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

[Nikon 2]

Unsurprisingly, lenses for national security and scientific applications are better than those available to photographers.

Leica might take umbrage to that…!

Unsurprisingly, some developers provide more acutance than others, so pick the one most appropriate for the look you are trying to achieve. None of this is rocket science.

 

[Helge]

• Deleted
• 38 minutes ago
• Reason: Due to my ignorance and apparent stupidity, I posted the wrong data and illustration...

Don’t be so hard on yourself. You are in good company with that. A lot of very clever people would draw on that common analogous factoid if asked.
It’s like the theory of lift for wings. Oft misunderstood and the common explanation “proved” by abundance of quoting.

 

[chuckroast]

More than that, it completely ignores the factor noise, and it compares a wave, which is observed over time, with a wave, which is observed as a whole packet in one. Whatever the differences between analog vs. digital may be, they are not explained in this chart.

I know the referenced post was deleted but just one fine point of order. Sound and Light propagate by rather different means. Sound propagates by compressing the medium in which it exists. Light follows - IIRC - Maxwell's Equations and is a form of electromagnetic radiation. That's why there is light in the vacuum of space, but no sound.

In reference to this thread topic - it makes no difference how light is captured - analog, digital, a combination ... The light arrives in exactly the same way, all other things being equal. But, aye, there's the rub - they're aren't quite "equal" even if you assume a theoretically perfect taking lens to capture both ways.

With film, you have a single optical path - the taking lens - to worry about things like diffraction, especially. And, of course, you have to worry about the post capture optical paths for sharpness - either the enlarger reproduction chain or the scanner optics.

With digital, you can think of the sensor as being an array of lenses that capture the light being delivered by the taking lens. The sensor is essentially an array of additional tiny "lenses", one per pixel. These become relevant in the question of diffraction. This is why you get diffractive effects at larger f/stops with digital than with film and why early point and shoot digital cameras set their default shooting aperture around f/5.6 or so. These days, a lot of this has been remediated by better sensor design and in-camera digital post processing, but the underlying physics doesn't change.

In many respects, the difference comes down to film being asked to capture/record what is there, whereas digital - with it's many sophisticated in-camera optimization algorithms - is, to one degree or another - constructing an image of what is there. If you then look at the post processing tools like Topaz AI, they take this "construction" of the image to a whole new level using a whole bunch of machine learning tools to try to "fix" or "improve" the image to what it should have been irrespective of what was actually captured.

I say this without judgement, only to note that they are rather different approaches to capturing the same light.

P.S. All of this isn't new to digital, BTW. Many years ago I worked in the music recording business and one of our competitors was recording a famous classical pianist's new album. He had a particularly bad session but had to leave to perform with the local orchestra that night. The recording engineer told him "Don't worry about it. We'll take the best bits from each of the many takes today, and splice them together to create one perfect performance." In those days "splice" was literal. You copied the "best bit" onto new (analog) tape, and literally glued these bits of tape together, end-to-end, to make a new master recording.

When the performer heard the final, now flawless master, he was really impressed. The engineer looked at him and said (allegedly), "Yes, it's really great. Don't you wish you could actually play that way?"

 

[DREW WILEY]

Helge - I am 100% with you there, concerning Delta 100's "little more bite" of edge effect in comparison to TMX100's more "velvety" look (which I prize for portraiture, but not for landscapes). But TMX is more useful to me because the straight line or semi-straight line extends about a full stop deeper into the shadows. And this gives me two benefits : true 100 speed (whereas I have to shoot Delta at 50 to get the shadows further up the curve), and better performance in very high scene contrast ranges often typical of my work in the mountains especially. TMY also has a little better spectral sensitivity for my purposes. But I did take trouble to figure out how to make Delta 100 work as a substitute if necessary (some of its filter factors are different). Hence my Perceptol 1:3 tweak for sake of a little more "tooth" to the TMX grain.

I always keep both rolls of TMX 120 on hand for general 6x7 and 6x9 shooting, as well as sheets of 4x5 and 8x10 for sake of masking in the labs, but which get used at times for shooting too. But for general sheet film shooting, TMY400 is my preferred choice, and it gives good edge acutance with my routine PMK pyro development. But I sometimes shoot it in rolls too, if the wind and weather is so bad as to demand handheld work. And in the case of 35mm, I nearly always shoot handheld, and only print that small, so TMY400 is my routine choice there.

