Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-08-2012, 11:44 AM - 1 Like   #46
Veteran Member
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
Some humble uses of compression by changing distance

QuoteOriginally posted by jsherman:

Another thing you can do is use both zooming *and* a different distance/position to increase or decrease compression effects in neat ways. I can show you some of my own results from doing that tomorrow.
As promised...

These would have been more effective as before/after comparisons, but I can't find the 'before' shots, apparently I didn't save them, so I'll just tell what I did:

In this shot, (Tamron 70-200 f/2.8 on K20D,) I was originally shooting that photographer from a lot closer at 70mm. I wanted to get a nice detailed shot of that Mamiya because I thought it was interesting, but although I had some subject isolation, I didn't like the background. The top of the trees came down into the frame and was distracting and not very appealing, and I wanted it to just be a 'wall' of trees behind her. So, I moved back a bunch, re-shot at 200mm (and cropped a little bit,) and the compression effect worked for me to give me that 'wall' behind her, which I liked a lot better than the closer, 70mm, less-compressed shot.





In this shot below, I was shooting these guys with a 85mm f/1.8 on D700, and getting nice isolation, blurred background, etc, but when they started to be pulled parallel to that treeline I wanted to 'bring the trees in' for a more compressed effect, so I changed to a 300mm f/4, moved back, and re-shot (they were actually being pulled through a giant pile of leaves in front of them several times fast, so I had more than one opportunity to shoot )

I was able to soften the hard lines of the trees and lessen the background contrast by shooting wide-open, so the trees were 'present' via compression but not dominating the shot. I think it works well and is much more interesting than the nice-but-more-ordinary less compressed 85mm shots I had been taking.




I wish I had the 'before' shots, but hopefully you can see how there are some compression effects here. My main point is that from the original position, there was no FOV change, focal length change, format change, etc I could do that would give me the same compression - I had to move back.


.


Last edited by jsherman999; 05-08-2012 at 11:53 AM.
05-08-2012, 04:37 PM   #47
Veteran Member
Ikarus's Avatar

Join Date: Apr 2010
Location: California
Photos: Albums
Posts: 471
That second shot is fabulous!
05-08-2012, 05:57 PM   #48
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by jsherman999 Quote
I think it is exactly what we (and the OP) were talking about. If you have another definition for compression, that would explain this disconnect, though.
Indeed. It is unfortunate that there are actually two senses of the term "perspective" that are used in photography, and *both* tend to be misunderstood. You are absolutely correct that perspective - in the sense of ordinary linear perspective - is dependent on distance and distance only. And this does tend to be misunderstood by many, many people, as they erroneously believe that ordinary perspective is determined by focal length. It isn't, and indeed, you don't even need to have a camera, or lens, to talk about perspective in this sense. Distance to subject is what determines the linear perspective, and that's all there is to that.

However, there is an additional phenomenon known as "perspective distortion", that does have to do with linear perspective, but it refers to a *very specific effect* that can only be discussed with respect to 2-D representations of a scene (eg, photography, painting) and angle of view.

QuoteQuote:
"Compression", in the context of this thread (assuming I know in what context the OP was asking,) means "An effect of linear perspective distortion in which objects further away appear to be closer to each other, or compressed, and become more so the further the subject is from the viewer....
This is a very good description of one of the properties of linear perspective - as objects recede, the space between them appear to diminish. Yes, that much is indeed dependent only on distance to subject. But the term "telephoto compression effect" as generally used in photography actually refers to something *above and beyond that*, as I explained in my previous post.

If the AOV depicted within the image matches the AOV subtended by the image itself, our eyes and brain can correctly interpret this apparent reduction in distance. The space between the objects *is* diminished because of the distance, but we can look at the image and correctly deduce the actual distance to to subject and thus the actual space between objects. That is, we are not *fooled* by the appearance of reduced space into thinking the objects are closer together than they are - we see right through the effect.

However, if the AOV of the image is narrower than that subtended by the image itself, not only do we see the reduced space in the image, but we are also *fooled* into thinking that there was less space between the objects in real life as well.

