Does the same scene look different in 4x5 and 8x10?

[Severi Salminen]

This is beginning to look like numerous "how does the 1.6x cropping factor change the perspective on my Canon DSLR" threads.

Well, as many have pointed: it does not.

(Let's forget resolution, and depth of field for now.)

The size of the negative and the focal length _only affects cropping_ - nothing more. No "spatial or telecompression", no change in perspective or anything else. This is exactly what Adams also tries to explain in The Camera. Unfortunately I don't have the book with me now, but Jerald, if you read the text carefully and observe the pictures, you should get it. Íf I recall right, he got two sets of pictures: two where the position is the same and focal length changes and two where the camera position also changes. The first two pictures (same position) are identical, only the cropping is different. The one taken with wider lens could be cropped and printed to look exactly like the one taken with longer lens. This is not possible if camera, and thus perspective, changes.

You can allways make the same, identical, print of a certain subject using a shorter lens, larger negative, and cropping when printing. There will be no difference at all. (If the resolution would be infinite...). So basically all we need is a 1mm lens with infinite resolution ;-)

This is not a rule of thumb, but a rule: perspective changes if, and only if the distance between the camera and the subject changes.

[Jeffrey Goggin]

While the angles of view might be identical when using a 135mm lens on 4x5 and a 270mm lens on 8x10, there's still the matter of depth of field, as Tim pointed out in his original comments.

You'll need to stop down the 270mm lens two additional stops to match the DOF of the 135mm lens, something which may or may not make a perceptible difference when comparing the two images. And once you've done that, there's the difference in diffraction effects to consider, not to mention subtle differences in lens quality (and design), exposure, etc.

In my (very limited) experience, images shot on 8x10 do indeed look different than images shot on 4x5. Whether these differences are significant or not is something that will vary from person to person, depending upon their individual tastes and preferences, and visual acuity.

FWIW, in my case they're significant enough that I'm this >< close to selling all of my medium-format and 4x5 gear, and committing myself exclusively to shooting 8x10 despite the additional costs (in terms of time, money, and effort) that go along with it.

[Gary Meader]

Tim- For my money, I think it's about resolution. It's the same effect that Tech Pan has over standard films. This is what I mean; If you do all the things you mentioned in order to objectify the test, (equivalent lens lengths and all that), if the texture on that far off tree bark has .001mm in which to resolve itself on a 4x5 neg., and I just picked some wild number out of my head, so don't all the mathmeticians jump on me, but 4 times that area in which to do the same resolution on an 8x10, I sure think the film and resulting print looks and 'feels' diferent. I don't think it has anything to do with angle of view or distortion or anything like that, I think it's image capture capability. And I think sometimes we describe 'feel' by things we can't even really 'see' ourselves. I give a talk in my area to libraries and museums about the team of Darius and Tabitha Kinsey , who worked in the woods in my area at the turn of the century. And I've printed from his original 11x14 negs, in contact, and that experience is what makes me think it's all about resolution. They are glorious and I can't for the life of me make a 11x14 image from a 4x5 neg look like an 11x14 neg. Does this make sense?

[tim atherton]

Thanks for all the replies - it is as I suspected - not so much an optical/mathematical difference as an optical (perceptual) illusion (I'm going to stick with optical illusion - it's common parlance - those two wiggly lines - is it a vase or two faces...).

"A 135mm lens will produce the IDENTICAL "angle of view" on a 4X5 sheet as a 270mm will on an 8X10 sheet. If you took the two pics, one with 4X5/135 and the other with 8X10/270, and enlarged the 4X5 pic to 8X10 and layed it next to the 8X10 contact print made with the 270 the scene would be identical. But the 810 might be better quality. :>))"

Jim (and Severi), I am therefore assuming that there is no difference in "compression" of the view - i.e. what you might call telephoto effect - from one to the other?

"I feel that you are correct about the way in which 8x10 renders a scene differently to a 4x5 camera with a lens of equivalent angle of view. I believe that it has to do with the compression of space that is a property of all longer lenses. The background seems to get pulled forward lending the image a greater sense of solidity (for want of a better term)."

this is indeed part of the effect I was talking about - I think it probably comes more from some of the other things people have mentioned, that can perhaps all add together to make the same scene look subtly (and sometimes very) different from 4x5 to 8x10 (and it's something I can see often enough to make photographing with 8x10 worth the extra effort). Among others these would be:

the subtle difference between a 2 or 3 times enlargement from 8x10 and a 4 or 6 times one in 4x5 to get the same size print.

the potential differences between the type and quality of lenses often used between the two formats

the way the scene itself is viewed and chosen by the photographer because of the different relationship to the ground glass - 4x5 small, peering in close etc. Whereas 8x10 is certainly more comfortable and "natural" (and yes, probably more so than 11x14 in a way) - TV/Monitor screen like - under the darkcloth it is very easy to view the whole overall scene from a comfortable distance - when I think about it, I do compose quite differently from 4x5 to 8x10 - in good part because of this.

i think this also ties into the effect someone mentioned of an 8x10 lens just not seeming as wide as a 4x5 lens. On 4x5, 90mm often seems plenty wide for me - oft times too wide. My 75mm lens usually seems far too wide unless I'm in some cramped interior. But on 8x10 my 159mm lens (say an 80mm equivalent) rarely seems too wide and is often not wide enough. And a 210mm lens doesn't feel too wide at all - even though the proportions of the two formats are identical.

add in the obvious things - like often a smaller aperture and longer exposure in 8x10 and for me they often add up to enough of a difference in the final image (and hopefully, as far as I am concerned a difference that is more interesting...) that it's worth photo ring with 8x10 in many circumstances.