 

[Helge]

With digital, you can think of the sensor as being an array of lenses that capture the light being delivered by the taking lens. The sensor is essentially an array of additional tiny "lenses", one per pixel. These become relevant in the question of diffraction. This is why you get diffractive effects at larger f/stops with digital than with film and why early point and shoot digital cameras set their default shooting aperture around f/5.6 or so. These days, a lot of this has been remediated by better sensor design and in-camera digital post processing, but the underlying physics doesn't change.

In many respects, the difference comes down to film being asked to capture/record what is there, whereas digital - with it's many sophisticated in-camera optimization algorithms - is, to one degree or another - constructing an image of what is there. If you then look at the post processing tools like Topaz AI, they take this "construction" of the image to a whole new level using a whole bunch of machine learning tools to try to "fix" or "improve" the image to what it should have been irrespective of what was actually captured.

I say this without judgement, only to note that they are rather different approaches to capturing the same light.

Interesting post. I take it that is numerically higher f stops you mean? Only thing that makes sense in my mind.

This is most likely connected in kind to why insects has segmented eyes.
It would make no sense optically to have small image forming eyes with a single lens and iris.

Their eyes are more about establishing textures and navigating with the aid of easily recognizable textures and shapes. Hence the whole concept of flowers.

Helge - I am 100% with you there, concerning Delta 100's "little more bite" of edge effect in comparison to TMX100's more "velvety" look (which I prize for portraiture, but not for landscapes). But TMX is more useful to me because the straight line or semi-straight line extends about a full stop deeper into the shadows. And this gives me two benefits : true 100 speed (whereas I have to shoot Delta at 50 to get the shadows further up the curve), and better performance in very high scene contrast ranges often typical of my work in the mountains especially. TMY also has a little better spectral sensitivity for my purposes. But I did take trouble to figure out how to make Delta 100 work as a substitute if necessary (some of its filter factors are different). Hence my Perceptol 1:3 tweak for sake of a little more "tooth" to the TMX grain.

I always keep both rolls of TMX 120 on hand for general 6x7 and 6x9 shooting, as well as sheets of 4x5 and 8x10 for sake of masking in the labs, but which get used at times for shooting too. But for general sheet film shooting, TMY400 is my preferred choice, and it gives good edge acutance with my routine PMK pyro development. But I sometimes shoot it in rolls too, if the wind and weather is so bad as to demand handheld work. And in the case of 35mm, I nearly always shoot handheld, and only print that small, so TMY400 is my routine choice there.

Not much to add here. TMX and TMY are truly remarkable emulsions. They are in many ways the epitome of B&W technology.
Pity about the price hike on especially TMY.

 

[DREW WILEY]

Well, I paid less for a huge freezer than what a box of some 8x10 films cost now. It was a very wise investment, and the only realistic way I can afford to keep using TMY, TMX, and Ektar film - out of my own frozen stockpiile.

 

[Helge]

Well, I paid less for a huge freezer than what a box of some 8x10 films cost now. It was a very wise investment, and the only realistic way I can afford to keep using TMY, TMX, and Ektar film - out of my own frozen stockpiile.

Missed a good deal on three ten package “sticks” of TMY recently. Still kicking myself. There are currently more people than usually emptying out their freezers, for whatever reason, in Denmark.

 

[chuckroast]

Interesting post. I take it that is numerically higher f stops you mean? Only thing that makes sense in my mind.

I mean that digital will start showing diffraction effects at a wider opening than film will, ignoring post processing corrections. Where a film SLR may not diffract significantly until f/16 or f/22 (depending on lens), the equivalent digital capture system might start to show diffraction as early as, say, f/8.

See this for the gory details:

https://www.cambridgeincolour.com/tutorials/diffraction-photography-2.htm

 

[DREW WILEY]

I'm still buying my favorite Kodak films in smaller formats. There will come a time when I'll just be too old to enjoy lugging around 8X10 gear anyway. These newer films with their high detail capacity, along with really good modern lenses, has sure opened up the possibilities of Medium Format more than before.

 

[SodaAnt]

First, analog is a lot clumsier to post process in the darkroom than something from a digisnapper going onto a Mac for quick fixes. I have worked in the darkroom for 50 years and I am well aware how much effort it takes to really get a silver print right.

This is what originally sold me on computer post processing—the ability to do manipulations undreamed of in the darkroom. That, and sitting in a comfortable chair in normal room lighting, makes for a different, and sometimes more pleasant experience.

I agree with earlier posts in the thread that the controls programs like Photoshop offer are frequently overdone to the point the result looks artificial. I try to use the minimal amount of sharpening possible and rarely boost saturation to try to get that Velvia look.

