F8 and be there

[blockend]

I've just finished testing three 50mm lenses on a d*g*t*l camera. It was a comparison for video, though the test was still photographs. The lenses were a Canon 1.8, a Yashica ML f2 and an Industar f2.8 in M39. What was interesting is how similarly they performed at f8 - there was nothing in it. Even if I blew the images up massively, there was no difference you'd be able to see in a normal sized print. At f2.8 and f11 there were differences in performance, but at f8, no.

It will be interesting to compare three run of the mill lenses that can be had for £30 or less with more exotic glass. I'm betting there'll be no change at f8.

 

[Richard S. (rich815)]

Were these full frame digital? If not comparing lenses at f/8 and from only the very center of the image circle you're not going to see much difference if at all. Even with full frame or with film any fairly modern lens (60-70 years old or newer) closed down 3-4 stops from open the differences are very, very minor, especially 50's.

 

[snapguy]

which leads

Which leads to the notion that it isn't the box or the glass, it's the eye that counts.

 

[Jim Jones]

At f/8 or smaller stops, diffraction usually limits performance in full frame cameras. Its effect is affected by focus, film grain or pixel size, and subject matter. A rigorous optical bench test for diffraction will not precisely predict practical imaging. Therefore, you may find hot debates on diffraction between technicians and real photographers.

 

[thuggins]

It is well known that image sharpness increases as you close down the aperture and then will decrease again at increasingly smaller apertures due to diffraction. That is to say that any lens has a "sweet spot" for maximum sharpness. This is generally in the f8 to f11 range. F11 is generally not considered problematic for diffraction, with most folks picking f16 or f22 as the point where this becomes noticeable. I suspect the problems you saw were the result of the d!@#$%l sensor and not the lens. Olympus has published various articles regarding the idiosyncrasies of sensors when compared to film, and why film camera lenses do not provide optimum performance on d!@#$%l cameras.

 

[Alan Johnson]

Some time ago I tested the Canon EF 50mm 1.8 and the Industar LD on microfilm.The Canon was getting around 125 lppm and the 4 element Industar around 65 lppm (with higher contrast) in my test at f5.6-f8.The 6 element Jupiter 8 was around 90 lppm.
I expect you are looking at the resolution limit of the sensor, they are not especially high by film standards but hold the contrast up to the resolution limit better.Here is an example of sensor resolution for a reputable camera:
http://photo.net/nikon-camera-forum/00bF5a

 

[Xmas]

Some time ago I tested the Canon EF 50mm 1.8 and the Industar LD on microfilm.The Canon was getting around 125 lppm and the 4 element Industar around 65 lppm (with higher contrast) in my test at f5.6-f8.The 6 element Jupiter 8 was around 90 lppm.
I expect you are looking at the resolution limit of the sensor, they are not especially high by film standards but hold the contrast up to the resolution limit better.Here is an example of sensor resolution for a reputable camera:
http://photo.net/nikon-camera-forum/00bF5a

If you look at MTFs for the post 79 summicron M 5cm and post 94 Elmar M 5cm /2.8 at/5.6 there is not much difference. They both were on the web.

 

[blockend]

The images were taken on an APS-C "crop frame" sensor, which will eliminate edge aberrations. I agree that sensors handle focal nodes differently to film, digital is more demanding of field flatness. I wasn't promoting the comparison as a scientific exercise, just an aesthetic one.

 

[Paul Howell]

I've just finished testing three 50mm lenses on a d*g*t*l camera. It was a comparison for video, though the test was still photographs. The lenses were a Canon 1.8, a Yashica ML f2 and an Industar f2.8 in M39. What was interesting is how similarly they performed at f8 - there was nothing in it. Even if I blew the images up massively, there was no difference you'd be able to see in a normal sized print. At f2.8 and f11 there were differences in performance, but at f8, no.

It will be interesting to compare three run of the mill lenses that can be had for £30 or less with more exotic glass. I'm betting there'll be no change at f8.

Your right at F 8 perhaps not much of difference, but the quality of lens it is not how well it shoots stopped down, it how it performs wide opens. There are lens that are very sharp coming close to Leica, the Konica 1.7, is a very good lens. Often overlooked because the quality of the camera build was not very good is the Petri 55 mm 1.8 7 elements in five groups, well coated, excellent lens, sharp as a tack from 1.8 to 16.

 

[marcmarc]

The rule of thumb I've always heard was that (generally) lenses perform their best 1-2 stops from wide open. I suppose there are differences based on how fast the lens is, such as the same focal length by the same manufacture but one is a f2 and the other a f2.8