Determining Film Resolution

[venchka]

Resolution matters to those of us selling images, in competition with others to
whom resolution matters.

That leaves me free to do whatever I please. I don't have to worry about sales.

[percepts]

This, to me, is the most important decision we make. If we screw this up, the rest doesn't matter.

That said, having good understanding of the technical underpinnings of photography in general and LF in particular are important. The craft side is nearly equally important, because without it we can't reliably realize the art that standing in the correct place allows us to create. IOW, the craft has to serve the art.

For me understanding the resolution thing is nice, an interesting puzzle, and somewhat useful in the field. But even if I know that I should shoot a particular lens at f/11 for best resolution, I'm going to shoot it at f/32 anyway because that's what I need at the time to make the photograph that I'm envisioning.

The following are what a very good quality modern lens can achieve.

f-no. resolution (line pairs per millimeter)
45          35
32          50
22          70
16         100
11         140
8          200
5.6        280
4          400
2.8        560

And that is in the plane of focus. Off that plane but within the so called depth of field, those figures will drop by 50% and more. They also assume perfect film alignment, something rarely acheived in large format photography. And also assume zero camera vibration.

In other words there will be parts of your image achieving 50% or less of those figures meaning at f22 you can expect 4 to 6 enlargemnt max from a 4x5 neg exposed at f22 if you want very high print resolution. Get anything wrong, such as slightly misaligned enlarger and it will be less.

Difficult this large format photography isn't it. Mind you, because most of us never achieve these high standards, we accept lower ones without realising it could be much better. If only we paid attention to the technical aspects and applied them by doing something simple like aiming to expose at f8 whenever possible. That way we achieve much higher on film resolution which in turn gives us more room for error elsewhere in our process.

Thats why resolution is important. Just using 3 stops wider aperture makes a huge difference.

[Bruce Watson]

...meaning at f22 you can expect 4 to 6 enlargemnt max from a 4x5 neg exposed at f22 if you want very high print resolution.

The term "very high" is highly subjective. I've got a number of beautiful 10-12x enlargements that show amazing amounts of very sharp detail. Nice big 125 x 100 cm prints. I've had a lot of comments about how sharp they are, and never had a comment about them being soft in any way. These were all made between f/16 and f/32 with any of five different lenses.

Now it's true that they could be even sharper, but at the expense of things outside the exact plane of focus being even softer. Which would in general be a detriment to the print, which is why I chose the aperture I chose.

I'm not saying that resolution, sharpness, acutance, whatever you want to call it, isn't important. It is important. But it's not any more important than any of the other myriad parameters and decisions that go into making an LF photograph.

Resolution is but one parameter among many.

Difficult this large format photography isn't it.

You can make it as difficult as you like.

[percepts]

The term "very high" is highly subjective. I've got a number of beautiful 10-12x enlargements that show amazing amounts of very sharp detail. Nice big 125 x 100 cm prints. I've had a lot of comments about how sharp they are, and never had a comment about them being soft in any way. These were all made between f/16 and f/32 with any of five different lenses.

Now it's true that they could be even sharper, but at the expense of things outside the exact plane of focus being even softer. Which would in general be a detriment to the print, which is why I chose the aperture I chose.

I'm not saying that resolution, sharpness, acutance, whatever you want to call it, isn't important. It is important. But it's not any more important than any of the other myriad parameters and decisions that go into making an LF photograph.

Resolution is but one parameter among many.



You can make it as difficult as you like.

Correct, resolution is but one parameter. But many people here scan and they will quite happily expose at f22 or f32 and wonder why they don't get more than 2000 - 2500 spi in their scans. Answer, simply because it wasn't in the neg to start with which is why I posted the picket fence method of real world neg resolution testing. So people can do a very simple practical test to see how good their system really is.
All you need to do is to take two images. One at f22 and one at f8 or f5.6 and the difference is resolution immediately obvious just from looking at the negative.

[Dan Fromm]

The following are what a very good quality modern lens can achieve.

f-no. resolution (line pairs per millimeter)
45          35
32          50
22          70
16         100
11         140
8          200
5.6        280
4          400
2.8        560

Percepts, you're confusing the theoretical limits set by diffraction alone with the actual limits set by other aberrations and diffraction. I own some lenses that are in fact diffraction limited, but not wide open and only over small fields.

As a practical matter, we all do very well to resolve as much as 60 lp/mm on film or sensor. No real lens comed close to the limits you've given. Don't even think of telling me about the very expensive lenses made for aerial cameras. I have a number of them, some quite fast, and none is particularly good wide open with normal ordinary white light.

This is why getting a good big print requires a good big negative.

Cheers,

Dan

[percepts]

Percepts, you're confusing the theoretical limits set by diffraction alone with the actual limits set by other aberrations and diffraction. I own some lenses that are in fact diffraction limited, but not wide open and only over small fields.

As a practical matter, we all do very well to resolve as much as 60 lp/mm on film or sensor. No real lens comed close to the limits you've given. Don't even think of telling me about the very expensive lenses made for aerial cameras. I have a number of them, some quite fast, and none is particularly good wide open with normal ordinary white light.

This is why getting a good big print requires a good big negative.

Cheers,

Dan

Not quite, the theoretical limits give more resolution than that. The figures I gave are from Zeiss and have been achieved on 35mm cameras. But yes I agree that with large format lenses, particularly when wide open, you can start to see a fall off in resolution from the figures I posted due to other lens aberations. And that makes it even more difficult to achieve a 20x16 print that is of the highest quality possible.
I know many aren't worried about it but many also assume that because they are using 200 lp/mm film, that will improve their resolution. It won't unless they are technically savvy enough to utilise the films capabilities. Again, that is another reason for doing the simple parctical test of photographing a white picket fence from a known distance. You actually get to see what is really happening. And to also begin to understand that a piece of 4x5 or bigger film flopping about in a film holder really doesn't help things.

[ic-racer]

The basics of photography. You control timing, framing, light-dark and contrast, and focus (or sharpness).

[Greg Blank]

Resolution matters to those of us selling images, in competition with others to
whom resolution matters.

Resolution "Always matters" How one applies it is alway a personal choice. Soft focus can be just as lust worthy as hard line and biting.

Gb 6-12-09

[Dan Fromm]

percepts wrote "The figures I gave are from Zeiss and have been achieved on 35mm cameras."

Please give a citation. The numbers you quote are uncannily close to the Rayleigh limit.

[percepts]

http://www.zeiss.de/C12567A8003B58B9...256CED0054968D

Camera lens news No 2:

last section "Resolution of camera lenses – Where are the limits and why?"

read the full section as they say they can reach the limits of colour films upto 200 lp/mm. At least that seems to be what they are saying. And if you take their example of a 4mm object at 400 meters photographed on 100mm lens at f5.6 and do the maths as in the first post of this thread, then thre result is 500 lp/mm unless I got the maths wrong. I interpret them as saying they reach the film limit cos obviously the film isn't capable of that but the object resolves.

Rayleigh limit or no, they say their lenses are capable and who am I to argue with zeiss.