Any technical explanations as to how or why film sharpness looks different from digital sharpness ??

But what about that “no body hair” criteria? I must be part Neanderthal, or part gorilla.

chuckroast said:

I believe the next cycle has begun with products like Topaz AI. They are using machine learning farms to augment the editing process to give the post production process the ability to make the print what is should have been - in focus, color corrected, and so forth. This combined with the mindless snapshottery of iPhones and other personal devices (Topaz isn't very expensive) will make digital photography increasingly the domain of AIs. Whether this is yet another branch off the tree of light or the destruction of human driven digital photography remains to be seen.

I'll still shoot film, thanks...

Click to expand...

I'll continue to shoot film too, because I love it. But once you've refined an approach to generative fill and other AI techniques (with digital/film/hybrid) it's easy to see the usefulness and utterly seamless improvements that can be made to images, no matter the capture medium. These are not "hey look at the shiny new buttons I can push" changes, but rather bedrock changes that sober photographers can use productively. These are amazing times for digital, film, and print media.

MurrayMinchin said:

Okay...prune away

Click to expand...

I can prune, but it is a lot better if you start the new thread, than if I start the new thread.
In order to move a post, I need a destination.

Steven Lee said:

If we are talking only about what's published online (because what else can we see in large numbers?), sharpening is rarely a problem. What we see are mostly downsampled images, where downsampling was done after sharpening, so the artifacts are simply not visible as much in relatively tiny online photos. What I see is the opposite: excessive sharpening of film scans! In fact, film is allergic to sharpening for reasons I won't go, but the artifacts of film scan sharpening are visible even in downsampled JPEGs published online, including this forum.

IMO, the overuse of sharpening is the chief reason why so many people think that Rodinal "increases grain", for example.

Click to expand...

The developer Rodinal is what is known as an acutance developer (Acutance = Sharpness) This was achieved pre war when Rodinal was first formulated. It depends upon high dilution of the stock chemical minimum of 1-25 and the optimum is 1-50 How this works is in conjunction with minimum agitation. Usually 10 inversions of the tank immediately after filling and one inversion every 30 seconds hereafter. What this does is create an 'edge effect' between areas of different tone because in darker areas of then negative the developer is exhausted more quickly than that in lighter areas.

This has two effects. The individual grains are made more prominent which is why the grain appears larger and the individual areas of tone become more obvious if looked at on a large print (no smaller than 12x16) where it will be seen that between the darker and lighter tones there is a very fine clearer line of emulsion where the developer has exhausted more quickly on the darker parts of the negative.. In fact, because this happened between INDIVIDUAL grains is the very reason the grain appears more obvious. It is also why in the original Agfa instructions, Rodinal was not recommended for films above 200iso.

It (Rodinal) is very definitely is not what is termed a 'fine gain developer' where the apparent lack of grain is due to a mild solvent effect on the grains of the emulsion and side by side comparison of two identical negatives, One developed in Rodinal and the other in a fine grain developer will show Rodinal has an APPARENT extra sharpness over the other

Some may argue about this but it was a question posed to all students undergoing the British City and Guilds examination back in 1968. This was the answer I gave and was marked correct. Anyone who does not agree with this explanation, I suggest do not know what they are talking about.

There was one very similar developer which I used out of preference despite it's higher cost and no sadly no longer produced. It had the same effects as Rodinal but made by Ilford and called Hyfin. It came as dry sealed packets which were dissolved in water and had to be used more of less immediately because it started to 'decay' within an hour to an hour and a half. It was withdrawn from sale by Ilford sometime in the early 1970's and despite being asked by myself, they will not release the formula. If anything, it gave even a little more of a sharpening effect. They, Ilford only gave developing times for 2 films FP3 and Pan F which was 18 mins 68F for both, with even less agitation than Rodinal. 10 inversions at the start and one inversion every minute

BMbikerider said:

The developer Rodinal is what is known as an acutance developer (Acutance = Sharpness) This was achieved pre war when Rodinal was first formulated. It depends upon high dilution of the stock chemical minimum of 1-25 and the optimum is 1-50 How this works is in conjunction with minimum agitation. Usually 10 inversions of the tank immediately after filling and one inversion every 30 seconds hereafter. What this does is create an 'edge effect' between areas of different tone because in darker areas of then negative the developer is exhausted more quickly than that in lighter areas.

This has two effects. The individual grains are made more prominent which is why the grain appears larger and the individual areas of tone become more obvious if looked at on a large print (no smaller than 12x16) where it will be seen that between the darker and lighter tones there is a very fine clearer line of emulsion where the developer has exhausted more quickly on the darker parts of the negative.. In fact, because this happened between INDIVIDUAL grains is the very reason the grain appears more obvious. It is also why in the original Agfa instructions, Rodinal was not recommended for films above 200iso.

