How much sharpness is needed?

[Montanawildlives]

I'm just starting with 35 mm film photography with a Nikon Fe. I'm looking for a wide angle AIS lens and have read quite a bit about the 20, the 24, and the various versions of the 28. I'm wondering how important sharpness is going to be if I am shooting typically with 400 or 800 speed film. Other than the terrible versions of these lenses, would I even be able to tell the difference on fast 35 mm film between, for example, the 28mm AI and the 28mm AIS? Wouldn't both of these lenses, for example, out -resolve 800 speed 35 mm film? And with this also be true of 100 or 200 speed film?

I suppose if I get the negatives scanned at various resolutions that might make a difference also? Standard scans versus enhanced or high-resolution scans? A popular online film developing and scanning service provides standard scans at 1024 by 1536. Surely any decent lens could accomplish this and anything better than decent is just wasted sharpness isn't it?

 

[E. von Hoegh]

Film at 400 or 800 iso will be the limitting factor.

 

[MattKing]

I'm just starting with 35 mm film photography with a Nikon Fe. I'm looking for a wide angle AIS lens and have read quite a bit about the 20, the 24, and the various versions of the 28. I'm wondering how important sharpness is going to be if I am shooting typically with 400 or 800 speed film. Other than the terrible versions of these lenses, would I even be able to tell the difference on fast 35 mm film between, for example, the 28mm AI and the 28mm AIS? Wouldn't both of these lenses, for example, out -resolve 800 speed 35 mm film? And with this also be true of 100 or 200 speed film?

I suppose if I get the negatives scanned at various resolutions that might make a difference also? Standard scans versus enhanced or high-resolution scans? A popular online film developing and scanning service provides standard scans at 1024 by 1536. Surely any decent lens could accomplish this and anything better than decent is just wasted sharpness isn't it?

To your first question, any of the lens options you mention will be capable of giving excellent results, no matter what film you use. Assuming of course you are talking about lenses in excellent condition.

With almost all modern (as in the last quarter of the 20th century or later) lenses, sharpness differences are really quite small. You need to be printing quite large to see the difference between most lenses.

And if you are working with scanned files, then yes the scanning resolution - and equally importantly the quality of that scanning - has a huge effect on the results.
Oversimplifying greatly, if you want to make a good quality 11x14 print from a file, you need at least a 3300 x 4200 pixel file. If you don't have that information in your file to start with, you need to use digital tools to add it artificially.

More generally though, don't let the initial scanning resolution for cheap scans determine any choice you might make. A 1024 x 1536 file will basically give you a 3.5" x 5" print, and won't even display full frame on a so called "HD" screen.

With decent technique even a mid-range lens used with most 35mm film will give you a negative that is easily enlarged optically to an 11"x16" print. And if you shoot slide film, a nicely projected 3' x 4.5' image on a screen.
You want your negatives or slides to have those capabilities, so that you can use them in the future, even if your initial use is less demanding.

 

[Les Sarile]

None of the OEM prime lenses that I've tested so far have been the limiting factor. Poor focus, user error and scanning are the weakest links followed by film and then lens.

 

[warden]

....

More generally though, don't let the initial scanning resolution for cheap scans determine any choice you might make. A 1024 x 1536 file will basically give you a 3.5" x 5" print, and won't even display full frame on a so called "HD" screen.
....

Much agreement here. Also don't let fast film influence your decision on lenses. Sure as day follows night you'll eventually be pining for slow films, bigger scans, bigger prints, etc. Having the best lens you can reasonably afford will allow you to experiment without concern for the lens.

 

[RalphLambrecht]

To your first question, any of the lens options you mention will be capable of giving excellent results, no matter what film you use. Assuming of course you are talking about lenses in excellent condition.

With almost all modern (as in the last quarter of the 20th century or later) lenses, sharpness differences are really quite small. You need to be printing quite large to see the difference between most lenses.

