Print size / viewing distance?

[Didier Lamy]

<p>I have already read something about the optimal minimal distance for looking at a photo print in relation to its size, but I have forgotten to remember it..<br />I guess there is no upper limit..<br />Regarding my prints, I am a frequent user of Tmax films<br>

Thanks for your advice</p>

[Bill C]

Hi, it is frequently said that people tend to view prints from a distance roughly equal to the diagonal of the print. I've never seen a formal

justification for this, though.

 

I made it a point to pay attention to this sort of thing - how far away people view from - about 15 years ago and haven't found that rule very

accurate. But I don't know of any better way to predict, at least a general rule.

 

As an example, when prints are small, say 4x6 inches, many seem to view from perhaps a foot or more, although a few may inspect from

close up - perhaps 5 or 6 inches. In an exhibit of larger prints, they seem to follow the diagonal rule more closely, although many do seem

to also step in for closer inspection.

 

Regarding a "correct viewing distance," this also depends on the focal length and degree of enlargement of the camera image. But these

ideas are not well accepted on photonet, so I don't want to belabor the issue.

[Alan Marcus]

<p>Photographic images can be viewed from most any distance. However when it comes to perspective, we can calculate a distance that will replicate the human perspective. Before I explain, you need to know that viewing a picture from a “correct” viewing distance is not critical. For most images, a variation of 50% or more is unimportant. However, portraiture is a different story. We see facial distortion when the lens used is too short or long as to focal length. <br>

<br>

The human perspective: We can gaze out a window at a vista, and with wax pencil, trace the outlines of the objects we see. Such a drawing presents the human perspective. Can we duplicate this perspective with the camera? Yes, if we replace the position of the human eye with a camera. We expose this scene and make a print or slide, computer image the same size at the film frame or imaging chip (contact print). We now view this image from a distance equal to the focal length of the taking lens. Voila! The image we see replicates the human perspective. <br /> <br /> <br>

Now such a viewing distance is impractical or impossible with a modern camera. In the first place, the image is too small and in the second place we likely can’t focus our eyes at this short distance. However, we routinely enlarge the image for display. Say you are using a full frame 24 by 36mm (Fx) camera. We enlarge the image on our computer or to make a print. Typically this will be about 10x (240mm by 360mm or 10 by 14 inches). To find the correct viewing distance we multiply focal length that took the picture by the magnification. If the lens was a 50mm then the viewing distance will be 50 x 10 = 500mm (20 inches). Now most people gravitate to a viewing distance about equal to the diagonal measure of the image. For the 10 x 14 this will be 17 inches (430mm). <br>

<br>

Now for portraiture: Most images will be 8x10 inches placed on a mantel and viewed from about 1 yard (1 meter). The magnification to from full frame to 8x10 is about 9.5x. If viewed from 1 meter (1000mm) the focal length of choice to deliver the human perspective is 1000 ÷ 9.5 = 105. In other words a 105mm is the lens of choice for these conditions. <br>

<br>

Also, you need to know that most images are OK even if the “correct” viewing distance is rejected. Art is art and has no rules, you are free to follow your heart. I call all of this math gobbledygook – anyway. </p>

[Bill C]

Thanks Alan, your example of "wax pencil tracing on a window" is as elegant an explanation of perspective as I've seen, and with no

math either. If one thinks about it a bit, they can probably see that all the details about print viewing are simply to duplicate the same

viewing angle - the secret to "realism" in photos.

 

On another note, your name seems familiar; did you once co-chair an SPSE symposium session? I've wondered for some time, but didn't want to ask and risk losing my "amateur status" here (I've since decided to give it up).

[Alan Marcus]

<p>Hi Bill C<br>

I can't recall being a co-chair but I did many presentation over my career - thanks for asking.</p>

[Alan Marcus]

<p>Hi again Bill C.<br>

The idea of the window and the wax pencil is elegant. The credit goes to C.B. Neblette of Kodak and RIT. His book "Photographic Lenses" covers perspective and focal length and viewing distance. Paraphrasing -- if a photograph is viewed from the correct viewing distance, focal length multiplied by magnifications, there will be no distortion. A 35mm slide made with a 50mm lens, projected with a 100mm, the proper viewing distance is 1/2 the distance projector to screen. The 35mm slide viewed with the aid of a magnifier with a focal length about the same as the taking lens delivers correct perspective. Remember those hand-held 35mm slide viewer. Hats off to C.B. Neblette </p>

[Bill C]

Hi Alan, C. B. Neblette was certainly a giant in the field, but I first saw the "drawing on glass" technique in a book, "Freehand Drawing, Self-

Taught," by Arthur Guptill, published in 1933. So I don't know if Neblette predates him. I suspect it was a well-established teaching tool, but I

don't know. It's only in fairly recent times that I realized how well it demonstrates the photographic situation.

