Portrait perspective: Quiz and two questions

[Jerry Fusselman]

Jerry, do you ever get to taking a picture?

Nice of you to ask. I took pictures three nights so far this week. I am thrilled with my pictures from Monday. I am going to scan some of them---my first 6x24 scanning.

Do WE have to take a test to fulfill your need for mathematical precision which IMHO does not matter one bit when you are taking photographs?

You misunderstood me entirely. No one requires that you find every topic interesting. No one requires you to read all posts. No one requires that you answer posts. As I said in paragraph 1, the quiz was supposed to fun; if not, no one requires you to do it. If you were shocked and appalled that a quiz appeared in my post, I suggest you read titles of posts a little more closely.

I was hoping the word "quiz" in the title would bring the kind of poster who thinks a little quiz on perspective might be fun. Paraphrasing Scarlett O'Hara, "Quiz fun?!? Fun for teachers, you mean."

Maybe I should have included the word "math" in the title? Would that have helped you steer clear?

Here is another confusion of yours, but it is shared by many others: I never said I worry about mathematical precision while taking pictures. As it happens, I do no math worth mentioning while taking pictures. The math in my post is for camera design and lens choices almost exclusively. Math sometimes helps me figure out what to build, buy, or bring. When I am out there shooting, I use what I brought and do the best I can. At that point, math is too late.

The reason for math is that is often faster and easier for me than trial and error. I also enjoy it. But the amount of math we are talking about here is about five orders of magnitude less than what appears in some photography books, so I am just a math dabbler by comparison to some of the heavyweights.

[Helen Bach]

Well, I think that I've read through every post and nobody has pointed out that the entrance pupil is the centre of perspective of a lens, not the front nodal point. That is why when you use a fixed-lens camera to take a panorama with multiple shots you should rotate the camera about the entrance pupil. Apologies if I have missed a post that has already pointed this out.

Best,
Helen

[Jerry Fusselman]

Well, I think that I've read through every post and nobody has pointed out that the entrance pupil is the centre of perspective of a lens, not the front nodal point. That is why when you use a fixed-lens camera to take a panorama with multiple shots you should rotate the camera about the entrance pupil. Apologies if I have missed a post that has already pointed this out.

Helen, yours is definitely the first to make this point. Thanks. Do you remember any reference to this idea of rotating the camera about the entrance pupil and that being the location that determines perspective? I seem to recall that there was some intelligent disagreement about the best rotation point for panoramas, but I cannot recall that it was well resolved anywhere.

[Ralph Barker]

. . . Do WE have to take a test to fullfill your need for mathematical precision . . .

Juergen - the test is being conducted on Sunday afternoon at my place in Rio Rancho. But, you have to bring the tequilla for the pre-test margaritas.

[Phong]

[*]What is this you are saying about requiring a lot more skill from the photographer? Are you saying that a short perspective is rather poor and therefore takes some work to fix? But how can too-close perspective be fixed? You are still close---there is no way undo that except by moving back, I would think.

I was alluding to non camera/lens related stuff like communicating with your subjects, establishing a rapport with them, getting the right expressions, etc. Sometimes these issues dictate subject distance, hence lens choices. Yousuf Karsh is recognized as one of the great masters of portrait photography (I certainly think so); yet in analyzing his photos I often find the lens choice too short, making the hands looming in the foreground in some cases. My guess is that he wanted to be close enough to his subjects, hence forego a more "ideal" focal length. Communicating with the subject is the number one skill in a portrait photographer in my opinion.

I agree that 85mm is adequate, sure it's okay, but 300mm really looks better to me. When you and others call shorter focal lengths "adequate," that does not sound like a ringing endorsement to me. It almost sounds like you are agreeing with my point, and saying the best head-and-shoulders portrait perspective is taken with longer lenses than are commonly used. Am I misreading you?

