no similar article that examines clearances on Canon DSLRs. JeffConrad (talk) 21:29, 31 August 2008 (UTC)

The article to which Jeff refers: Ken Rockwell article Motorrad-67 (talk) 21:35, 31 August 2008 (UTC)

Lost resolution in the more distant areas of the subject

yes, you lose resolution in the distant areas because they're recorded on a smaller area of film

I suppose it's possible to look at it this way, but it's mighty unusual statement. And even if one were to do so, this would be true for any lens; what in the world does it have to do with perspective control? And what is “lost depth of field due to the angle of the film/sensor plane to the subject”? Tilting a lens changes the orientation of the plane of focus and changes the shape and orientation of the depth of field, but there is no “lost” depth of field. And again, what does it have to do with perspective control? JeffConrad (talk) 03:32, 1 September 2008 (UTC)

Perhaps I should have clarified my comment on the first issue: obviously, it's more difficult to resolve details on an object at great distance from the camera than it is to resolve those details on the same object close to the camera. What's unusual is to say that resolution in distant objects is “lost”; the resolution was never there. I cannot imagine anyone thinking that software perspective control would somehow improve resolution in distant objects; unless I've really misunderstood the intent, I cannot see why the statement I deleted belongs in this article. JeffConrad (talk) 04:01, 1 September 2008 (UTC)

To use Motorrad-67's photographs as an example, applying a digital distortion to photograph #2 might make it looks like #3, but #3 will have more detail in the distant areas of the subject (the roof) because they were originally recorded over a larger area of film. Same holds for the fact that a distorted version of #2 will never have its field of focus perpendicular to the ground, whereas #3 does. I'm just trying to drive home the point that digital stretching isn't a substitute for movements. --Stybn (talk) 14:41, 1 September 2008 (UTC)

We're in agreement; I obviously misunderstood that sentence—the “loss” is in relation to what would have obtained by using rising front. You're really saying much the same thing as I was trying to say with “pixel interpolation”. I still find the reference to “loss” confusing, because the information never was really present; I think the article would be more clear if the description were something like

The results are not the same as those obtained from shift movements, however; parts of the image that are stretched require interpolation of information that was not present in the original image. Additionally, the depth of field remains perpendicular to the original line of sight rather than perpendicular to the image plane as would be the case if shift movements had been used.

This needs a bit of work, but I think it gives the general idea. In light of your explanation, I moved my comment about the inability of post production to duplicate the effects of tilts or swings to a separate paragraph to avoid confusing two different issues. There is some question whether it's even needed, but we again get back to whether this article is primarily about perspective-control lenses of yore or about current lenses, nearly all of which include tilt/swing as well as shift. JeffConrad (talk) 19:26, 1 September 2008 (UTC)

Aperture control

Most of this section really doesn't belong in this article; I originally added it only because there wasn't another good source to explain what an automatic diaphragm was. If there is no objection, I'll move most of this section to the Aperture article, leaving only material that specifically relates to tilt/shift lenses. JeffConrad (talk) 00:14, 12 December 2008 (UTC)

Use of shift in aerial photography

The paragraph on using shift to negate the effects of motion blur in aerial photography is nearly incomprehensible as well as unsourced. Absent a reliable source, as well as a better explanation, I think this material should be removed. JeffConrad (talk) 03:47, 4 September 2009 (UTC)

I agree. I've never heard of anyone doing that myself. howcheng {chat} 04:37, 4 September 2009 (UTC)

Shifting the lens to compensate for ground speed is how spy planes (including the U2) take photographs without needing exceedingly ridiculous and physically impossible (1/100000th or faster) shutter speeds. I'm sure any reference on the U2 would include this information. NetJohn (talk) 22:50, 28 September 2010 (UTC)

I must admit this is an area with which I'm unfamiliar, but even if correct, it sounds different from how we use shift here (it sounds more like image stabilization). In any event, it would need a better description and support from a reliable source. JeffConrad (talk) 01:02, 30 September 2010 (UTC)

“Tilt-shift miniature faking”

The linked article Tilt-shift miniature faking doesn't really have anything to do with the use of tilt; it's almost as if that article attempts to describe digitally faking a technique that doesn't really fake miniatures. In the second paragraph under Techniques, that article states

It is perhaps improper to refer to this technique as "tilt-shifting".

That statement is correct for two reasons:

1. There is no such thing as “tilt-shifting”, and use of that phrase as a noun or compound adjective almost invariably reveals a person who doesn't understand camera movements. What's employed in “miniature faking” with a TS lens is tilt.
2. The shallow DoF that obtains from using tilt to rotate the plane of focus doesn't look anything like the shallow DoF that obtains from close-up photography.