It (Rodinal) is very definitely is not what is termed a 'fine gain developer' where the apparent lack of grain is due to a mild solvent effect on the grains of the emulsion and side by side comparison of two identical negatives, One developed in Rodinal and the other in a fine grain developer will show Rodinal has an APPARENT extra sharpness over the other

Some may argue about this but it was a question posed to all students undergoing the British City and Guilds examination back in 1968. This was the answer I gave and was marked correct. Anyone who does not agree with this explanation, I suggest do not know what they are talking about.

Click to expand...

I have a achieved both high acutance/edge effects with dilute Pyrocat-HD agitating only 3 times over an hour of stand time - Extreme Minimal Agitation. So fine apparent grain an edge effects absolutely can coexist. See Post #31 here.

MattKing said:

I can prune, but it is a lot better if you start the new thread, than if I start the new thread.
In order to move a post, I need a destination.

Click to expand...

Thanks, no worries.

It was just a conversational tangent, so can disappear.

@BMbikerider no need to burst into spontaneous lectures on developers. Everyone knows what Rodinal is but very few have the equipment and skill to scan it without digitally enlarging the grain. That is the point of my comment above: sharpening is a far bigger problem for film users than digital users.

And if you want me to be more direct, I can: DSLR users on average have better digital image editing habits and skills than film photographers who scan. Therefore, pointing at excessive sharpening as "digital problem" is unfair. It's the other way around. You want proof? Go and explore thousands of botched/over-sharpened scans in the media section.

And if you dismiss the over-sharpening problem (and I think we should), then the answer to the OP's question has been posted above: film resolution depends on subject contrast and exposure to a greater degree than digital. Discussing people's habits is besides the point. Case closed.

what is your definition of sharpness or what are you lookingfor?

chuckroast said:

I have a achieved both high acutance/edge effects with dilute Pyrocat-HD agitating only 3 times over an hour of stand time - Extreme Minimal Agitation. So fine apparent grain an edge effects absolutely can coexist. See Post #31 here.

Click to expand...

We are talking about Rodinal which is a very old developer, I believe the oldest in the world still in production. (My research indicates it was first sold in the 1920's in Germany. In that time span there will be other developers formulated with will mimic Rodinal but that is the original and I would say, still possibly the best acutance developer still available. It is a very much a personal opinion and I will stand by it unless I see something better. The very limited agitation with Pyrocat-HD will probably lead to the edge effects.

I had an article published in a UK magazine around the middle of the 1980's (For which I was quite well paid for) which was illustrated with pictures that clearly demonstrated the edge effect between two areas of differing tones.

Anon Ymous said:

Have you ever heard of unsharp masking? Do you know it's origins?

Click to expand...

Unsharp-masking is an analog technique, working pretty much the same as the digital method of the same name

Steven Lee said:

@BMbikerider no need to burst into spontaneous lectures on developers. Everyone knows what Rodinal is but very few have the equipment and skill to scan it without digitally enlarging the grain. That is the point of my comment above: sharpening is a far bigger problem for film users than digital users.

And if you want me to be more direct, I can: DSLR users on average have better digital image editing habits and skills than film photographers who scan. Therefore, pointing at excessive sharpening as "digital problem" is unfair. It's the other way around. You want proof? Go and explore thousands of botched/over-sharpened scans in the media section.

And if you dismiss the over-sharpening problem (and I think we should), then the answer to the OP's question has been posted above: film resolution depends on subject contrast and exposure to a greater degree than digital. Discussing people's habits is besides the point. Case closed.

Click to expand...

I don't 'burst' as you put it into 'spontaneous lectures' on anything, but it was apparent to me that the difference between artificially sharpening a digital image either with 3rd party software, or in the camera and the sharpening effect produced by acutance type developers is/was not fully understood.

I do not digitally scan a lot of anything either. I mostly print in the darkroom. However even quite basic scanners have a facility to reduce grain. How well they work I have no Idea, I have never tried it. The main scanning of any negative colour or B&W I do, is to make a low-res digital reference sheet of images in thumbnail size where sharpening is not really required.

Rudeofus said:

You can tell your enlarger to render the film section with that one grain "medium gray", but then you'd need a large area with many grains to create another area which is just 1% brighter. This is where film resolution completely falls apart. If you want to render two patches with an 1% translucency difference, you need at least 100 grains in each patch.

Click to expand...

No you don’t. Grain is not binary. You get a grain with 1% more or less filament structure.

RalphLambrecht said:

what is your definition of sharpness or what are you lookingfor?

Click to expand...

