Help with lenses please.

[madlights]

Yeah I think we all are standing the same distance from the subject just seeing it with a differing field of view. Kind of ironic I'd think

[sean_mcr]

:D Yeah I think we all are standing the same distance from the subject just seeing it with a differing field of view. Kind of ironic I'd think

All part of the fun huh;)

I don't actually mind some distortion if it's the right shot. It's all about the intent really

I think this explains it better then i did

http://cornicello.blogspot.com/2007/05/ ... ve_08.html


Cheers fellas

[madlights]

Firstly my mind wasn't too "in focus" when this thread began...since it's been a long time since I've tried to describe this visual stuff with verbal terms. If a viewer views a photo, say a print, and has the same field of view, then the photo's appearances are exactly the same as the cameras. If a person put their nose right up to the print (or screen) on a wide angle shot they would come close to approximating the lines of perspective that the camera sees...since our peripheral vision is not the same as a lens. But technically it IS correct that one lens in no different than another in regarding perspective. How can it be? the laws of perspective are just that...but our perception of perspective does change...since our eyes and brains don't behave exactly like cameras. I think that's where the confusion between us in this thread lies...between the "apparent to the viewer" and the technically correct. It is correct that a crop makes no difference in perspective...BUT...to get the exact same shot with a different crop sensor or a different field of view lens , say a head and shoulder portrait, the photographer must MOVE which changes the position of the camera...and not the perspective of the lens, but the actual perspective. If we think a telephoto changes perspective, no...distance changes foreshortening...but often times a tele just shows us a small circle at distance...since it is only a crop of the scene. I think the confusion has been between the apparent to the viewer and the technically correct...and between the 'real' properties of lenses and the 'real world' properties of them. Yes I do believe the effect of a 1.6 or 1.5 crop camera does change the "real world" properties of a lens. It requires a photographer to place themselves in a different position to get exactly the same framing of a shot as they would on a full frame 35mm. Much the same as a lens of a given mm would require a different position (and perspective) to get the same field of view on a 35mm camera as opposed to a medium format. My first answer in this thread was in less than correct terms..confusing perceived with actual...and could have been explained much better by myself...The one variable I'm still unsure of is the variance in the field of view between different lenses at a given mm setting (discussed in another thread ), so that could influence camera to subject distance to obtain the same framing... So.....I do think we're all on the same planet here.

[marxz]

.... and to sum it up, I guess, using a 50mm prime on a 1.6 crop means that to frame a shot the same way as on a full frame sensor DSLR/35mm film SLR with an 80mm lens you will have to stand the same distance away from the subject for both camera/lens combos.

As such you get the same effect on foreshortening in your image for both setups

[madlights]

.... and to sum it up, I guess, using a 50mm prime on a 1.6 crop means that to frame a shot the same way as on a full frame sensor DSLR/35mm film SLR with an 80mm lens you will have to stand the same distance away from the subject for both camera/lens combos.

As such you get the same effect on foreshortening in your image for both setups

LOL..sort of. We tried this on another thread...well it was with differing zoom lenses set at exactly the same zoom setting, say 70mm. Each of the lenses had a different field of view...at the same zoom! I figured it would be a little bit maybe, but it was quite a bit. This is all a very complicated issue LOL...and I don't really know how much it all means...since if you like the shot...snap it! The perspective is of course the same at the same distance...but the other perspectives introduced from a wider field of view (since more of the scene is seen) would have a bearing on perception and where you'd have to stand to frame exactly the same shot. If the field of view of each lens was exactly the same yes...but I think every lens is different unless focused to infinity. My 70-300 has a much tighter field of view at 70mm than my 24-85 at 70mm when not focused to infinity. So if you see what I'm getting at...you might have to stand at a different place...and you might not...depending on the lens optics at the focal length. Focused to infinity they should be the same from my understanding which might be limited.. dang remembered this about lenses (he's pretty sharp about such stuff) and was very correct. Here is the thread http://forum.pbase.com/viewtopic.php?t=30349

[marxz]

.... and to sum it up, I guess, using a 50mm prime on a 1.6 crop means that to frame a shot the same way as on a full frame sensor DSLR/35mm film SLR with an 80mm lens you will have to stand the same distance away from the subject for both camera/lens combos.

