Full resolution of film!

Helge · Sep 1, 2023

chuckroast said:
Better tell that to everyone you ever see that says "I'm pushing my FP4+ to ISO 1000". It is a widely held belief by many photographers that overdevelopment increase effective film speed dramatically. I'm glad you know that's not true, but in my experience, at least, this is not widely understood.

Developer like all other analog amplifiers is nonlinear. It works much more on strong signals than weak. Which is of course the exact opposite of what we want.
As long as the amplification is not too much we can get away with it, with build in inhibitors and various development techniques. But once we push it (pun intended) we get a steeper curve.

You still get more development of the lower part of the curve though, to a point.

What people seem completely oblivious to, is the option of preflashing and latensification which lifts the bottom up again, with the added effort of doing that before exposure or before development.
That, in combination with pushing is very real upping of the speed.

DREW WILEY said:
As far as resolution goes, it's all back to, You're only as good as your weakest link, whether that be the lens, the film, stability, development choice, etc etc etc. So of course, if a film is inferior to the task, blame that; if the lens is inadequate, blame it instead, or more realistically blame all kinds of interactive factors, including using too small a film format to begin with.

135 is superb with the right amount of care and slow film. What most people miss is the last step. Either scanning or printing. That is where you need to put at least as much effort as in the recording.

alanrockwood · Sep 1, 2023

Mr Bill said:
I hear you but I don't understand your thinking.

Perhaps you would be so kind as to make an opinion about the example I gave: if a lens, alone, can deliver max 200 lpmm and a film , alone, can deliver a max 200 lpm, what would you say is the max resolution that the combination can produce? Do you think it would still be 200 lpmm? Or something less?

Not 200, but something more like 140.

It is actually an oversimplification to say that resolution is set by the weakest link in the chain. What it really means is that resolution can never be as high as the weakest link in the chain. It is always lower. If there are two links in the chain of equal resolution, and if the resolution-degrading mechanisms are independent, then the resolution is significantly less than one of the links alone. The formula I gave earlier, which is based on rigorously correct mathematics, predicts that in the case of the example you gave the best you can hope for is 141.2, and that's if there are no other resolution degrading mechanism.

I need to repeat that this applies to the case when the mechanisms that degrade resolution are (statistically-speaking) independent. If they are not independent then resolution could be better or worse than the predicted amount, but as a practical matter this would be an unusual case.

Edit added later: For clarity purposes I should add that the squared equation in post 199 is the same as the equation I advocated in post 151.

Mr Bill · Sep 1, 2023

faberryman said:
I don't which, if either, of the formulas you set forth in post #199 are correct, so I'll refrain from speculating.

Well thanks for the response anyway. My curiosity has been somewhat sated.

Fwiw I don't think that either is strictly "correct." Rather they are given as empirical methods where sometimes one seems to be a better predictor than the other.

Nikon 2 · Sep 1, 2023

chuckroast said:
Only if your lenses are well corrected

I kid, I kid, but it IS interesting how - even the best manufacturers - have some vintages of lenses that have, shall we say, "other" issues. I recently bought a Nikkor Ai-S 35mm f/1.4 that needed a bit of work (which I supplied). I wanted the extra speed to replace an Ai 35mm f/2.8. Well, imagine my surprise to discover that the Ai-S has terrible coma problems at 1.4. It's quite usable and even has a sort of "veiling effect" that may be nice for portraits, but it's clearly not well "corrected". Down range at about f/11, the lens is just tack sharp, though. By comparison, the 85mm f/1.4 Ai-S is a razor blade.

Don't even get me started on my pre-war 50mm f/3.5 collapsible uncoated Elmar. It's super sharp but when faced with specular highlights or very bright light sources, the lack of coating gives it an ... interesting look. Again, this can be useful aesthetically for a kind of dream like look.

Yet more stuff that impacts image quality and perceived sharpness, I guess

Very impactful...!

faberryman · Sep 1, 2023

Mr Bill said:
Well thanks for the response anyway. My curiosity has been somewhat sated.

Fwiw I don't think that either is strictly "correct." Rather they are given as empirical methods where sometimes one seems to be a better predictor than the other.

The first formula was offered by Leidolf in post #146 as applicable for determining total resistance in a series of resistors in electrical circuits. The second formula was offered by alanrockwood in post #151 as applicable in statistics and signal processing. I don't know if either is appropriate for determining resolution in a system comprised of an optical lens and photographic film.

Alan Johnson · Sep 1, 2023

George Mann said:
This mystery film is 3.5 times the resolution of the highest commercially available film that I know of, which is ADOX CMS 20 II.

I believe CMS 20 II has a resolution of 800 lppm with a 1000:1 contrast target.

The best things in life are analog | CMS 20 II & ADOTECH IV

www.adox.de

Mr Bill · Sep 1, 2023

alanrockwood said:
Not 200, but something more like 140.

alanrockwood said:
If there are two links in the chain of equal resolution, and if the resolution-degrading mechanisms are independent, then the resolution is significantly less than one of the links alone. The formula I gave earlier, which is based on rigorously correct mathematics, predicts that in the case of the example you gave the best you can hope for is 141.2, and that's if there are no other resolution degrading mechanism.