And if you are working with scanned files, then yes the scanning resolution - and equally importantly the quality of that scanning - has a huge effect on the results.
Oversimplifying greatly, if you want to make a good quality 11x14 print from a file, you need at least a 3300 x 4200 pixel file. If you don't have that information in your file to start with, you need to use digital tools to add it artificially.

More generally though, don't let the initial scanning resolution for cheap scans determine any choice you might make. A 1024 x 1536 file will basically give you a 3.5" x 5" print, and won't even display full frame on a so called "HD" screen.

With decent technique even a mid-range lens used with most 35mm film will give you a negative that is easily enlarged optically to an 11"x16" print. And if you shoot slide film, a nicely projected 3' x 4.5' image on a screen.
You want your negatives or slides to have those capabilities, so that you can use them in the future, even if your initial use is less demanding.

+1
This calculates to about 60 lp/mm and all decent Nikon lenses are capable of that.

 

[tedr1]

I'm not a Nikon expert however as you are probably aware the differences between AI and AIS are mechanical not optical. The 28 AI and AIS may be identical optically and therefore yield the same sharpness. Lenses of the same focal length and different apertures on the other hand are certain to be optically different. Lens performance evaluation under laboratory conditions may be able to reveal small differences in optical quality that under real-world conditions (hand-held, subject motion, depth of field, focus error....) cease to be important.

See also this thread on a closely related subject https://www.photrio.com/forum/threads/how-much-sharpness-do-you-really-need.162742/

 

[Montanawildlives]

+1
This calculates to about 60 lp/mm and all decent Nikon lenses are capable of that.

Ralph can you clarify for me which calculation you are referring to? Thanks.

 

[Montanawildlives]

Thanks everyone. I'm sure it will be a long time before anything other than user error causes me difficulty.

 

[Ste_S]

I use a Nikon FE, mostly with the AI 35mm f2.8 amd AIS 50mm f1.7. No problems with sharpness with either of those lenses.
For scanning, get at least an 8MP equivalent scan. I find they give nice A4 ink jet prints at home and A3 C-type prints at a lab

 

[Vincent Peri]

I'm not a Nikon expert however as you are probably aware the differences between AI and AIS are mechanical not optical. The 28 AI and AIS may be identical optically and therefore yield the same sharpness....

The 28 2.8 AIS is an all new (at the time) optical formula, different from the 28 f2.8AI.

https://kenrockwell.com/nikon/28f28ais.htm

 

[tedr1]

As I said I am no Nikon expert I wonder if KR is correct? And are the differences such as the OP is likely to find valuable in his work? I don't know and don't need to know.

 

[cooltouch]

When it comes to evaluating image sharpness based on various film ISOs, I agree the faster ones will show less information than slower ones, but these days the difference is not all that significant. I do think, however, where you'll find the biggest difference in film resolutions is a comparison between a slow slide emulsion, such as Fuji Provia with pretty much any negative film, with the possible exception of Portra 160 or Ektar 100.

 

[Montanawildlives]

When it comes to evaluating image sharpness based on various film ISOs, I agree the faster ones will show less information than slower ones, but these days the difference is not all that significant. I do think, however, where you'll find the biggest difference in film resolutions is a comparison between a slow slide emulsion, such as Fuji Provia with pretty much any negative film, with the possible exception of Portra 160 or Ektar 100.

Please excuse my stupidity but I've never used slide film. Are you saying that the slide film would show more resolution?

 

[cooltouch]

Yes. I've shot many, many slides over the years and the reason why is because I've always felt that I got better, sharper images from slide film. I still believe this to be the case, although as I mentioned, Portra and Ektar do get very close.

 

[John Koehrer]

A slide is the original image. Any print is second generation thing influenced by the printing process.

 

[cooltouch]

Yes, but when I'm evaluating slides vs. negatives, typically I'm examining dupes of each (dupes shot with my d*tal and a Nikon 55/2.8 macro lens). So there's no paper involved to soften things up.