As such you get the same effect on foreshortening in your image for both setups

LOL..sort of. We tried this on another thread...well it was with differing zoom lenses set at exactly the same zoom setting, say 70mm. Each of the lenses had a different field of view...at the same zoom! I figured it would be a little bit maybe, but it was quite a bit. This is all a very complicated issue LOL...and I don't really know how much it all means...since if you like the shot...snap it! The perspective is of course the same at the same distance...but the other perspectives introduced from a wider field of view (since more of the scene is seen) would have a bearing on perception and where you'd have to stand to frame exactly the same shot. If the field of view of each lens was exactly the same yes...but I think every lens is different unless focused to infinity. My 70-300 has a much tighter field of view at 70mm than my 24-85 at 70mm when not focused to infinity. So if you see what I'm getting at...you might have to stand at a different place...and you might not...depending on the lens optics at the focal length. Focused to infinity they should be the same from my understanding which might be limited.. dang remembered this about lenses (he's pretty sharp about such stuff) and was very correct. Here is the thread http://forum.pbase.com/viewtopic.php?t=30349

OK I did actually have several long paragraphs in my post that I removed before posting to simplify things but first an aside comment-

The mathematics of foreshortening were established by Filippo Brunelleschiare in the early 1400's and refined by later artists such as di Vinci (who noticed and included the perceptual part) and are well know by almost all artists, architects and draftsmen to this day (even if they may not know the name or the actual formula's they understand and apply the technique) . Using this method artists and draftsmen have been able to produce the same perspective effects photographers use today, even to the point of being able to manipulate certain numbers and ratio's to emulate what photographers do to exploit the effect of barrel and pincushioning distortion (in lens or in post processing) to enhance or flatten _perceived_ perspective or manipulated to such an extent they create Echer like brain teasers.



In the case raised by the original post's question primes like the Canon 50mm primes (1.2 and 1.8 ) and Nikkor's equivalents actually have very little perceivable distortion so will not be unduly influenced by our perception of perspective so would, when mounted to a 1.6 crop camera exhibit relatively similar perspective flattening effect as a 80mm lens on a full frame camera when framed the same (as they would be taken, effectively, at the same distance from the subject)


- so here are the omitted paragraphs (with some addendum's and edits):

(insert at the end of my original post) ....this is all, of course, discounting 2 factors that will present them self's in many "real world" comparison between lens.



the first factor is distortion (pincushion and barrel distortion) that have the effect of "squeezing" more (in the case of barrel ) or less (pincushion) of the peripheral (non central) view relative to a rectilinear (non distorting) lens of the same mathematically calculated true "Effective Focal Length".

This occurs becasue the magnification factor varies across a distorted lens in a certain ratio (varies from lens design to lens design) as it radiates out from the center to the edges.
In the case of barrel distortion it causes the center part of the image to appear larger/closer in proportion to the edges and gives the perception of increased foreshortening.

This sort of distortion is most noticeable in non rectilinear wide angle lenses (and at its most extreme in fish eye lenses)

So a non rectilinear wide angle lens will, as they almost always display barrel distortion rather than pincushioning, actually "pull in" more peripheral sceanery but at the expense of distorting the image (straight lines that do not pass directly through the center-point will appear as if they are bowed away from the center-point).
the peripheral image material will also smaller relative to the central image. This means a lens with significant barrel distortion will actually have a wider "FOV" than a non distorting lens of the same "true" Effective Focal Length.

When taking a traditional portrait almost inevitably you center the face in the view finder.

If you take this portrait with a lens that has significant barrel distortion the effect is that the central part of the lens magnifies the center part of the image (oh noz thatz da nose) to a greater degree than the ears that are about 1/2 to 3/4 of the way out from the center causing them to appear to be smaller, combined with the "real" perspective effect caused by you needing to be closer to the subject with a wide angle lens this greatly increases the perceived perspective.