Sharpness is usually used for perceived contrast. The transfer of contrast is a property of the lens (MTF). In analog photography the focal length of the lenses are much larger than in digital photography because of the difference in film size and sensor size. The consequence is the differences in the properties of the lenses. One of those properties is the quality of the ' sharpness and unsharpness'. How is the transition of the sharpness into the unsharpness? In digital photography the focal length is short, and as a result the optical properties cannot compare. I personally prefer a 150 mm lens Apo sironar S 150/5,6. I use it on my 4x5". A 150 mm Sonnar on my Rollei, a 135 mm Xenar on my Leica.

It is not just the lens that makes a difference. The film, the film developer are factors too. And then the difference in informatioin content. The pixel amount is so small compared to the 'pixel equivalent' of the human eye. In analog the quality of the human eye can be obtained, using the proper equipment. In addition the behaviour of light in a baryta paper cannot be compared with that on a digital print.

Jed

Helge said:

No you don’t. Grain is not binary. You get a grain with 1% more or less filament structure.

Click to expand...

In this case this either happens due to crystalline/physical/whatever differences between two grains, i.e. noise, or because a different size/number of develop centers was formed. So you may end up with 11 instead of 10 silver atoms forming a development center and indeed the larger one will likely develop to a slightly larger size. Latent silver specks with 100 atoms would only happen in very strong exposures. Therefore, if you want to correctly represent an 1% difference in luminosity, then you still need 10-20 silver grains to make this happen.

BMbikerider said:

A pure non technical opinion along the lines of 'Beauty is in the eye of the beholder' may be the reason.
Resolution or sharpness with film is due to the lens, tripod and how sharp film ran resolve the image taking into consideration contrast and colour.

Resolution or sharpness with digital sensor is down to the lens, tripod and the quality of the sensor and any in- camera sharpening done by the camera or by 3rd party software.

The main problem being a lot of images from a digi are over-sharpened. (My opinion)

Click to expand...

In discussions like these you quickly run into platitudes like the one above or “to each his own” or “you can’t discuss art”.
Those resolve nothing, satisfies no ones curiosity and are simply not true.
Even if there is not exact science pertaining the subject (there rarely is in anthropology and humanities) that doesn’t mean that there are not general rules of the senses and a lot of research to back them up.
The field of psycho optics is huge.

There is the famous Pepsi Challenge conundrum.
People clearly preferred sips of Pepsi to Coke. But overwhelmingly chose the latter when buying cola.
The human senses like to be rubbed with the grain and “helped” in small doses.
That includes edge enhancement, lifting of and even detached treble in speakers for extra “clarity”, and enhanced flavor and more sugar that taps right into ancient survival instincts of “stock up while it’s there”.
But in larger doses it gets cloying and too much.

Trouble is, when we collective agree something is “better” for the above reasons in a moment of global insanity, the common person really doesn’t have the vocabulary to argue against that later on.

Soon a new normal is created and anything else is going to feel subtly weird and off, instead of being a valuable and worthy alternative.
And even perhaps really overall better.

Rudeofus said:

In this case this either happens due to crystalline/physical/whatever differences between two grains, i.e. noise, or because a different size/number of develop centers was formed. So you may end up with 11 instead of 10 silver atoms forming a development center and indeed the larger one will likely develop to a slightly larger size. Latent silver specks with 100 atoms would only happen in very strong exposures. Therefore, if you want to correctly represent an 1% difference in luminosity, then you still need 10-20 silver grains to make this happen.

Click to expand...

Noise is always a factor. Either substrate noise or readout and quantization noise.
And noise is almost always worse at the extremes of the range.
That goes for CMOS, CCD, film even painting if you get philosophical. And certainly for audio recording of any kind.
A single crystal can in theory hold the whole range. Whether it does it, is another matter.
Same way the capacitance of a cell in a sensor is affected by all kinds of noise and nonlinearities too, that is invisibly corrected and compensated for in-camera.

But again, silver halide crystals are stacked, offset to different degrees and of different sizes and shapes through the stack.

Helge said:

A single crystal can in theory hold the whole range. Whether it does it, is another matter.
Same way the capacitance of a cell in a sensor is affected by all kinds of noise and nonlinearities too, that is invisibly corrected and compensated for in-camera.

But again, silver halide crystals are stacked, offset to different degrees and of different sizes and shapes through the stack.

Click to expand...

I found two images on petapixel, which seem to support my claim of "subject matter contrast determines film resolution". In this article they make the umpteenth comparison between analog and digital and (among others) compare image resolution from a Mamiya 7 against the output of a PhaseOne IQ180.

If you look at the comparison between IQ180 and Velvia 50, you see very comparable resolution in the "Nikon" label (high contrast), but more or less mush in the Velvia 50 sections with low contrast.

If you look at the comparison between IQ180 and CMS 20, the low contrast areas look very similar, but the CMS 20 "Nikon" label looks a lot crisper.