QuoteQuote:
the exact same level and rate of object compression happens in every shot you take, regardless of focal length/FOV, if you take it from the same position. The telephoto lens (or cropping) just makes it more apparent, but it doesn't create anything that doesn't already exist in the image, were it taken with a wider-angle lens.
Again, you are describing the ordinary effects of linear perspective, and you are quite correct. There is nothing different *about the images themselves*. However, the effect I refer to is not just dependent on what it is in the image, but also, how you *view* that image. Again, it's all about AOV. When the AOV depicted in the scene the scene matches that subtended by the image itself. we *perceive* it as "normal". When the angles don't match, we *perceive* a distortion. Nothing has actually changed regarding the linear perspective - I'm describing an optical *illusion* here. Even with a narrow AOV, the linear perspective is correct for a scene shot from that position - it's just that the narrower AOV fools us into thinking we are closer than the photographer really was, and the linear perspective actually in the image is not correct for an image shot from where we *think* the image was shot from.

Your examples correctly show that the *actual linear perspective* is dependent on shooting position only, but again, the *optical illusion* of "telephoto compression" (or perspective distortion in general) doesn't care about shooting position - it cares about AOV.
05-08-2012, 06:38 PM   #49
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteQuote:
So would it be safe to say that if you took a picture of a tomato with a pineapple behind it at a distance of 1 metre, on the top of your dining table or something, no matter what focal length or FoV you use, as long as you are 1 metre away from the tomato, the relative size of the tomato and the pineapple will be exactly the same?
Yes, of course. Doesn't even matter if you take a picture at all. From any given position, the relative sizes of the tomato and pineapple are fixed. I don't care if you own a camera or not, if you shoot a picture or not, if you take a picture with film or digital, with a wide angle or telephoto lens, if you crop the image or not, if you use a UV filter on your lens or not, if you use a tripod or not, if you turn SR on or off, if you shoot RAW or JPEG, if you stand on one leg while pressing the shutter etc. The nothing you do while in that position is going to change the relative size of that tomato and pineapple.

This is the very real and very important sense in which *linear perspective* is dependent on position (ie, distance to subject) only. Many people fail to comprehend this, and it is absolutely correct and appropriate to remind people from time time - linear perspective depends on position and position only.

However, the optical illusion referred to as "perspective distortion" (and "telephoto compression" is one example of this) is something different, and it actually does not depend on position, but on the the AOV depicted in the image and the size of print and distance from which you view it (which together determine the angle subtended by the image itself).

If you shoot with a "normal" lens and make a typical print and view from a typical distance, these angles will match. If that occurs, then a number of very important things follow as well, and the most important is this: the *absolute* size of the objects in the image will be exactly the same as they appeared in real life from the shooting position. That is, the tomato in the picture will appear the exact same size that the tomato in real life did, and the same is true of the pineapple. Not only will the *relative* sizes be correct, but so will their *absolute* sizes. If you shift your gaze from the actual tomato to the print, it will look perfect. However much larger or smaller the pineapple appears will make sense to your eyes and brain, because it matches real life.

However, if you shoot with a telephoto lens, the angle depicted in the image will be *narrower* than the angle subtended by the image itself, and the effect of this is that both objects will appear *larger* in the image than they did in real life. Obviously; that's the whole point of telephoto lenses. Relatively speaking, they will be in the correct proportions, but both will appear larger (by the same amount). If you compare the image to the scene, you will see that this is the case.

The optical illusion comes in if you show the image to someone who cannot also see the scene, and doesn't know what kind of lens was used. Absent information to the contrary, they will assume the lens is "normal" - that the scene depicted matches real life in the same way as in our first example. Since the tomato and pineapple appear larger than they really did from the original shooting distance, the viewer will mistakenly assume the photographer was *closer* to that tomato than he really was. However, in real life, if you get closer to the tomato, it would get larger *faster* than the pineapple, and indeed, from close enough, it would appear *larger* than the pineapple. So when this viewer sees a tomato that is apparently shot from close range, with a pineapple behind it that is nonetheless still much larger than the tomato, the viewer will mistakenly assume the pineapple is *closer* to the tomato than it actually was, because that would explain its size relative to the tomato.

This is the perspective distortion known as "telephoto compression", and you can demonstrate it from any distance you like, although it does help to be relatively far away or else a telephoto lens might show only a small slice of the fruit and thus make it difficult to evaluate the effect. It heklps if both fruit fit in the frame.

05-08-2012, 06:58 PM   #50
Veteran Member
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
QuoteOriginally posted by Ikarus Quote
That second shot is fabulous!
Thanks!

