Resolution

[tim]

The resolution of a lens depends on colour of the light (wavelength lambda), the clear aperture of the lens, d, and the distance from the lens to the image, f. A pair of lines, black on white, can be resolved (so that the brightness of the space between them is twice that of the lines) if the space between them is 1.22 x lambda x f/d. Assuming the width of the lines to be small, and the object distance much larger than the focal length, then f/d is close to 1/F, the relative aperture of the lens (the F number). For typical cine camera lens apertures, and for lambda = 550 nm (green light) resolutions for an otherwise perfect lens are

F No.-------Lines/mm

1.2-------1242

1.8 ------- 828

2.8------- 532

4 ------- 373

5.6-------266

8-------186

11-------135

Note that lens resolution should be around twice film resolution for reliable measurement.

[jessh]

I know I have heard otherwise so I did a little research, your statements that the resolution of a lens is higher when the iris of the lens is wider (lower fstop number) appears to be correct in terms of the Diffraction that occurs when the iris is stopped down: "Diffraction, an optical phenomenon, occurs when light passing through a small hole scatters and renders the image unsharp. Iris openings of f/16 or smaller may cause noticeable loss of sharpness. Difraction is a function of the absolute size of the aperture" (The Filmmaker's Handbook, pg 104) The problem is that that diffraction is not the only thing that decreases sharpness in a lens, Aberations cause lenses to get decresingly sharp the wider open the iris is, "Nearly all aberrations, except distortion, are most apparent when the lens iris is wide open or nearly so and they limit lens sharpness. As the iris is stopped down, the effect of the aberrations becomes less pronounced."(103) "Most lenses are sharpest when stopped down two to three stops from the widest aperature. For example, an f/2 lens would be sharpest around f/4 to f/5.6" (104)
I also ran across some camera tests the other day where the resolution was higher when the lens was stopped down some than when it was all the way open, I don't remember exactly where they were but I believe it was some link from: http://www.bealecorner.net/trv900/respat/index.html

Hope this was of help to someone. If I am wrong about anything please let me know.

~Jess

[tim]

The important point is that lens resolution cannot be better than that predicted from the aperture and focal length. If a lens is not effective at apertures of greater than say, f/8, then it cannot be used to test film resolutions of better than about 150 lines/mm.

[Guest]

oh, there is a commercial full-length movie film, produced in S8, blown up to 35 mm and distributed in cinemas. As far as I know it had been made in the end 1970ies and had been done all optical, of course. I watched the film in the mid 1980ies at TV, only I don´t like the style of the autor. It´s made by Herbert Achternbusch and the satiric title of the film is "heilt hitler" (in English "cure hitler"). He did the film in S8 only by lack of resources.

Pedro

[S8 Booster]

Although not s8 - actually medium format this site contains a lot of interesting info. Scroll down the page.

Site: Lens resolution testing: http://people.smu.edu/rmonagha/mf/resolution.html


Reference chart: http://people.smu.edu/rmonagha/mf/resolution1.jpg

R

[S8 Booster]

If the resolution drops 2:1 to the film plane due to the camera lens, will the resolution drop 2:1 with the projector lens too?

Target (1:1) -> Camera lens (2:1) -> Film (1:1) -> projector lens 2:1 = 900 lines? K40 3400K?

If this is correct is the real film resolution higher than 900? lines?
Or, will the lenses operate differently in "compressing" vs "decompressing" the images?

Just wondering:

R