This is not some strange miracle, but instead follows directly from the requirement from multiple grains to render a low contrast transition.

That linked article sounds about as authoritative as a political poll based on only two people sitting at the bar in a bowling alley. It's more article "filler" or fodder than anything else.

DREW WILEY said:

That linked article sounds about as authoritative as a political poll based on only two people sitting at the bar in a bowling alley. It's more article "filler" or fodder than anything else.

Click to expand...

I linked the article only to provide a proper attribution for these two images. These two images look more than "bowling alley talk" to me: they actually prove my point. Everyone is welcome to prove the opposite - with images please.

DREW WILEY said:

That linked article sounds about as authoritative as a political poll based on only two people sitting at the bar in a bowling alley. It's more article "filler" or fodder than anything else.

Click to expand...

Sometimes you are just too kind with your words.

Rudeofus said:

I found two images on petapixel, which seem to support my claim of "subject matter contrast determines film resolution". In this article they make the umpteenth comparison between analog and digital and (among others) compare image resolution from a Mamiya 7 against the output of a PhaseOne IQ180.

If you look at the comparison between IQ180 and Velvia 50, you see very comparable resolution in the "Nikon" label (high contrast), but more or less mush in the Velvia 50 sections with low contrast.

If you look at the comparison between IQ180 and CMS 20, the low contrast areas look very similar, but the CMS 20 "Nikon" label looks a lot crisper.

This is not some strange miracle, but instead follows directly from the requirement from multiple grains to render a low contrast transition.

Click to expand...

I believe all you have done is to show that the results for CMS 20 are limited by the scanner.
The article quotes 50 mpx for 645, 54x 41.5 = 2324 square mm. For 35mm, 36x 24 = 864 square mm.
So for a 35mm scan of CMS 20 that would be 50 x 864/2324 =18.6 mpx, a bit more taking the value 80 mpx for 645 also quoted.
This is pretty small compared to the true resolution provided by CMS 20.

It astonishes me that anyone would would rely on a Petapixel article for anything other than funny anecdotes from wedding photographers and fishing with a GoPro.

Alan Johnson said:

I believe all you have done is to show that the results for CMS 20 are limited by the scanner.
The article quotes 50 mpx for 645, 54x 41.5 = 2324 square mm. For 35mm, 36x 24 = 864 square mm.
So for a 35mm scan of CMS 20 that would be 50 x 864/2324 =18.6 mpx, a bit more taking the value 80 mpx for 645 also quoted.
This is pretty small compared to the true resolution provided by CMS 20.

Click to expand...

The Mamiya 7 is a 6x7 camera.

The article quotes 50 to 80 mpx for a Mamiya 645 scan , for the Mamiya 7 it quotes about 150 mpx.

faberryman said:

It astonishes me that anyone would would rely on a Petapixel article for anything other than funny anecdotes from wedding photographers and fishing with a GoPro.

Click to expand...

What's so "astonishing" that anyone would be interested in what this guy wrote?

About the author: Tim Parkin is the editor of a dedicated landscape photography magazine On Landscape and also the CEO of a drum scanning business.

Still not good enough for you? Do you expect the creator himself to chime in?

Alan Johnson said:

The article quotes 50 to 80 mpx for a Mamiya 645 scan , for the Mamiya 7 it quotes about 150 mpx.

Click to expand...

Given the author's credentials I should probably trust his numbers more than my guestimates about this. Since analog resolution has a strong dependency on recorded subject matter contrast, there is no such thing as a pixel value equivalence. These insanely high numbers are commonly thrown out in meaningless comparisons.

A few more things about sharpness and resolution: an F/4 lens turns a tiny point light source into a diffraction pattern of roughly 5 µm. A diffraction pattern diameter of 5 µm corresponds to 100 lp/mm. The lens stopped down to F/8 as used in the test creates twice the diameter. This is physics, so no film or sensor can do anything about this, but digital image "improvents" often try to do it anyway. That's where many people go "my Olympus with this and that lens creates 20MP with pixel sharp results!" and we all know, that this is nonsense.

Oh, and there is DoF, which is abysmally low, if you aim for 5µm resolution. These nice Zeiss formulas we all love to use in Dofmaster are for much larger CoC.

In the end there are only two ways to get huge resolutions:

1. Take the sharpest tele lens you can find wide open, and shoot a very flat wall perfectly perpendicular to your lens axis at moderately close range. Mirror lockup (where applicable) and very sturdy tripod don't need mentioning. Don't get too far away from the subject matter either, or diffusion due to air moisture gets in the way.
2. Take any lens on some digital camera with ridiculously small pixel pitch, shoot any subject matter at any distance and let some smart image processor immerse you in a silly illusion about high res results.

The first method will not give you results actually worth looking at, and the actual resolution you get with the second method will not really depend on the recording medium.