You may be. Perhaps I should have said "more than good enough" rather "adequate".
Aside from the issues I mentioned above, 300mm on 35mm film is a terrible perspective in my view; it flattens the face too much, and also gives the look from a distance (like what we see on newspapers of say presidents and heads of state). My taste goes for 90-105mm for heads and shoulders; as heads and shoulders make up a very small percentage of my portraits, I use wider lenses much more often. On 8x10, my current favorite focal length for portraits is 240mm (I have 4 different 240mm lenses at the moment). 4 240mm's and a 450mm. No 300mm (tried that), no 360mm (tried that too).

Do you think a 35mm professional who favors 300mm when he is trying to get the most beautiful look possible might read the Kodak book and decide to switch to 75mm?

???? You were quoting some unnamed "pros", so I offered the Kodak book as a counterpoint. In both cases, my reaction is "so what?", or perhaps more politely, "uhm, interesting, but that's not me". Anyway, even if 300m was my favorite focal length on 35mm, I wouldn't want to use a 1800mm lens on a 8x10 camera because it would be so unwieldy.

Regards,

- Phong

[Leonard Evens]

Panoramas and nodal point:

I did a google search, and everything I found suggested that for panoramic photos, you rotate about the nodal point. They don't say which nodal point, but it is fairly clear it must be the front nodal point.

[Jerry Fusselman]

Juergen - the test is being conducted on Sunday afternoon at my place in Rio Rancho. But, you have to bring the tequilla for the pre-test margaritas.

Great! Can I come to Rio Rancho too? I already finished my answer to the test---it appears in post 42 as well as the thread with the super-exciting title "Computing seven lens variables from three inputs." Since I have answered the quiz already, can I have two margaritas? For extra credit, if you like, I can bring my latest camera that required table 1 for its construction.

[Ralph Barker]

Great! Can I come to Rio Rancho too? . . .

Of course! Anyone who brings tequilla (or fresh chilis [red and green]) is welcome.

[Helen Bach]

Panoramas and nodal point:

I did a google search, and everything I found suggested that for panoramic photos, you rotate about the nodal point. They don't say which nodal point, but it is fairly clear it must be the front nodal point.

It's a very common error, and the internet is full of unreliable information. I'm not at home at the moment, so can't quote Sidney Ray. I can't find the Zeiss document(s) that my memory pictures so vividly*, but here is another.

It isn't difficult to demonstrate that it should be the entrance pupil.

Best,
Helen

*later edit: found an example "The entrance pupil position is the correct position of the axis of rotation when making a panorama image by combining individual images of a scene." Will you trust Zeiss? Better to understand why by working it out than take it on trust, no matter how reliable the source. The fact that it is the centre of the entrance pupil makes it quite easy to find the correct point by parallax when looking into the front of the lens.

[Leonard Evens]

Helen appears to be right about something, but I'm still not sure what.

It does appear to be true that it is the center of the entrance puil which should be over the axis of rotation when doing panoramic photography. When you rotate about that position, you don't have any parallel errors. According to several sources which appear to know what they are talking about, this is commonly miidentified as the front nodal point. Be that as it may, the methods recommended for finding the proper point rely on checking for parallax and in fact find it whatever it is called.

It also seems clear that the entrance pupil need not be centered on the front nodal point, although Jacobson, who presumably is the ultimate expert, suggests in his lens FAQ that the distance from the entrance pupil to the front nodal point is often negligible.

The other thing which seems to be clear is that the variables in the various formulas we have been discussing, which are all based on the lens equation, are measured from the principal planes, which are the planes centered on the nodal points and perpendicular to the lens axis. Also ray tracing is done using the principal planes and nodal points.

I am not sure how to make sense of all of this. In particular, just where is the center of perspective? It seems that one can make arguments for either the center of the entrance pupil or for the front nodal point. Parallax arguments suggest the former, while ray tracing and lens formulas suggest the latter. The parallax arguments seem convincing, but then we need an explanation for why the nodal point is wrong.

To add to the confusion, it would seem that the arguments used in deriving DOF formulas use the distance from the exit pupil to the film plane. If this distance is different than the distance to the rear principal plane, there seems to be a problem.

I will keep looking, but if someone else can clarify all this and provide a reference, I would appreciate it.