.
05-08-2012, 07:55 PM   #51
Veteran Member
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
QuoteOriginally posted by Marc Sabatella Quote

Your examples correctly show that the *actual linear perspective* is dependent on shooting position only, but again, the *optical illusion* of "telephoto compression" (or perspective distortion in general) doesn't care about shooting position - it cares about AOV.
I think you explained this well enough in your first post in this thread, which I liked - subsequent posts seem to have suffered from disassociation between the optical illusion you were trying to describe and what the photographer could do with linear perspective to manipulate that optical illusion. I'm certain your understanding of that was fine, but I didn't think your descriptions and emphasis were.

As the subtended angle decreases, one effect is that the optical illusion of object spacing compression increases (objects appear to get closer together.) What can one do to decrease that subtended angle? Move backwards, away from the subject

(BTW, the entirety of what you are describing was diagrammed in a very clever stick-drawing cartoon I saw a few years ago, complete with angles being pulled out of the scene and depicted on a printed image on a wall, with the angles no longer matching the expected subtended angle, which fooled the stick figure guy, etc. I tried to find it via google but no luck, I don't know the creator or where I saw it. It would be very good to have for discussions like this.)


.

Last edited by jsherman999; 05-08-2012 at 08:10 PM.
05-10-2012, 12:13 AM   #52
Veteran Member
Anvh's Avatar

Join Date: Sep 2011
Posts: 4,616
Marc when i wrote this you seem to disagree and now you seem to be saying the same thing i did so what is it? https://www.pentaxforums.com/forums/pentax-beginners-corner-q/178549-focal-le...ml#post1876265
05-10-2012, 02:13 AM   #53
Forum Member




Join Date: Sep 2010
Location: Brisbane
Posts: 57
Original Poster
Thanks everyone for replying. I have found this very educational and am quite surprised about the complexities behind all this... I thought it was going to be a simple yes or no

Thanks again and definitely soo much more I have to learn.

Cheers

05-10-2012, 11:35 AM - 1 Like   #54
Pentaxian
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,686
QuoteOriginally posted by Anvh Quote
Marc when i wrote this you seem to disagree and now you seem to be saying the same thing i did so what is it? https://www.pentaxforums.com/forums/pentax-beginners-corner-q/178549-focal-le...ml#post1876265
Very good question!

This is tricky, because your post contained lots of true statements. In fact, it was almost all true statements. Perspective - in the sense of ordinary linear perspective - really is dependent on position only. And yes, AOV tends to influence position, which is how a lot of people get confused into thinking AOV (or focal length) somehow affect ordinary linear perspective.

I was not disagreeing with any of the statements you made in that post except the one in the first paragraph: "it won't change my view on this". Because "this" - the thing we were discussing - was *not* "ordinary linear perspective. What I was dicussing was the *very special optical illusion* known as "perspective distortion", which as I have explained is not dependent on shooting position but on angle of view.

You had been essentially denying this second effect existed, saying that the only relevant effect was ordinary linear perspective. That is just not true. There are two *different* subject being discussed here. One is ordinary linear perspective, which works as we all agree it works. It is dependent on position only, and is indeed not an illusion at all - it can be captured on film, measured with a ruler and protractor, etc. The other subject is the special phenomenon - *an optical illusion* - known as perspective distortion. The existence of this illusion is dependent on AOV only.

These two sinjects are of course not unrelated. Perspecitve distortion comes about when the viewer misjudges the distance to the subject (because of the discpreancy in AOV), and this misjudgement of distance causes him to misinterpret the linear perspective in the image as telling him something other than what it actually is. So yes, the perspective distortion happens *because* of the fact that linear perspective depends on position. But it is the discrepancy between the actual distance to subject and the distance we mistakenly perceive that causes the illusion, and that discrepancy is caused not by the distance itself, but by the AOV.

Needless to say, there are some *extremely* subtle distinctions being made here. Please believe me when I say that my comments about the language gap were not meant in the sense of "you're a non-native English speaker, so you're incapable of understanding the concepts being discussed". I meant only, the language gap may partially explain why my own attempts to explain these subtle distinctions was not successful. Because the distinctions being made are so subtle, it's hard enough explaining them successfully to native speakers who would presumably have the best chance of understanding the nuances of each word I am carefully choosing to express these distinctions. Especially since my explanations are not perfect. So it just shouldn't surprise anyone if my attempts to explain these concepts don't work as well with non-native speakers. If I were better at explaining, maybe that problem wouldn't exist. And hopefully, I *am* getting better at explaining.

