Low frequency spatial resolution?

[Denton]

I have a tendency to ignore the low spatial resolution lines in MTF charts when comparing two lenses. But perhaps I might find more personal appeal in an image taken with a lens having poorer high frequency resolution, but "superior" low frequency resolution? What would be the qualitative differences?
Denton

 

[Rob-F]

The low frequency graph (top line) refers to the reproduction of major subject outlines or shapes. Then the next higher frequency line is for the coarse details, and the highest frequency (lowest line) is for the finest ones.

 

[Denton]

But how does the print REALLY differ?

But how does the print REALLY differ?

It seems surprising that the 40 lp/mm would have much to do with the sharpness of a print, being well beyond the perception of the eye, yet most of us likely evaluate a lens based on this metric. So the question is, what is the qualitative difference. A large shape of several mm should not be affected. I suspect the difference is in the edge effects, maybe digital accutance or close to unsharp masking.
I read graphs a great deal in my scientific work, and my experience is that there can be many ways to interpret such data, not always accurately.

Any lens designers reading?

 

[PaulDalex]

Denton,

your remark is a very smart one.

For misterious reasons that you should not ask to me, a certain mathematical topic is adopted in various technical fields and is completely twisted.

Actually it is much more meaningful to represent the transfer function as a function of frequency at each point and for the two orientations.

In the acoustic field they correctly represent the transfer function as a function of frequency.

In optics this does not happen and it is a nonsense.

And as you say (and you are completely right) one should ideally have a flat diagram with a decay at higher frequency (the higher the better).

If that is not so, it indicates a bad reproduction (low fidelity).

But how could one possibly obtain the diagram from 3 only frequencies?

It is a total nonsense

Regards

Paolo

 

[Rob-F]

You may be giving too much importance to resolution. The MTF is about percent of contrast. Of course, while resolution is a function of local contrast--contrast across a boundary--it is the contrast that determines our subjective impression of sharpness. Leica caught on to this in the early '70's (or late '60's), and began designing for contrast first, and worrying a bit less about resolution.

The MTF is a record of the lens's transfer function: contrast out over contrast in, times 100 percent. So a perfect lens would graph out as a flat line across the top of the graph, no loss in contrast at any aperture or any spatial frequency, from the center to the corners. No one has yet made such a lens that I know of; but if they did, I don't think everyone would like it. When people complain about a lens, the complaint may not be about a lack of sharpness. It might instead be that the lens has too much contrast. Sometimes this is said about some CV lenses, or the Leica asphericals. Some find the picture "harsh" in appearance. I think that's why folks still look for some of the older German lenses, like a 50mm Summarit or a collapsible Summicron. I've read that certain directors wants their films shot with older lenses, finding the current state of the art ones too sharp--even when blow up to movie screen size. Traditional spherical lenses get praise for having "3-D pop" or the picture has "glow" or it's "liquid."

A frequent comment here is that high contrast is great for color, but too much contrast is bad for black & white work. A lower contrast lens might have more flare, reflecting a little stray light into dark areas, giving them an impression of volume rather than empty blackness.

So I think the MTF may be telling us as much or more about how contrast affects the overall look--harsh vs. smooth--vs. how resolution affects that look.

I guess it's a bit analogous to sound reproduction. One listener may prefer the detailed sound of a solid state amplifier, while another finds that a tube amp sounds more like real music.