On the other hand, mixing up chemicals, loading reels, and developing film is also relaxing, in a different way. Ditto for printing, although I’m not at the point yet where I can do that.

But in the end, it’s all photography, and it’s all good, whether analog, digital, or a combination of both.

 

[Nikon 2]

This is what originally sold me on computer post processing—the ability to do manipulations undreamed of in the darkroom. That, and sitting in a comfortable chair in normal room lighting, makes for a different, and sometimes more pleasant experience.

I agree with earlier posts in the thread that the controls programs like Photoshop offer are frequently overdone to the point the result looks artificial. I try to use the minimal amount of sharpening possible and rarely boost saturation to try to get that Velvia look.

On the other hand, mixing up chemicals, loading reels, and developing film is also relaxing, in a different way. Ditto for printing, although I’m not at the point yet where I can do that.

But in the end, it’s all photography, and it’s all good, whether analog, digital, or a combination of both.

I get the picture…!

 

[Rudeofus]

I mean that digital will start showing diffraction effects at a wider opening than film will, ignoring post processing corrections. Where a film SLR may not diffract significantly until f/16 or f/22 (depending on lens), the equivalent digital capture system might start to show diffraction as early as, say, f/8.

See this for the gory details:

https://www.cambridgeincolour.com/tutorials/diffraction-photography-2.htm

This effect may be weaker in practice, since typical Bayer filters are not all that selective. Therefore most demosaicing filter yield pixel accurate luminosity resolution. It's only color info which gets blurry.

 

[Alan Johnson]

There is also the factor that the largest grains in "ordinary" films of 25-100 ISO may be 75-200 x the size of the smallest grains.

Argenti Nanotomic-X B&W Film Review

Sroyon is here with another thorough film review, this time focusing on Argenti Nanotomic X, a respooled microfilm for 35mm cameras.

casualphotophile.com

This means that such films may show irregular [not straight] edges on enlargement due to the large grains, which results in less resolution of fine detail. On the other hand some viewers may count the appearance of large sharp grains as an improvement in apparent sharpness.
Microfilms may have a ratio of large grain to small grain size of only 10-20 x and are thus better at resolving fine detail straight lines but don't show apparent sharpness due to large grains.
So as Matt noted the meaning of "sharpness" has to be defined in the question.

 

[Nikon 2]

Just watched a YouTube video showing two gentlemen shooting their Leica M-A cameras at the Leica park in Europe. The photos of the Leica compound buildings were “perfect”. No distortion, straight lines, very sharp and couldn’t believe it was film! I suspect they used the top tier Summicron 50mm f/2 APO, A Perfect Optic!
There was also no soul. Too clinical for my tastes. Digital without the steroids…!

 

[Alan Johnson]

Even if the Youtube video was 4k, 3840 pix wide, that would be 53lppm on a 35mm negative.
Any consumer zoom should manage that.

 

[Helge]

This effect may be weaker in practice, since typical Bayer filters are not all that selective. Therefore most demosaicing filter yield pixel accurate luminosity resolution. It's only color info which gets blurry.

It has more to do with the topography and shape of the sense cells. They are almost always in a well structure, almost always with a micro lens on top and always in the shape of a polygon, never the optimal round.

Sense cells can also affect each other electrically and even optically.

Bayer filters ride the uneasy edge between prioritizing luminance information or colour.
Never able to reach a perfect compromise.

Luminance information is affected, even with the usually desaturated filters. Especially if it’s on a coloured surface like red fabric or even a brick wall at a distance.

The solution is of course to have three sensors, like it was common for video. But then you run into problems with exact optical matching of the three sensors.

 

[faberryman]

It has more to do with the topography and shape of the sense cells. They are almost always in a well structure, almost always with a micro lens on top and always in the shape of a polygon, never the optimal round.

Sense cells can also affect each other electrically and even optically.

Bayer filters ride the uneasy edge between prioritizing luminance information or colour.
Never able to reach a perfect compromise.

Luminance information is affected, even with the usually desaturated filters. Especially if it’s on a coloured surface like red fabric or even a brick wall at a distance.

The solution is of course to have three sensors, like it was common for video. But then you run into problems with exact optical matching of the three sensors.

Digital sensors seem to do a pretty good job. Film grain and dye clouds are not uniform and uniformly dispersed, and they do a pretty good job too.

 

[Helge]

Digital sensors seem to do a pretty good job. Film grain and dye clouds are not uniform and uniformly dispersed, and they do a pretty good job too.

They do the job quite differently. With appropriate artifacts to match.
That’s the whole point of this thread.