Getting off topic (so what's new), but that is one reason I tend to have so much patience for these long involved discussions. Some would call them arguments, and I suppose sometimes, that's all they are. But in many cases - like this one - is just an educational process as far as I am concerned. I am, both by nature and by profession, a teacher. It is my inclination and often my job to keep explaining something until the light bulb clicks on. Each time I encounter a new obstacle that causes one of my explanations to not work, I learn something about how to explain differently the next time.
05-11-2012, 06:20 PM   #55
Site Supporter




Join Date: Dec 2006
Location: Gabriola Island
Posts: 591
QuoteOriginally posted by tyronfall Quote
Thanks everyone for replying. I have found this very educational and am quite surprised about the complexities behind all this... I thought it was going to be a simple yes or no

Thanks again and definitely soo much more I have to learn.

Cheers
One simple answer to your original question is that you have the option of looking throught the viewfinders of different format cameras and seeing for yourself.

I really don't see much point in agonizing over theories about perspective and compression when it is so easy to observe things in the real world.

The main thing you need to learn is to be concious of what you see through the viewfinder. Optical theories be damned!

John
05-13-2012, 09:16 AM   #56
Veteran Member




Join Date: Jan 2012
Location: Somewhere, USA
Photos: Gallery
Posts: 453
This whole discussion blew my mind. Compression is a result of distance, not optics. Mind blowing. I never thought of that and it explains why I cannot get the same perspectives by moving forward with a wider lens. Coming off a superzoom, I've really missed being able to dial in a composition. The only zoom I use regularly is my 60-300, and its sharp enough to do landscape and buildings up to 200mm....anything higher and its single objects in the center, which is fine for a lot of birding and nature shots. The macro comes in handy too....anyways I'm getting away from the discussion. I typically leave a 28mm on the camera all the time. I've learned. To love the classic look of 40mm, and it has forced me to be far more creative with composition and framing. Same goes for the 50mm I use for up close portraits and street shooting where I want to be a bit far from my subjects. At f2 there is plenty of light at night to shoot handheld easily. Win. The subject isolation at f2 is rather nice too. As far as portraits and flattening....aps equivalent will give you the same flattening as a full frame focal length. This much is fact. My superzoom was roughly 8mm (if I remember correctly) and was the. Equivalent of 27mm on a full frame. Distortion due to the lens was pretty similar. The lens and sensor are just smaller. Aps does have the. Added advantage of taking images from the sweet spot in ff lenses. So those unsharp corners and barrel distortion are cropped out or. Cut down. Every lens is a compromise. I find that older glass from the 60s and 70s is really sharp, especially in the center where aps takes it from. I love WA photography and really need to build my kit below 28mm. A 35mm would be nice. Too, but really I think. Id prefer the 28. I don't think you can have too many. I still use the kit lens every once in a while for wa....thinking about upgrading that and getting a 55-300 next. Id love a da*, but jesus it just got expensive.

Thanks everyone. I've struggled with understanding this for a long time, and always felt like aps was missing something. In reality you lose dof the most and that's about it. With fast lenses I can still get plenty dof control. I really must admit I love the crop factor working in favor for long zooms. 450mm on a 300mm is nice. Sorry for all the typos. Did this on my phone and they need to fix the mobile input badly. I can barely see what I'm typing because it scrolls off the screen. The box is too wide! @adam you hear that?
05-14-2012, 08:17 AM   #57
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 27,424
QuoteOriginally posted by zosxavius Quote
This whole discussion blew my mind. Compression is a result of distance, not optics. Mind blowing...
Indeed! I remember the first time I did the simple "crop" experiment with my favorite image editor. The realization that the position of the lens determines perspective and that focal length allows us to control FOV and framing changed the way that I approach the subject.


Steve
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
compression, effects, k-mount, pentax lens, slr lens
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Full frame/APS-C conversion chart for dummies jonhock Pentax SLR Lens Discussion 31 07-03-2014 08:08 AM
APS-C vs Full frame background blurriness on same lens same subject Enkhbayar Photographic Technique 30 02-07-2012 03:03 AM
Is a full frame lens on an aps-c, a negative? outsider Pentax SLR Lens Discussion 36 03-30-2011 09:34 AM
Lens hoods on full frame, APS-C... nater Pentax Camera and Field Accessories 4 08-04-2010 08:50 AM
Full frame Lens vs DA lens(APS) bobmaxja Pentax SLR Lens Discussion 8 04-17-2010 12:03 PM



All times are GMT -7. The time now is 05:29 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top