The emphasis of that article is a bit difficult to discern, but it appears to describe a technique (e.g., Smallgantics that actually does simulate DoF that obtains from close-up photography. JeffConrad (talk) 09:09, 7 September 2009 (UTC)

{{advert}} tag added on 25 March 2010 by 90.195.131.21

The section on lenses doesn't strike me as an advert; perhaps the sentence on the 35 mm PC-Nikkor could be slightly rewritten so it sounds less like bragging rights, but otherwise, I'm not sure how this section could be significantly changed and still adequately describe the lenses. IP editor made no comment, so it's impossible to address his objections; perhaps he doesn't understand the significance of things such as automatic diaphragms. In any event, unless the editor can indicate why the section isn't neutral, I recommend we remove the tag. JeffConrad (talk) 23:02, 25 March 2010 (UTC)

Done. Motorrad-67 (talk) 23:44, 25 March 2010 (UTC)

I've updated the section to reflect current status (except for the Leica S 30 mm tilt/shift, for which I have no information). I've put the manufacturers in alphabetical order so that we don't imply that Canon and Nikon rule the world. I've also removed mention of the PC-Nikkor 35 mm; it's no longer available, and it's already mentioned in the lead section. Likewise, the image of the PC-Nikkor 35 mm is already included in the gallery. JeffConrad (talk) 02:06, 26 March 2010 (UTC)

I agree that the 1961 PC-Nikkor should be shown—I erred in saying the image in the gallery was of that lens. But the lens should be shown where it's discussed, not in the section on available lenses. And arguably at least as significant is the world's first tilt/shift lens, because all but one of the current available lenses includes tilt. Where do we stop? If we want to include every shift lens ever made, we should re-title the section or create a separate section for discontinued lenses. JeffConrad (talk) 06:07, 26 March 2010 (UTC)

I agree with sizing the image to match the diagram. Though normal WP policy is “let the user decide”, common sense must prevail. Here, I think aesthetics dictate.

On a side note: the left-hand side of the top of the figure isn't correct—it should show the limit of the angle of view cutting off the top of the building. Any chance of getting this fixed?

Upon closer examination, I see that the rays correspond to the building edges rather than the angle-of-view limits, so the image is fine as it is. JeffConrad (talk) 03:46, 29 March 2010 (UTC)

Editor 90.195.131.21, who added the {{advert}} tag, further explained the edit:

Hi there,

I'd made a brief remark in the comment when I'd added the advert tag, or at least I thought I did. Re-reading the article, I'm not sure it's so much advert that is wrong with it, it's just far too much detail. I kept reading yet another paragraph about another manufacturer's offering and thinking "so what?". I wouldn't expect that under an article for say, prime or zoom lenses. I know these are very speciailised bits of kit with an interesting history, but I only think that lenses of note should be described, with perhaps a link to the manufacturer page (on wikipedia) for the other players.

Cheers, sorry I did'nt explain that at the time 90.195.131.21 (talk) 20:34, 26 March 2010 (UTC)

To an extent, I think the editor has a point. On the other hand, I think PC and TS lenses are somewhat different beasts, because they incorporate, to varying degree, special features unfamiliar to many photographers. And I also wonder if we ever could agree on which lenses are “notable”; for me, the list might include

• Nikon PC-Nikkor 35 mm (1961): first shift/PC lens
• Canon TS 35 mm (1973): first tilt/shift lens
• Canon TS-E 24 mm, 24 mm, and 90 mm (1991): first shift or tilt/shift lenses with automatic aperture control
• Hartblei Super-Rotator: first tilt/shift lens with independent rotation of tilt and shift functions

A case might be made for including the original Nikon PC Micro-Nikkor 85 mm lens as the first “macro” TS lens, but I think its 0.5 magnification is simply an incremental gain from the Canon TS-E 90 mm lens's 0.29. Of course, others well might disagree.

Perhaps we could say something like “24 mm, 45 mm, and 90 mm PC-E lenses” rather than giving all the details. I'm not sure the maximum aperture is of interest to most readers, but I do think features such as automatic aperture control and independent rotation of shift and tilt are significant and merit mention.

Thoughts? JeffConrad (talk) 22:35, 26 March 2010 (UTC)

Hi there, agree with most of that (except the suggestion I don't understand diaphrams!). I think the section is may be a bit the wrong way round, discussing the features that are particularly relevant, and how they have evolved with examples of the first lens incorporating that and/or the first one that had market success would be easier to read and more informative.

