Method for testing for lens resolution/sharpness

[Greg]

Have been trying to come up with a way to test my lenses for resolution and sharpness at each of their f/stops and then creating a quick reference chart for each lens. I don't have a 120 roll film back any more and shooting probably an average of 6 sheets of 4x5 film for each lens seems like a very daunting expensive project to start and then follow through with.

So came up with the following: Mount my Nikon FX digital camera body onto a Sinar lens board using an older extension tune to get clearance for the body. Then mount it onto the rear standard of the 4x5 Sinar Norma, and mount the view camera lens on the front standard. Create a resolution chart/target mounted in front of a lightbox say about 20 feet away (all done under controlled and repeatable conditions in the basement). Turn on the camera to LIVE VIEW to focus the lens. Now increase the magnification of the image on the back screen and focus. Then I figure I should be able to stop the lens down at each aperture and view and note the image sharpness change on the rear screen. Many years ago did the same for the macro lenses that I was using on a Nikon Multiphot but back then had to do it with film and had to buy a microscope glass slide with a resolution target on it. One exposure for each T or f stop, process the film, then view the negatives through a low power microscope, and then carefully make lens sharpness notes for each lens at each of its apertures. If I recall, this took me 3+ nights to do for 5 lenses, but the lens performance info was invaluable from then on. With the magnified LIVE VIEW screen, should take me 5 or 10 minutes per lens. Also should be able to detect any focus drift. I'm not interested in calculating lines/inch, but only in getting a general idea of how each lens performed.

Has anyone tried something like this? Would appreciate any advice before I permanently dedicate and epoxy a Nikon extension tube to a Sinar lens board.

Thanks

[Bob Salomon]

How will you get short lenses to focus to infinity. How will you check edges and corners?

[Greg]

Mock-up of the set up (using a WA bellows) seems will be able to work with lenses short as 120mm. True won't work with shorter lenses, but the shortest lens I want to test for my 11x14 is a 5.9" F/14 R D Gray Extreme Angle Periscope.

Again totally true won't be able to check edges or corners, but testing with 11x14 sheets of film just too costly. Film costs are around $8.00 per sheet and if the lens has 6 full f/stops that's $48.00 in film costs alone. Looked into shooting X-Ray film but box of 100 sheets goes for $439.00.

I should theoretically be able to check for focus drift by moving the rear standard back and forth (fortunately the Sinar Norma has really precise and repeatable movements).

I already have good notes on the coverages of my lenses, and on some lenses (the 5.9" Gray especially) know from looking at my negatives that the sharpness in the corners deteriorates greatly.

Greg

[Ken Lee]

Have been trying to come up with a way to test my lenses for resolution and sharpness

Are your lenses unique or obscure designs ?

If not, others have probably tested them: you can feel confident that your results will match theirs and you can save yourself some time and trouble.

Stopping most LF lenses down to f/16 - f/22 we reach the sweet-spot of coverage and sharpness.

[Greg]

thanks for the advice. Will do searches in the Forum for the G-Clarons which most certainly someone out there has a tested at one time or another. If this set-up does work, testing costs will be nil and take no time at all. Major source of info on the lenses I've found in The use of Historic Lenses in Contemporary Photography by Paul Lipscombe. Other lenses like the 5.9" F/14 R D Gray Extreme Angle Periscope border on unique and obscure, and 2 brass lenses have no writing on them.
thanks
Greg

[Ken Lee]

You can find some G-Claron lenses tested here (http://www.hevanet.com/cperez/testing.html). Some have criticized the methodology of these tests, but a quick study reveals that for a given focal length and design, modern lenses for LF are fairly similar in performance. The best ones get very close to what is optically possible, "diffration limited" as they say. The others come quite close.

Here is a another series of tests (http://www.arnecroell.com/lenstests.pdf), quite rigorously done. Several G-Clarons are mentioned.

From what I can tell, best-of-breed lenses perform no more than 10-15% better than their peers. Given variations between samples, it's hard to evaluate the actual number.

To find the best aperture on any lens, we can easily inspect the ground-glass with a loupe.

Using modern methods, small differences in contrast and color rendition are easily corrected. Other factors rise in importance, like maximum aperture, price, weight, filter-size and blur rendition.

[Lachlan 717]

Why not test them using a couple of sheets of 4x5" film taped in the 11x14 holder, one in the centre and one in the corner?

Use cheap and/or expired film. It won't matter.

