Low contrast, high resolution - I'm confused!

[jon_noble]

Allen

 

Ok, maybe i shouldnt have generalised my statement. granted a lot of nikkors are junk but what should say instead of "nikkors" is "the best nikkors".

 

But i must state i am not sure on this. These are just my impressions from reading a lot of second-hand opinions which count for nothing. Like mine for example

[Allen Herbert]

"resolution" is a characteristic of a lens (optical system),

 

Yes,Bill, after doing a bit of research i think the above statement holds true.

[Rob F.]

Well, contrast is also an attribute of the optical system, in that the lens has a contrast transfer function. If the lens did not degrade contrast at all, then the transfer function would be 100%. In other words, what comes out is the same as what went in. A real-world lens will degrade contrast to some extent. What comes out might still be 98% of what went in, or it might be only 10%.

 

Now, this is not to say that subjects don't differ in contrast. But no matter what the contrast of the original subject--high or low--the image coming out the back end of the lens will always be lower in contrast than it was when it went in the front end. It's just a question of how much lower. A perfect lens would preserve all the contrast that was there to begin with. A 21mm ASPH (for instance)will come closer to that goal than a 21mm SA.

 

This has been an interesting thread that shows that this subject is not as simple as it appears. I've been following with interest, without having much to contribute, except for the above. Like others, I'm puzzling over whether it makes sense for a lens to be high-res but low-contrast. I think it must be a matter of degree. My 50mm DRS is very sharp but only has medium contrast. But I don't think a lens could be extremely low in contrast, yet extremly sharp. There has to be a contrast limit, at some point, below which resolution breaks down (I would think).

 

Gotta go. I may have more thoughts later.

[Rob F.]

Possibly the method of "reduction to an absurdity" would be helpful. It seems clear that a lens with 100% preservation of contrast can be capable of high resolution/sharpness. It also seems clear that a lens that has no contrast at all--renders everything as gray--would not be capable of any sharpness at all, since it can't make an image. Those are the extremes. In between the extremes, it would then follow that sharpness would co-vary with contrast.

 

This is, however, armchair analysis. Again, it's a challenging topic. I wouldn't mind seeing more discussion.

[vivek iyer]

Contrast is more of a subject for a given lighting condition.

 

Change the lighting and the contrast varies.

[Rob F.]

Yes, Vivek, that's true as far as it goes. But for any given subject contrast, the lens still has its contrast transfer function. A 50mm Summicron, current version, will deliver a higher contrast image of any given scene, than a collapsible Summicron will render of the same scene. Try the experiment with a different scene, and although the contrast level may change with the scene, the current version Cron will still come out on top, with higher contrast compared to the collapsible.

 

It's a little like comparing brightness with and without your sunglasses. Brightness depends on whether you are indoors or out; whether cloudy or bright, whether day or night. BUT! Whether indoors or out, cloudy or bright, day or night, if you take off your sunglasses, things will get brighter for you. Your sunglasses have a brightness transfer function that operates in every circumstance.

 

Now, brightness isn't the same as contrast. This is an analogy. But all lenses have a contrast transfer function. That is in fact what an MTF graph is showing you. And that transfer function operates on everything you point it at, whether the original scene is more contrasty to begin with, or less so.

 

Comments?

[vivek iyer]

Rob, Quite a few of my scanned images are locked up in another hard drive so, I can't post any examples at the moment.

 

I have experimented with various 'high resolution' lenses (mostly in 1/5X to 5X magnification range).

 

Some of the lenses with lower contrast (diffused lighting) yield perky images under directed, axial or pseudo axial lighting, for example.

 

The veteran p.netter, Luis Triguez (has a language problem with English so his expressions in words are limited), "knows" what lens to use for a particular lighting condition. I am sure a lot of the experienced photogs "know it" from their experience.

[Allen Herbert]

Hmm, my old dad, used to have his lenses tested by a Company called Wallace Heaton in London. He was into serious giant enlargements for London West End Cinemas etc.

 

The same lens, by the same Company, often had different resolution so i was told......hence; all his new lenses were tested.

 

So, it�s about the individual character of the lens tested. However, i must admit contrast to the eye looks sharper.

[Rob F.]

"However, i must admit contrast to the eye looks sharper."

 

And that, I think, is what is missing from our discussion. I'm not sure we've properly addressed the difference between sharpness and resolution. In my mind, sharpness is the subjective appearance of sharpness--apparent sharpness--as opposed to resolving power, which the ability to separate closely spaced points. I feel certain that apparent sharpness depends on contrast. I've posted images here in the past that show that a lens of lower contrast (the DRS) can have excellent resolving power.

[vivek iyer]

<i>I feel certain that apparent sharpness depends on contrast.</i>

<p>

Agreed. Also the edge "sharpness" enhances the apparent sharpness.

[david j.lee]

and there's where acutance come to play Vivek.

"acutance is the contrast of the edges between distinct tones in the

print" ( from "edge of darkness" by the late Barry Thornton).

[david j.lee]

Barry Thornton also used a chart similar to the one Bob posted, but instead of

low resolution/high resolution he used low/high acutance. in Bob's chart you

can still count the pairs of lines, that means that the resolution is the same for

both examples, it's the acutance that changes.

[vivek iyer]

David, Thanks for injecting that word! Great discussion!

[peter]

""However, i must admit contrast to the eye looks sharper."

 

This is the point I was making in my original post