Same with the PC/TS distinction.

I don't know whether even listing the focal lengths is that necessary, other than to show where choices have emerged, or a particularly peculiar offering is.

I feel it's just far too much detail for this article and the point gets a bit lost. It is interesting information, though, Maybe that should be placed in an article specific to the manufacturer's product line and simply linked over? I'm thinking this sort of location http://en.wikipedia.org/wiki/Canon_EF_lens_mount#List_of_EF_lenses

90.195.131.21 (talk) 23:49, 26 March 2010 (UTC)

Automatic aperture control is explained in the section on Aperture control. Because almost all current lenses, including Canon and Nikon TS lenses, include it, it's perhaps becoming mainly an issue of historical interest. But the third-party lenses still don't provide it; it's not really a big deal until you don't have it. Even then it's manageable, but it entails several extra steps, and the photographer must remember to stop the lens down to working aperture before taking the picture. So I think it's important to know which lenses have it.

The distinction between “PC” and “TS” often isn't significant for architectural photography, but it certainly is for landscapes, where tilt is usually more important than shift. As I've said several times, because almost all current lenses provide tilt as well as shift, “perspective control” is an artifact of an earlier era, and remains in use mainly because of Nikon branding—I've always preferred the more descriptive “shift” to “PC”, as have many manufacturers. But perhaps this is simply an argument for a separate article on tilt/shift lenses.

I think a case can be made for a general historical summary of important developments. But I think a case can also be made for leaving things much as they are, perhaps with the elimination of some detail as I suggested. I honestly don't have strong feelings either way. JeffConrad (talk) 02:46, 27 March 2010 (UTC)

Edits of 28 March 2010

I've cleaned up the lead section a bit, and updated a couple of comments to reflect the improvements in the image circles of recent lenses, and the availability of longer focal lengths that almost certainly aren't used primarily for architecture. I removed reference to “other technical photography” because I have no idea what it means; I have no objection to it's being restored if an adequate explanation can be provided. The first sentence in the second paragraph should probably be supported in any event; I know of many photographers who use 24 mm lenses for landscape work.

I don't especially like the description of shift, but it's by far the most common I found in about a dozen books I examined. I would prefer something to the effect of

“Shift allows changing the line of sight without moving the camera back, allowing the position of the subject in the image to be adjusted without changing the subject's shape”.

But I'm not sure everyone would understand what the “line of sight” is, and this explanation isn't common in sources (Merklinger uses the term, but not in relation to translational movements). Howard Bond speaks of using shifts to point the camera, which I think is actually a pretty good description of what happens, but it strikes me as a bit informal for an encyclopedia. In any event, it's important not to say that perspective is controlled by shifting the lens, because that's not at all what happens—perspective is controlled by the orientation of the camera back relative to the subject. On a view camera, perspective is directly controlled by tilting or swinging the back; with a PC lens, this movement is emulated by reorienting the camera and using the shifts to adjust framing.

I think it may be worth adding a brief section on the convergence of parallel lines: they result from changes in magnification when parts of a subject are at varying distances from the camera, such as when it's pointed up at a tall building. Another consequence of the camera back not being parallel to the subject is that the subject is foreshortened. This important when perspective is corrected in software; two steps are usually needed: one to correct converging lines, and a second to correct the aspect ratio (removing the effects of foreshortening). There's a good Rodenstock article to support this.

The great remaining task is to address tilt/shift lenses, since they now represent almost all available lenses. Offhand, I don't see a good way to adequately cover them in this article without some major changes, and possibly making a mess of things. I'm consequently leaning towards a separate article to avoid some of the problems. As I've mentioned, there would necessarily be considerable duplication, and one significant question would be how to handle the Nikon PC-E lenses. They're tilt/shift lenses in everything but branding, but I'm sure some still use them only for the shift functions, especially the 24 mm. But I'm interested in ideas others might have. JeffConrad (talk) 00:26, 28 March 2010 (UTC)

Nikon centricity and other gripes

This article is terribly Nikon-centric in both terminology and examples. A tilt-shift-swing lens doesn’t and can never ‘correct’ perspective. Perspective comes, by definition, from a point of view, a position in space. Unless the camera moves, the perspective is the same, regardless of focal length or tilting, shifting, or swinging the lens. You can change the focal plane or crop from a different area of the image circle, but that doesn’t change perspective.

And statements like ‘Short-focus perspective-control (PC) lenses’ really don’t help. A macro lens would be short-focus, a wide angle lens has a short focal length. — Preceding unsigned comment added by Hugowolf (talk • contribs) 00:43, 7 November 2013 (UTC)