What is the biggest, perfectly sharp format you can get from your sharpest negatives?

[richard ide]

You can not change the information recorded on the negative.

You can print the information by varying magnification or tonal values but cannot change the negative parameters.

 

[Allen Friday]

Situation 1. A viewer stands at the normal viewing distance from a 16x20 inch print. The viewer moves to look at a 24x30 inch version of the same image, made from the same negative. Again the viewer stands at the normal viewing distance from the print. To the viewer, the prints will appear equally sharp.

Situation 2. A viewer sits in a chair at the normal viewing distance from a 16x20 inch print. The print is replaced by a 24x30 inch version of the same image, made from the same negative. The viewer does not move. To the viewer, some areas which were sharp in the 16x20 inch print will appear less sharp in the 24x30 inch print.

Ralph or Q.G., do either of you disagree?

 

[A49]

You can not change the information recorded on the negative.

You can print the information by varying magnification or tonal values but cannot change the negative parameters.

That is absolutely true. But what quality in terms of sharpness and resolution you need to catch in the negative depends on the size of the negative that you shoot and the print size that you intend to make. At least if you want to have a sharp print not only from the normal viewing distance from where you can overlook the whole picture but also in the critical nose on the print distance.

Best regards,
Andreas

 

[Edward_S]

One possible source of dissenting views lies in this statement from Ralph (post #79):

"The factors controlling DoF are subject distance (u), circle of confusion (c), and aperture (N) and focal length (f) of the taking lens. None of these factors can be changed after the exposure. DoF is locked into the negative."

In fact, the circle of confusion is arbitrary, depending on the physiology of the viewer; the circle of confusion that matters most lies in the eye of the beholder. So whereas the subject distance, aperture and focal length of the lens are all indisputable quantities for a given exposure, the circle of confusion is not necessarily even defined at that point, and I would be perfectly at liberty to choose a different value from Ralph, Q.G., or indeed anyone else on APUG for that matter. In so doing, I would then alter the DOF as calculated by Ralph's equations, after the exposure had been made.

If I place almost any of my slides on the light box and view them with the unaided eye, they all seem to be 'perfectly sharp' and have unlimited depth of field. But if I look at them through a loupe, I see that some do have unlimited DOF, some have narrow DOF and some aren't really sharp at all. But this judgement is made based on the size of the circle of confusion on my retina as much as by the properties of the image on the slide.

With respect,

Edward.

 

[JS MD]

Uhh... my prints don't have pixels.

Line pairs per mm

 

[RalphLambrecht]

... You say, if I get you right, that the DoF is "locked" into the negative, as the enlargement is supposed to be made in a way that the larger the enlargement, the larger the viewing distance, so that enlarging is out of the equation and in fact in the equation you give "enlarging" does not seem to take part. ...

Exactly.

... If you compare equal magnifications of the negatives, micro formats don't have any more DoF than "macro" formats. ...

I'm willing to compile the following pictorial evidence if it is accepted as proof beforehand:

1. Two images, one taken with a 35mm the other with a 80mm lens from the same viewpoint and subject distance (u) and at the same aperture (N).
2. Both images enlarged to the same subject magnification.

Then we can visually investigate the images and compare their DoFs.

I claim it will disprove Q.G.'s statement in post #75:

But enlarge both to the same size print, and you will have not only achieved equal magnification, but also equal DoF.

 

[JS MD]

new eyeglases ?

 

[Diapositivo]

For the sake of added clarity, I think that we should specify that when it is said that DoF, CoC etc. depends on magnification, it is meant that they depend on magnification of equal sized portions, in absolute terms, of the negative.

Say we have a landscape with a bell tower. We take this landscape with a 100mm on a 135 camera, and with a 100mm with a 4:3 film camera (again :wizard: ).

If the bell tower is tall 10mm in the negative 135 camera, it is equally tall 10mm in the negative of the 4:3 camera. The difference being that in the 4:3 camera the bell tower will occupy a proportion of the side of the negative that is twice as much as in the 135 camera.

This means that in the 4:3 format the bell tower will seem, considering the entire picture, to occupy the negative as it would have looked in a 135 camera if it had been photographed with a 200mm lens. The 4:3 camera will give a "bigger tele flair" to the image in comparison to the 135 camera, both regarding perspective and DoF, when and only when we consider both images in their entirety.

I note here en passant that if we take our scissors and cut the 135 negative in exactly the same absolute dimensions of the 4:3, we obtain a negative which is identical to the one of the 4:3 format. Identical in "tele flair", in perspective, in DoF. Not a surprise, as both images have been taken with a 100mm so they project an image behind them which has the same absolute dimensions.

Now let's revert back a moment. We did not cut the 4:3 negative. We have two different negatives, a 135 and a 4:3. The bell tower is 10mm in both negatives, but there is "more real estate negative" around the bell tower in the 135 than in the 4:3 negative.

If we print those two negatives on two pieces of sensitive paper, and we print them in such a way that we have the same absolute height of the bell tower in the paper, we print the two negatives at equal magnification. Let's say that the bell tower was 10mm in both negative, is now 10cm in both prints, both negatives have been enlarged 10 times, so it is equal magnification.