If you were to take the same portrait from the same distance with the same lens but shift the subject off to one side so that the edge of the frame passes through the face's line of symmetry then the nose would be "shrunk" by the lesser magnification at the edge and the ear (just the one this time ... unless the subject is a nuclear waste technician) in the center of the frame would appear to be enlarged by the centre's relative to an image taken with a rectilinear lens

there are other factors at work here with wide angle lenses in that it is more difficult to achieve a rectilinear lens design as you get wider - to the point that it is, apparently, actually impossible to get a perfectly rectilinear 180 degree FOV represented on a flat surface unless you have an infinitely wide/tall projection surface... My understanding of it is that this is because as the FOV angle increases you need to increase the size of the focal plane, or the projected image or the print so that it occupies a roughly similar FOV to our eye - this is why IMAX theatre screens are curved the way they are, and your need to sit near the center as the curvature of the screen presents a distorted image (sitting too far to the left would result in exaggerated enlargement of the left side image making it perspectively seem closer than it should and expanded away from the rest of the image and a non linear distortion would occur that would result in the center image and the far right image to appear overly compress perspectively

SO in the case of photographic images on a print or screen the perception of perspective that we have is filtered by our brain based on visual clues calculated from how much of the FOV the image actually occupies when we view it and so to look at a image that is captured at, say 90 degrees FOV then printed to an image that may only occupy 10 degrees of our eye's FOV triggers "false" perspective cues in our perception of the image.

The second factor which relates directly to the above factor is the mislabelling of lenses,

back when I studied photography in school our teacher gave us a comparison between several so called 50mm lenses made for the Pentax K mount camera's that we used.

using a grid (ok it was a 4 story tall brick wall of our gym) and a tripod at a set known distance and some simple maths (that eludes me now) he showed that between the half a dozen different Pentax and 3rd party lenses labelled as 50mm that they actually ranged from (by memory) 43mm to 55mm in terms of real Effective Focal Length - that is, all other things being the equivalent in terms of FOV captured to a "crop factor" of almost 1.3 (so the difference between a EOS 1Ds and an EOD 1D) - so quite significant.

a couple of reasons were given for this 1st that there is some variance in that some companies marketed lenses that had a mathematically calculated "Effective Focal Length" other than 50mm but presented a FOV roughly equivalent to a non distorted true "Effective Focal length lens of 50mm.
eg. on lenses that had an true mathematical Effective Focal Length of more than 50mm they had enough barrel distortion to cause it to have a greater FOV than a non distorted lens of the same Effective Focal Length and that equivalent FOV was the same as a non distorted 50mm true Effective Focal Length lens and one had pincushion distortion so had an Effective Focal Length of less than 50mm but a FOV equivalent to a 50mm non distorted lens.
The other part of the whole discrepancy was marketing - one of the lenses was ACTUALLY a true 45mm Effective Focal Length and was the lens displaying the LEAST distortion so was displaying the FOV of a 45mm lens but because in the 80's people wanted 50mm prime lenses their lens, based on a 45mm lens from the 60's, was simply re-badged as a "50mm" lens knowing most people wouldn't notice the difference.


as for zooms... I don't trust the marketed so called "focal length" of them, one of my old 35-70 zooms actually had a significant overlapped with my old 75 to 300 zoom I would guess that was because the 35- 70 went in to pincushion distortion at it's long end (giving a FOV equivalent to a true Effective Focal Length greater than non distorting 70mm lens) and the 75 to 300 was displaying significantly noticeable barrel distortion at its wide end (giving a FOV wider than a true "Effective Focal Length" 75mm non distorting lens)... either that and/or one or both lenses were misrepresenting their true "Effective Focal Length".




Some useful links (OK they are mostly Wiki but didn't have time to dig deeper)

Mathematics of Foreshortening:http://en.wikipedia.org/wiki/Perspective_(graphical)

Focal length:http://en.wikipedia.org/wiki/Focal_length

Pincushion/barrel distortion: http://en.wikipedia.org/wiki/Image_distortion




Now I bet you're glad I (initially) over simplified my post?