[Greg]

Progress report:

Resolution target was made by cutting vertical parallel 1mm slits in a sheet of the black film interleaving paper. Slits were spaced 9mm apart and I left 1" of the paper above and below the slits uncut to keep the whole sheet intact. Mounted my 10x11" Graphiclite lightbox on a shelf at eye level height about 20 feet away and hung the slitted sheet of black paper in front of the lightbox.

Sinar Norma with its rather stubby but beefy rail mount is mounted directly onto a flat top round plate atop of a Linhof Heavy Duty tripod. Legs extended to place the camera at eye level. This is a very, very stable setup. I have one lens board that will accept Nikon lenses, and by using a short extension tube and various adapters was able to easily mount a FX body onto the board. Now need to find my rail clamp that has an adjustable horizontal piece to attach to the bottom of the FX camera body to stabilize it and the whole rig together.

Mounted a 305mm G-Claron onto the front standard. Shutter opened. Aperture wide open. Camera on with LIVE VIEW and LCD rear screen set at maximum magnification.

Focused in on the resolution target. It was like looking through a long range microscope. Put up an old film data sheet and was easily able to read the fine print.

Viewing the 3.2" (8cm) LCD screen looked great at full aperture; but as I stopped down the lens, image noise increased and soon totally degraded the magnified image displayed on the LCD. So now for each lens will just shoot an image at each f stop and compare the images in Adobe Bridge.

Edge/corner resolution: At first thought about shifting the front standard to the left and the rear standard to the right but this only added up to 4 cm off from center axis. Then I saw the obvious... swing the front standard to one side its maximum of 35 degrees from center axis and refocus the lens (per the Cosine Law?). In essence I am now able to measure the edge of a 70 degree image circle. The front Sinar standard is marked in degrees, and has an accuracy of less than 1 degree.

Each lens should take only 5 minutes to capture the resolution target in the center and at the edge for each f stop. Open the images for each lens in Bridge, run an action to crop out the center portion of the frame, and interpret the sharpness data, and maybe finally even have Photoshop print up a proof sheet of all the images for each lens, again probably taking only 5 minutes per lens. I will then have sharpness data for all my lenses without spending a cent. Ice storm due here tomorrow morning, ideal time to run sharpness tests on all my lenses, all in a warm basement.

[Bob Salomon]

Progress report:

Resolution target was made by cutting vertical parallel 1mm slits in a sheet of the black film interleaving paper. Slits were spaced 9mm apart and I left 1" of the paper above and below the slits uncut to keep the whole sheet intact. Mounted my 10x11" Graphiclite lightbox on a shelf at eye level height about 20 feet away and hung the slitted sheet of black paper in front of the lightbox.

Sinar Norma with its rather stubby but beefy rail mount is mounted directly onto a flat top round plate atop of a Linhof Heavy Duty tripod. Legs extended to place the camera at eye level. This is a very, very stable setup. I have one lens board that will accept Nikon lenses, and by using a short extension tube and various adapters was able to easily mount a FX body onto the board. Now need to find my rail clamp that has an adjustable horizontal piece to attach to the bottom of the FX camera body to stabilize it and the whole rig together.

Mounted a 305mm G-Claron onto the front standard. Shutter opened. Aperture wide open. Camera on with LIVE VIEW and LCD rear screen set at maximum magnification.

Focused in on the resolution target. It was like looking through a long range microscope. Put up an old film data sheet and was easily able to read the fine print.

Viewing the 3.2" (8cm) LCD screen looked great at full aperture; but as I stopped down the lens, image noise increased and soon totally degraded the magnified image displayed on the LCD. So now for each lens will just shoot an image at each f stop and compare the images in Adobe Bridge.

Edge/corner resolution: At first thought about shifting the front standard to the left and the rear standard to the right but this only added up to 4 cm off from center axis. Then I saw the obvious... swing the front standard to one side its maximum of 35 degrees from center axis and refocus the lens (per the Cosine Law?). In essence I am now able to measure the edge of a 70 degree image circle. The front Sinar standard is marked in degrees, and has an accuracy of less than 1 degree.

Each lens should take only 5 minutes to capture the resolution target in the center and at the edge for each f stop. Open the images for each lens in Bridge, run an action to crop out the center portion of the frame, and interpret the sharpness data, and maybe finally even have Photoshop print up a proof sheet of all the images for each lens, again probably taking only 5 minutes per lens. I will then have sharpness data for all my lenses without spending a cent. Ice storm due here tomorrow morning, ideal time to run sharpness tests on all my lenses, all in a warm basement.

One thing to keep in mind is diffraction. With 35mm a lens is diffraction limited at much larger apertures then for large format lenses.