If the two pieces of paper have the same dimensions, and we are printing all the 4:3 negative, this means that we are printing only the central portion of the 135 film. We don't have space on the paper to print the "extra negative real estate" of the 135.

If the paper where we print the 135 film is instead larger than the 4:3 (in the same proportion in which the 135 is larger than the 4:3 format) we can print both negatives at full fram, we will still print at equal magnification, and we have two prints of different size. The "real estate negative in excess" will now be printed in the "real estate paper in excess".

The central portion of the print obtained from the 135 negative will be identical to the print obtained from the 4:3 format. Identical not just in the size of the bell tower, but also in the perspective, tele "flair", DoF, proportions between plans nearer and plans further etc. Identical in all respects, as it is obtained with a lens with the same 100mm focal length.

Keeping all of the above in mind, and re-reading the thread, I hope the common fallacy will be evident of thinking that we can obtain a larger DoF by changing capture format.

Fabrizio

 

[Diapositivo]

I will make another example.

We take architecture photography with a 135 camera, and no movements. We use a 56mm lens as this gives the field of view that we want (we are maniacal about this). We are often struggling with want of DoF. We want to solve this problem. We also have a 28 mm lens. We notice that the DoF marks on the 28mm barrel tell us that the 28mm has much more DoF than the 56mm, which has DoF marks which are much less generous.

An idea crosses our mind. We have read, and everybody knows after all, that the smaller the format, the shorter the "normal" focal length, and also that the shorter the focal length, the higher the DoF. So, we reason, since now on, we will use a 4:3 film camera (there is one made on purpose for APUG forumers). Everybody knows that the 28mm on a 4:3 camera is the "equivalent" of the 56mm on a 135 camera. So we will have the same angle of coverage of the 56mm in 135 (we are maniac about this), BUT we will have the larger DoF of the 28mm as we are using a 28mm and nobody can deny it. Brilliant!

So we have our 4:3 film camera which we created for the sake of this discussion. We buy an adapter, and we use our 28mm on our 4:3 body.

When printing the negatives, we will soon find that the trick did not work. We have the same angle of view of the 56mm, and also the same DoF. We are using only the central part of the 28mm when using it on the 4:3. That's why it looks like a 56mm. And we are "enlarging", or "magnifying" the central portion of its image. That's why the DoF is reduced under every respect.

Fabrizio

 

[Diapositivo]

I will make another example.

We take architecture photography with a 135 camera, and no movements. We use a 56mm lens as this gives the field of view that we want (we are maniacal about this). We are often struggling with want of DoF. We want to solve this problem. We also have a 28 mm lens. We notice that the DoF marks on the 28mm barrel tell us that the 28mm has much more DoF than the 56mm, which has DoF marks which are much less generous.

An idea crosses our mind. We have read, and everybody knows after all, that the smaller the format, the shorter the "normal" focal length, and also that the shorter the focal length, the higher the DoF. So, we reason, since now on, we will use a 4:3 film camera (that one made on purpose for APUG forumers). Everybody knows that the 28mm on a 4:3 camera is the "equivalent" of the 56mm on a 135 camera. So we will have the same angle of coverage of the 56mm in 135 (we are maniac about this), BUT, we reason, with the DoF of the 28mm. That's a 28mm and nobody can deny it, and the DoF marks on it cannot change just because we mounted it on another camera, we reason.

So we have our 4:3 film camera which we created for the sake of this discussion.
We buy an adapter and we mount our 28mm on the 4:3 machine. We go taking pictures.

But when we enlarge the negatives, we will be very disappointed in finding that the trick did not work.

The angle of view, true, of the 28mm on the 4:3 camera is equivalent of the angle of view of the 56mm on the 135 camera.
But we find that also the DoF (and the perspective etc) of the image we obtained is equivalent to the DoF of the 56mm on the 135 camera, and not to the DoF of the 28mm as inscribed on its barrel.

What, we think. The DoF marks on our 28mm barrel lied to us?

They didn't. But they were traced for use on a 135 body. By using it on a 4:3 body, we are enlarging (magnifying) the central portion of the image. So all the Circles of Confusion are now much larger, and much more evident. So the "acceptable" focus is not any more corresponding to the DoF marks on the barrel of the 28mm lens. It is, rather, corresponding to the DoF marks on the barrell of the 56mm lens.

When the engineers calculated the DoF signs on the 28mm, they calculated a certain magnification in print, corresponding on a certain magnification of the negative. And they considered that the bigger the print, the bigger the viewing distance. So the DoF marks are the same for every print size.

But now that we use the 28mm on the 4:3 format, we are enlarging the central portion of the image, and for EACH of those full frame print size we are enlarging the negative twice as much, so the calculations resulting in the DoF marks on the 28mm barrel don't hold true any more.

Fabrizio

 

[RalphLambrecht]

Fabrizio

A simple 'yes' or 'no' to my proposal in post #107 will suffice.

 

[RalphLambrecht]

Ralph, make it more extreme if possible by using a 20-24mm lens compared to a 135-200mm lens. The difference in DOF should be greater.