[madlights]

Hi marxz
Yeah I guess I was confusing the issue there. If all things are technically equal...a 50mm on a 1.6 would require the same distance to get the same crop as an 80mm on a Full Frame. All a crop sensor does is crop the image...if all other things are the same. I'm beginning to see where this side of the discussion is coming from. Some of the other issues have been muddied however. Such as how a wide angle exposes us to differing perspectives etc...(for practical purposes unless we put our nose right on a print) but that's not what you were originally discussing (although explained well with the IMAX example). Then all that ANY optically correct lens does is crop the view...and correct me if I'm wrong. Your explanation may have been long but I think pretty clear...without reading the links even. I was getting terminology confused somewhat...and really the crop factor for all practical purposes does change the properties of a lens labeled for 35mm?...so in that I was plain incorrect when I think of it. However in my last post I wasn't trying to defend my original error...just point out some inconsistencies with lenses. And when I think of it most digicams have lenses labeled with the focal lengths relative to them...IE: 6.7mm to 35mm zoom etc. (much less numerically than the 35mm equivalents) I think you've done a very nice job at analyzing this whole thread.

[marxz]

Hi marxz
Yeah I guess I was confusing the issue there. If all things are technically equal...a 50mm on a 1.6 would require the same distance to get the same crop as an 80mm on a Full Frame. All a crop sensor does is crop the image...if all other things are the same. I'm beginning to see where this side of the discussion is coming from. Some of the other issues have been muddied however. Such as how a wide angle exposes us to differing perspectives etc...(for practical purposes unless we put our nose right on a print) but that's not what you were originally discussing (although explained well with the IMAX example). Then all that ANY optically correct lens does is crop the view...and correct me if I'm wrong. Your explanation may have been long but I think pretty clear...without reading the links even. I was getting terminology confused somewhat...and really the crop factor for all practical purposes does change the properties of a lens labeled for 35mm?...so in that I was plain incorrect when I think of it. However in my last post I wasn't trying to defend my original error...just point out some inconsistencies with lenses. And when I think of it most digicams have lenses labeled with the focal lengths relative to them...IE: 6.7mm to 35mm zoom etc. (much less numerically than the 35mm equivalents) I think you've done a very nice job at analyzing this whole thread.

LOL that and I'm an annoying pedantic at times

I don't confess to being a "expert" but, unfortunately, we were drilled by our junior high photography teacher with all this sort of stuff which may have been good if we were studying a university degree but for 14 year old high school students it just left us with a bad taste of "well, this is boooooooring"

Though it must have been effective as I can still remember, 30 years later and 25 years after giving up photography, all that theory stuff and lots of dark room stuff like how to do Gaussian Glow and an unsharp mask the "old school" analogue way.

[madlights]

Hi marxz
Yeah when I originally said "properties" I should have said "perspective" at the start of all of this...it would have been so much easier...LOL. The one thing I've found really kind of interesting...and that I'm not really certain about now is that previous thread I mentioned...as the elements in a lens go back and forth in focusing do they change the field of view between different primes in a differing way? Depending on lens design? I can see that could happen in zooms much moreso...would think? Not making a statement but asking an opinion...since I don't know...? Of course that wouldn't have any more bearing on the difference between a 50 on a 1.6 and an 80 on a 35mm, than it would probably between various models of any prime at a given mm. It just seems an interesting concept that I've started to wonder about...with that thread about lenses only being the same FOV when focused to infinity? Thanks

[marxz]

Hi marxz
<snip>...and that I'm not really certain about now is that previous thread I mentioned...as the elements in a lens go back and forth in focusing do they change the field of view between different primes in a differing way? Depending on lens design? <snip> t...with that thread about lenses only being the same FOV when focused to infinity? Thanks

sounds feasible given that you control magnification (viz FOV) with macro shots by using the focus - I'm no optics expert but the "experiment" that our photography teacher all the images were shot at a fixed length with the same aperture and focused to infinity