Thanks Alan.

I have the sense that a lot of photographers would see it as more akin to a physical chain, where if one link could withstand 200 "units" of force then a series of identical links would maintain the same overall capacity. That is, that the entire chain can still withstand the same force. (Which is clearly not the case for a series of stages in a reproduction system.)

Nikon 2 · Sep 1, 2023

Alan Johnson said:
I believe CMS 20 II has a resolution of 800 lppm with a 1000:1 contrast target.

The best things in life are analog | CMS 20 II & ADOTECH IV

www.adox.de

That's amazing. What`s more amazing is there`s no lens capable of resolving that film...!

chuckroast · Sep 1, 2023

Leidolf said:
Nope. save for old pre WW1 box cameras FILM is always the limiting factor. Especially colored films....
And deducing the resultant resolution is much like deducing the resultant sesistance in an electrical circuit, when adding resistors in a serial coupling :
1/N1 + 1/N2 + 1/N3 etc etc

I believe you mean resistors in parallel. Resistors in series simply have their resitance added: R1+R2+R3 ... for computing total resistance.

chuckroast · Sep 1, 2023

Nikon 2 said:
That's amazing. What`s more amazing is there`s no lens capable of resolving that film...!

No lens you know about. Nowithstanding their current used of digital sensors, the Five Eyes likely have optics that only an unlimited sum of (our) money can buy...

Sirius Glass · Sep 1, 2023

alanrockwood said:
I apologize. I relied on memory in making my post. I offer a correction below.

First of all, I found the article. It is in Modern Photography, October, 1978. The article starts on p. 108. The title of the article is "How Sharp Can You Get?". The authors were Bennett Sherman and Al Gordon.

They did not achieve 100 LPM using Panatomic X. With Panatomic X the best results were at infinity focus with the apertures set at f/4 for six different lenses. The results and resolution in lines/mm were as follows:

Sumicron 50mm f/2 lens: 88 at f/4
Nikor 50mm f/1.8 lens: 88 at f/4
Canon 50mm f/1.8 lens: 86 at f/4
Minolta 50mm f/1.7 lens: 86 at f/4
Pentax 50mm f/1.7 lens: 86 at f/4
Olympus 50mm f/1.8 lens: 86 at f/4

With Tech Pan film focused at infinity the results were a little better for the same lenses, ranging from 92 lines/mm to 96 lines/mm.

High contrast Copy film did a little better, reaching 102 to 106 lines/mm.

For Kodachrome II the results ranged from 80 to 86 lines/mm.

For Micro-Ektachrome the results ranged from 100 to 102.

They also tested some close focusing and macro lenses under under close up conditions. I couldn't quite figure out exactly what those close up conditions were. However, the results were somewhat less sharp than the tests of the 50mm prime lenses at infinity, roughly 10% lower resolution.

That same issue of Modern Photography also had an article addressing that age old question "Can you really "push" film???" I won't comment on that, except to say that their conclusion could be summarized with the somewhat vague characterization of "yes, sort of." (Not a quote from their article, but my condensed version of how I interpret their results.)

My 50mm lens is better than your 50mm lens!

Duh! They are all about the same.

Mr Bill · Sep 1, 2023

Nikon 2 said:
That's amazing. What`s more amazing is there`s no lens capable of resolving that film...!

Sure there are. That is, lenses capable of resolving 800 lpmm. Fundamentally they will be limited by the size of the lens aperture with respect to an f#. A rule of thumb in photography is that a certain number, roughly 1800 (?) divided by the f-stop number gives the maximum lpmm that the lens can produce. It's actually based on the smallest point of light the lens can make, aka the Airy disc. (If you can stretch that "point" out into a line, then space such lines at the limit of resolution this would give lpmm.) Anyway, an ideal lens with an aperture wider than f/2 will not be so limited.

As a note the base number of 1800 ( or so) is, I believe, based on the wavelength of "green" light. Bluish light has a shorter wavelength so would not be limited quite so much.

Nikon 2 · Sep 1, 2023

Sirius Glass said:
My 50mm lens is better than your 50mm lens!

Duh! They are all about the same.

You have a point, Tack sharp...!

Nikon 2 · Sep 1, 2023

Mr Bill said:
Sure there are. That is, lenses capable of resolving 800 lpmm. Fundamentally they will be limited by the size of the lens aperture with respect to an f#. A rule of thumb in photography is that a certain number, roughly 1800 (?) divided by the f-stop number gives the maximum lpmm that the lens can produce. It's actually based on the smallest point of light the lens can make, aka the Airy disc. (If you can stretch that "point" out into a line, then space such lines at the limit of resolution this would give lpmm.) Anyway, an ideal lens with an aperture wider than f/2 will not be so limited.