I can use 24 and 85mm.

 

[RalphLambrecht]

... In fact, the circle of confusion is arbitrary, depending on the physiology of the viewer; the circle of confusion that matters most lies in the eye of the beholder. ...

Edward, the CoC is not that arbitrary. Human eye resolution is limited to about 7 lp/mm at closest comfortable viewing distance distance (10 inches). That distance is about equal to the diagonal of an 8x10-inch print. Take your film format and calculate the magnification required to achieve an 8x10 print and you will find the CoC required to get to the 7 lp/mm resolution. For 35 mm this is about 0.025 mm for example, and requires that you have a negative resolution of 45 lp/mm.

Of course, you can pick a different eye resolution, and that will change your CoC, but you can also accept the standard of 7 lp/mm and have a standard CoC. Without it, lens makers couldn't add depth-of-field markings on their lenses.

 

[A49]

1. Two images, one taken with a 35mm the other with a 80mm lens from the same viewpoint and subject distance (u) and at the same aperture (N).
2. Both images enlarged to the same subject magnification.

Then we can visually investigate the images and compare their DoFs.[/I]

You´re undoubted right with that. I checked it and calculated some examples. Even if you adapt the 35mm shot´s CoC to a proportional smaller amount (to 35/80 multiplied with the CoC of the 80 mm shot) so that the resulting resolutions in both print enlargements are equal, the DoF of the 35mm is wider. That is because the DoF is simply not linear to the CoC and also to the focal length. The math behind DoF is a little more complex and therefore not very demonstrative.

Kind regards,
Andreas

 

[Diapositivo]

Ralph, the test you propose in post #107 is, if I get you right, the same already performed and published in the Luminous Landscape web site, and which I have linked in this thread:

http://www.luminous-landscape.com/tutorials/dof2.shtml

The conclusion which you can derive from this "paper" are extensible to our discussion. DoF is a function of "magnification", which is the result of two factors, focal length and enlarging. You can compensate focal length with enlarging so that you maintain the same magnification, in this case you will have the same depth of field.

You seem to be more mathematically inclined that me. This page by Norman Koren should be more understandable to you than to me:
http://www.normankoren.com/Tutorials/MTF6.html#DOF_focal_length

The math just makes me but the text which seems to be in accordance with what I say is:

It is well known that short focal length lenses have large apparent depths of field and long telephoto lenses have small apparent depths of field. There are some very practical reasons for this conception, but it isn't quite true. DOF is much more closely related to magnification and f-stop; DOF expressed in distance is nearly independent of focal length. It appears smaller with telephoto lenses because it is smaller when expressed as a fraction of the lens-to-subject distance, s.

(bold mine)

Cheers
Fabrizio

 

[RalphLambrecht]

Fabrizio

No, the luminous-landscape test is not what I'm proposing. In that test, the camera was moved and did not remain stationary (look at the background tree in relationship to the gremlin). The same subject size was achieved by altering the viewpoint. I said all along subject distance (u) must remain the same, as does aperture (N) to test the influence of focal length alone. In the test I propose, the camera will remain stationary, only the focal length will change. Same subject magnification will be achieved during enlargement, as proposed by Q.G.'s statement in post #75:

But enlarge both to the same size print, and you will have not only achieved equal magnification, but also equal DoF.

which I claim to be incorrect.

 

[JS MD]

silly thread - about nothing

 

[holmburgers]

silly thread - about nothing

*thumbs up*

 

[RalphLambrecht]

OK, I won't post the test results then.

 

[cowanw]

silly thread - about nothing

Well I have been wondering about the difference in Magnification on film vs on print for a long time. So I am looking forward to the conclusion.
Time life would seem to be on Ralph's side. Page 116 -119 of The Camera in the series shows Ralph's test. Not only does the focal lenth affect depth of field in and of itself, so does it by way of the fact the f2.8 (for example) is a much smaller hole in a 28mm lens than f2.8 in a 200 mm lens.
I hope Ralph presents his tests whether or not others approve.

 

[cowanw]

Aw sh*t. too late

 

[RalphLambrecht]

Aw sh*t. too late

Bill

Don't worry. Not posting them doesn't mean I won't email them to people who are interested. I may even keep them for another publication to clarify the issue and a common misunderstanding.

 

[A49]

silly thread - about nothing

For those who shoot with pinhole cameras yes. Surely there is a pinhole topic anywhere on APUG.

I´ve learned a lot from this enlargement factor and DoF discussion and I think people who already know all this or don´t like to learn should show refined manners and be quiet.

As long as there at least two people interested in this discussion it is worth to be kept alive.

Kind regards,
Andreas

 

[cowanw]

Well I would appreciate the contact.
Besides which I wanted to get in touch with you about cyan in B&W enlarging vis a vis your book:confused:

 

[MattKing]

Well I would appreciate the contact.
Besides which I wanted to get in touch with you about cyan in B&W enlarging vis a vis your book:confused:

Bill:

Offer Ralph a "virtual" beer.

I have it on good authority that he needs one .