 

I first had some realization of what proper perspective in a photo could do at the age of 10 or12. I was bored silly in some class (history or

geography?), and started looking at some of the photos in the book with a magnifier. Certain ones had this incredible sense of being in the

scene, and I went through the whole book looking for them; they were just fun to look at. It was many years later before I realized what was

happening - those particular photos had to be viewed from much closer for the perspective to be right. I've occasionally described this to

other photographers over the years, but you practically have to force them to view prints from different distances before they become

believers. I don't know why the strong resistance to the idea, but I think it may just be a form of arrogance, "I already know all about that!"

 

BTW, I have to rate presenting SPSE papers as much more prestigious than chairing a session.

[Didier Lamy]

<p>Thank you everybody for your answers, I am much impressed by their expertise (I was ready to get just a one sentence answer by a compassionate pro..).<br /> So by default the distance should be a function of the diagonal of the print, and then one should move depending on the photo itself. <br /> Applying Allan's rule of D = F x magnification, I get for my A4 prints a magnification of ~8x. With a 50mm lens photo, that makes a distance of 40cm, which, fortunately, is a bit less than the length of my arms. The diagonal is then ~35cm.<br /> My concern with the viewing distance comes from occasional landscape Tmax400 prints with more grain than usual in the clouds. And contrary to people who were born with digital photography, I see nothing inspiring in granulates. I am trying to keep the problem under control by playing with the usual parameters, but at the end I have found that the best remedy is to increase the viewing distance, which of course is not fully satisfying.. Using a slower film (Tmax100) makes things better, but then details get easily lost in the shadows, a problem that can sometimes be solved by getting closer to the print..<br /> Thank you again, I will write the Marcus / Neblette rule on a mental post-it.</p>

[bob_bill]

<p> I like Dan Dano Steinhart's description of viewing distance relating to pixel peepers. He says for most artists, the viewing distance is approximately the length of the diagonal of the painting. However, for some photographers, the viewing distance is the length of one's nose. </p>

[Alan Marcus]

<p>You observation is correct for those of us that used 35mm slide film. Our cameras lash-up was likely a 50mm lens and we often used a hand-held viewer. These handy viewers sported a 50mm magnifying lens. The result was, with this device the perspective was always correct regardless of camera to subject distance.</p>

[art_thomas1]

<p>Hmmmmm....<br>

I must do everything wrong. I hold most prints 2x3" to 12x18" with my forearms about 45 degrees to perpendicular. My 20x30" tend to be less than an arm's length away.<br>

The diagonal of a 2x3' would be about 2.5" which is too close for me. Is there really a "correct" distance? If so, who determined it?</p>

[Bill C]

[Art] "Is there really a "correct" distance? If so, who determined it?"

 

If you want the perspective to appear "realistic," assuming that your photo has depth cues in it, you want the same viewing

angle for the print as was had in the original scene. If you carefully consider the mechanics of the "drawing on glass"

technique, I think the concept will be clear.

 

I don't know when this was first realized, but it was certainly long before my time. Older photo books, from perhaps the

1940s used to describe it in terms of focal length times print magnification.

 

I once referred to Rudolf Kingslake's "Optics in Photography," like so:

 

"By far the most important rule for correct perspective in photography is that the final print must be viewed from

approximately its correct center of perspective, so that the angles subtended at the eye by the various images in the

picture will be the same as the subtense angles of the original objects at the camera lens...."

 

...

 

[note: Kingslake then describes the situation for prints, including an example of enlargement from a 35mm negative]

...

 

"The gain in realism obtained by enlarging small negatives in this way is quite marked and often astonishing."

[frode]

<p>I usually look at an image from various distances. Any of them might be more optimal than the others depending on what I am looking for.<br /> There is no "optimal distance in relation to its size" per see. Optimization always need at least one varying input value (here "viewing distance in relation to its size"), at least one output value (here unspecified) and an optimization criteria (a rule/function that says when the output is most optimal, here unspecified).</p>

<blockquote>

<p>I guess there is no upper limit..</p>

</blockquote>

<p>Oh yes, there might be. Most people would for example not even be able to spot a 4 in. x6 in. image at a viewing distance of 1 mile. :-)</p>

<p>Cheers,</p>

[Didier Lamy]

<p>"There is no "optimal distance in relation to its size" per see"<br /><br />Not "per see", but regarding artefacts due to film, artefacts, sharpeness, imho, yes.</p>

[Alan Marcus]

<p>For most images, correct perspective which is intertwined with viewing distance is unimportant. Some images, particularly portraits, correct perspective makes or breaks. This distance is a function of geometric optics. Optimal viewing distance is focal length of the taking lens multiplied by the magnification used to make the final display. Say a 4 x 5 inch print is made on a 35mm camera with a 50mm lens mounted. To make the 4 x 5 with no added cropping, the magnification is 4.2X. Thus viewing distance is 50 x 4.2 = 210mm = 8 1/4 inches.<br>

For an 8 x 10 inch print, magnification is 8.5X thus 50 x 8.5 = 425mm = 16 3/4 inches.<br>

These distances yield correct perspective. Again few images require "correct perspective". </p>