 

[Rudeofus]

They do the job quite differently. With appropriate artifacts to match.
That’s the whole point of this thread.

I guess the best suggestion so far was "because at low contrast, film doesn't show much resolution", which by far exceeds all effects of diffraction, lens sharpness and nominal film/sensor resolution. This, and the massive onslaught of image improvement algorithms on anything which happens to end up on a sensor ("maximum likelihood correct data to the nearest typical subject matter").

 

[Helge]

I guess the best suggestion so far was "because at low contrast, film doesn't show much resolution", which by far exceeds all effects of diffraction, lens sharpness and nominal film/sensor resolution. This, and the massive onslaught of image improvement algorithms on anything which happens to end up on a sensor ("maximum likelihood correct data to the nearest typical subject matter").

You could turn that around and say that “because of necessary edge detection, and contrast enhancement digital shows selective sharpness and smudge elsewhere.
And because of aliasing from various sources in the matrix and digitization, digital shows harshness and artificial sharpness where there isn’t any”.

Film shows resolution in a different kind of way. A way I personally find much more pleasing and in a way that leans into and parallels what psycho acoustics tells us that our ear like.
A frequency response with a slightly down slanted curve toward the highs.
Generally there are more overlaps than not with the very general way in which we respond between the two senses of sight and hearing.

 

[Nikon 2]

Digital sensors seem to do a pretty good job. Film grain and dye clouds are not uniform and uniformly dispersed, and they do a pretty good job too.

Actually, my Leica M-D 262, shows a more realistic landscape than my Nikon F2 before the CLA. I’m waiting for my prints from Blue Moon and ordered a premium drum scan and flash drive...!

 

[chuckroast]

You could turn that around and say that “because of necessary edge detection, and contrast enhancement digital shows selective sharpness and smudge elsewhere.
And because of aliasing from various sources in the matrix and digitization, digital shows harshness and artificial sharpness where there isn’t any”.

Film shows resolution in a different kind of way. A way I personally find much more pleasing and in a way that leans into and parallels what psycho acoustics tells us that our ear like.
A frequency response with a slightly down slanted curve toward the highs.
Generally there are more overlaps than not with the very general way in which we respond between the two senses of sight and hearing.

I am put in mind of the wars that erupted when the Moog synthesizer first hit the music world. There was this back and forth about the end of "real" music, piano manufacturers going out of business and so on. Well, here we are some 50 years or so later and Steinway and Bosendorfer are still around, classical pianists are still making marvelous recordings, and electronic music has just become its own thing.

In an analogous way, this thread illuminates a deeper truth. Digital and film are just entirely different media. They are no more related to each other than they are to oil painting. Digital, film, and oil painting all capture light by one means or another, but no one would compare, say, the resolution of film to a Van Gogh.

Different media help the artist to see their world in different ways. I shoot film one way. I shoot digital an entirely different way. More importantly, I don't try to map one to the other. They are just ... different and I treat them as nearly unrelated activities (except for the part where my Nikon AI-S lenses work well on my D750 ... but that's a discussion about types of paint brushes).

 

[MurrayMinchin]

Some photographers, like Dick Arentz, straddle both worlds.

"Those familiar with my work may notice a stylistic change in vision. I consider palladium prints made from film and those from digital techniques to be distinctly different media. Therefore, I adjusted my vision to accommodate those different characteristics..."

"...He works with a historical Leica f/1.4 lenses noted for its soft focus and flare effects, many times simulating the lenses used 100 or more years ago."

In the link posted below, there are examples of his palladium cathedral prints from a digital camera. (You can find other work of his there, including 12x20 LF work).

https://www.dickarentz.com/recent-work/the-english-cathedrals/

I remember reading a discussion about how pin registered sharp & unsharp masking was bad, because the writers had seen some harsh examples where the effects went too far and were painfully obvious. We've all seen bad dodging & burning, but that doesn't automatically mean the technique itself is the problem.

Same can be said of digital, where many prints I see (to my eye) tend to be over sharpened.

The largest prints I made from my 4x5 negatives were 11x14. While different, the prints of about the same size I've made through the digital realm hold their own in comparison.

 

[MurrayMinchin]

...Different media help the artist to see their world in different ways. I shoot film one way. I shoot digital an entirely different way. More importantly, I don't try to map one to the other. They are just ... different and I treat them as nearly unrelated activities...

Perfect post to sneak in right before mine

 

[chuckroast]

Perfect post to sneak in right before mine

Hehe - I was peering over yer shoulder ...