As a note the base number of 1800 ( or so) is, I believe, based on the wavelength of "green" light. Bluish light has a shorter wavelength so would not be limited quite so much.

Can you name a few...?

Nikon 2 · Sep 1, 2023

chuckroast said:
No lens you know about. Nowithstanding their current used of digital sensors, the Five Eyes likely have optics that only an unlimited sum of (our) money can buy...

Which do you know about that resolve that film...?

George Mann · Sep 1, 2023

Alan Johnson said:
I believe CMS 20 II has a resolution of 800 lppm with a 1000:1 contrast target.

The best things in life are analog | CMS 20 II & ADOTECH IV

www.adox.de

Not in the real world.

Nikon 2 · Sep 1, 2023

George Mann said:
Not in the real world.

The site also mentions it having to have a special developing technique and chemicals to achieve its resolution...!

Alan Johnson · Sep 1, 2023

George Mann said:
Not in the real world.

Agreed.
240-260 lppm for CMS 20 II at 4:1 contrast as per Henning Serger linked in post 11 is more like it.

George Mann · Sep 1, 2023

Nikon 2 said:
The site also mentions it having to have a special developing technique and chemicals to achieve its resolution...!

You can only achieve the maximum attainable resolution by utilizing a carefully contrived laboratory test environment.

But most of these extremely high quotes are theoretical, none of which is remotely achievable thru real world photography.

Mr Bill · Sep 1, 2023

Nikon 2 said:
Can you name a few...?

Not offhand, but I would imagine that just about any f/1.4 cine lens (a decent one) would do it. Almost for sure just about any good quality microscope lens (with moderately high N.A.) would do it.

I'd GUESS that a lot of higher end lenses for 35mm cameras would also do so in the center. But they gotta be faster than f/2. If they don't do it, as is, narrowing down the wavelength range with a sharp blue or green gel filter would probably do it.

Can I prove it? Not offhand but I'd probably be willing to make a pretty healthy bet I could come up with some lenses but it might take some doing to prove they do. Offhand I'd say that perhaps one could examine the aerial image with a microscope, but it would also need an aperture as fast (or faster) than the lens.

Fwiw you asked chuckroast "Which do you know about that resolve that film...?". With all due respect this tells me that you are not understanding the issues of interaction between lens and film. With an equally capable lens, meaning roughly an ideal (in the center) lens of about f/2, and a nominal 800 lpmm film, the expected result will be roughly somewhere between 400 and 560 lpmm. Not the 800 lpmm the film is said to have.

Let me ask you, just offhand, how would you quantify the resolving power of that film? Let's say I'm skeptical about those numbers. How would you go about proving what it can do?

Let me say one last thing about measuring resolution on film, using something like the standard 3-bar black & white test charts. When I was a young pup fairly fresh in photo lab QC I thought it was necessary to have a clear delineation of 3 bars. But... we had some ANSI publications on how to read the charts. The official procedure, when examining test film under a microscope, was to find the smallest bar set where you thought it was just as likely as not that there were 3 bars. That is, about a 50% confidence that there were 3 bars. So I'd guess that most photographers testing film are gonna underrate things.

chuckroast · Sep 1, 2023

Nikon 2 said:
Which do you know about that resolve that film...?

I don't know about any of them either. Although I have been called Four Eyes in the past, I've never been privy to the inner workings of the Five Eyes. So, I'm just guessin' here ...

Here, Let Me Google That For You

Passive-aggressively teach your friends how to Google. For all those people who find it more convenient to ask you rather than search it themselves. Not associated with Google.

googlethatforyou.com

Bill Burk · Sep 1, 2023

Found some light easy reading on the topic of lens and film resolution…

It says Minox (8x11mm) lenses have great resolution about 163 lp/mm.

Most great 35mm camera lenses are around 100 lp/mm.

And my favorite high resolution film Panatomic-X has around 170 lp/mm.

https://cool.culturalheritage.org/videopreservation/library/estimating_historic_image_resolution_v9.pdf

Nikon 2 · Sep 1, 2023

chuckroast said:
I don't know about any of them either. Although I have been called Four Eyes in the past, I've never been privy to the inner workings of the Five Eyes. So, I'm just guessin' here ...

Here, Let Me Google That For You

Passive-aggressively teach your friends how to Google. For all those people who find it more convenient to ask you rather than search it themselves. Not associated with Google.

googlethatforyou.com

Nikon 2 · Sep 1, 2023

Me too…!

Nikon 2 · Sep 1, 2023

Bill Burk said:
Found some light easy reading on the topic of lens and film resolution…

It says Minox (8x11mm) lenses have great resolution about 163 lp/mm.

Most great 35mm camera lenses are around 100 lp/mm.

And my favorite high resolution film Panatomic-X has around 170 lp/mm.

https://cool.culturalheritage.org/v...imating_historic_image_resolution_v9.pdf[/URL

Very informative, light and easy…!

Full resolution of film!

Privacy & Transparency

Privacy & Transparency