My interest in starting this thread (answered I think...) was - is this difference due to a simple mathematical difference between format size and the different lenses used or was it something more subtle, harder to pinpoint and perceptual.

Interestingly, it seems to be the latter - which of course makes it more fun and a greater challenge, because you can't figure it out just by buying a calculator from Rodenstock or The View Camera Store - it's much more down to intuition and fine tuning all these subtle little differences and perceptions to make an interesting image that draws on these differences and ambiguities.

[Severi Salminen]

There is, indeed, no "compression" (the way Matthew, for example, described) involved with different focal lengths or film sizes. It is _all_ about field of view -> cropping. The background will not get "pulled forward" when using larger negative or different focal length. This would mean non-linear light path from subject to lens and it happens only near black holes...

I think one of the biggest differences in "feel" comes when looking at the ground glass. There is a big difference in perceiving the scene when looking at 5" wide ground glass compared to 10" wide - when they cover the exact same field of view. The same effect can be seen when looking final prints either too close or too far away.

[Andy Eads]

I think there are a few factors that generate a genuine difference. First, an 8x10 neg has four times the area of a 4x5. All other factors being equal, the non-image noise will be 1/4th as great (grain, dust spots, processing anomalies, etc. The depth of focus is greater with longer focal length lenses and so may contribute to the image. Likewise, the resolution of the optical system is related to the area of the image formed; the same 4x factor applies. A contact print from a larger negative will not suffer from local contrast degredation caused by an enlarger's optical system. Finally, and this is purely opinion on my part, the way the depth of field falls off on a longer focal length lens meets the expectation of the human vision system better than does a shorter lens.

This discussion has got me thinking I should perform a simple test. I'll make a series of photographs using my Nikon, Rollie, 6x9 back, 4x5, and 8x10 using the same film. I'll make a baseline contact print from the 8x10 and then enlarge the other negs to match the contact exactly for size and tonality. Then I'll round up some people to pass judgement on the results. It would be fun if others on this forum would try the same thing. We could compare results and perhaps learn a few things along the way.

[Martin Miller]

There is a mathematical subtlety to this question that everyone has so far missed. The assertion has repeatedly been made that a 4x5 camera with its normal lens set at exactly the same position as an 8x10 camera with its normal lens will lead to exactly the same perspective, and I would not argue with this. However, the assumption that the two different format cameras will be colocated is incorrect if you want to fill the frame of each with the same subject. Say you want to fill the frame of each camera with a tree. The image magnification for the 8x10 is twice that of the 4x5, because the image is twice as big. The distance between the center of the lens and the tree, u, is related to the magnification, M, by the simple relation, u = (1+M)F/M, where F is the lens focal length. Using this relation one can show that the distance of the 8x10 camera (more precisely the center of the lens) from the tree is exactly a distance of F further from the tree than is the 4x5 camera. For an 8x10, this is about a foot. For objects on a distant horizon, this will obviously not be perceptible, but many picture scenes have significant objects in the foreground, and camera placement a foot further away will give the impression that the scene was shot with a longer focal length lens as compared with the 4x5 image. The 4x5 image, enlarged to 8x10, will not be a perfect overlay with the 8x10.

[Jim Galli]

Martin, You're saying the "nodal point" of both lenses would be in the same place but the 8X10 image would "land" further away? Right?

[Martin Miller]

Jim:

No. It's the nodal point that lies at different distances from the subject for the two camera formats. In the formula u is defined as the distances from the nodal point. If the cameras are placed so that the nodal points coincide, then the perspective will be the same but tree will slightly magnified in the 8x10 compared with the 4x5 image. In that case the 4x5, when enlarged to 8x10 will again not exactly overlay the 8x10, even though the perspective is the same because the field of view of the 8x10 image will be slightly smaller.

[Martin Miller]

CORRECTION: In my first post, I said "Using this relation one can show that the distance of the 8x10 camera (more precisely the center of the lens) from the tree is exactly a distance of F further from the tree than is the 4x5 camera. For an 8x10, this is about a foot." The first sentence is correct but ambiguous. The F referred to here is the focal length of the 4x5 lens NOT the 8x10 lens. The second sentence is correct also, but the implication is that the 8x10 camera is a foot farther away from the tree than the 4x5. This is incorrect. The 8x10 camera is a HALF foot farther from the tree than the 4x5 when the tree fills the frame of both. The perception of the effect will vary with the distribution of subject distances in the scene, but, in principle, one can never take exactly the same FULL-FRAME image with two different formats. Of course, the depth-of-fields of the two normal lenses will differ too as pointed out by Jeffrey. This is true even allowing twice the circle of confusion for the 8x10 because it suffers half the enlargement.