35mm Film Resolution

[Ambro51]

Advertisement (gone after registration)

The lenses on the Stereo Realist CUSTOM “Rare Earth Germany” are said to have resolution of 300 lpm.    I’d imagine K25 would have been able to handle that.  

Edited November 13, 2019 by Ambro51

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[John Miranda]

In re-experimenting with film, my scanner can produce a fairly large size file (6k or more pixels horizontally).  What I do see, more than resolution, (it can be good or bad, depending on the intent) is a much narrower dynamic range than the digital counterpart 35mm file.  In B&W though, I am enjoying the contrasty images right out of the film.

[A miller]

2 hours ago, John Miranda said:

In re-experimenting with film, my scanner can produce a fairly large size file (6k or more pixels horizontally).  What I do see, more than resolution, (it can be good or bad, depending on the intent) is a much narrower dynamic range than the digital counterpart 35mm file.  In B&W though, I am enjoying the contrasty images right out of the film.

that doesn't feel right, John.  What scanning software are you using?

[John Miranda]

Hi A miller - I am using VueScan as the software and PrimeFilm XE for the scanner.  But your thoughts, what aspect doesn't feel right, that I can get a scan with 6k pixels horizontally from 35mm film, or that film has a lesser dynamic range then new digital camera bodies?

[Ko.Fe.]

If you need resolution, you don’t need film. Film quality is in gradations of colours and tones.
The only film resolution I impressed with is LF.

 

[adan]

I don't know about A miller, but it strikes me as odd that you are getting less dynamic range from film than from digital (although digital has gotten quite good).

Film (except perhaps for the exotic, low-iso, high-contrast micro/document films) should be able to handle about 13 stops of DR (a scene brightness range of 1:8192 ). Per Kodak.

However, that maximum can be degraded by exposure, development, and scanning techniques. Especially with roll films, where 12, 24, or 36 exposures all have to undergo the same development, regardless of the type of subject/scene in each picture.

Underexposure can lead to clipped shadow detail (the darkest areas register the same as blank, unexposed film). Overdevelopment can lead to crushed highlights. And the scanner software settings  (especially "auto" settings) can also clip away detail that is actually present in the film. Scanner illumination type (exposed LEDs vs. flat-plate diffusion panels) can also change just how much detail is recorded. (It is not clear to me whether the PrimeFilm XE uses diffused or "bare" LEDs). See the Callier effect (works for scanner lighting as well as enlarger lighting): https://en.wikipedia.org/wiki/Callier_effect

_________

Below is an experimental exposure I made on TMax 400 film (6x6), of a scene with extreme brightness range. The f/numbers show which f/stop would be needed to get a "medium gray" tone for various parts of the scene - 13 stops difference. I in fact exposed by metering the darkest area (f/0.7, black curtain with white polkadots, shaded by the coffee stand), and then stopping down 3 stops from that meter reading to make that area a dark, but not pure-black, tone (in Zone System parlance, I "placed" that area in Zone II with exposure). Hand-held meter rather than internal camera meter.

I then developed with reduced ("pull") development - the developer (HC-110) was overdiluted to 1:55 instead of the normal 1:32, and in addition I cut back on the agitation - once every 2 minutes (120 seconds) instead of once every 30 seconds (Kodak's "box" technique).

I also used Vuescan, with a diffuse-illumination scanner (Epson flatbed). I used Vuescan's "Auto Levels" exposure setting - BUT I also included the black border around the actual image  (film-base-plus-fog density), so that Vuescan's autoexposure algorithm "calibrated" itself to give me pure black only in that completely unexposed area. I also made an overall camera-meter reading - which would have resulted in a higher shutter speed (influence of the white wall in the scene) and in less exposure and lost shadow detail.

Result - detail on both black cloth in deep shadow, and in white-white paint in direct sunlight.

A bit of a "party trick" just to test Kodak's 13-stop claim - but it did verify the range that can be captured.

There are more convenient ways to expose and process film (overall reflected camera metering, auto-exposure, automated lab processing machine) - but they can extract a price in dynamic-range quality.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Ko.Fe.]

Advertisement (gone after registration)

2 hours ago, adan said:

I don't know about A miller, but it strikes me as odd that you are getting less dynamic range from film than from digital (although digital has gotten quite good).

Film (except perhaps for the exotic, low-iso, high-contrast micro/document films) should be able to handle about 13 stops of DR (a scene brightness range of 1:8192 ). Per Kodak.

However, that maximum can be degraded by exposure, development, and scanning techniques. Especially with roll films, where 12, 24, or 36 exposures all have to undergo the same development, regardless of the type of subject/scene in each picture.

Underexposure can lead to clipped shadow detail (the darkest areas register the same as blank, unexposed film). Overdevelopment can lead to crushed highlights. And the scanner software settings  (especially "auto" settings) can also clip away detail that is actually present in the film. Scanner illumination type (exposed LEDs vs. flat-plate diffusion panels) can also change just how much detail is recorded. (It is not clear to me whether the PrimeFilm XE uses diffused or "bare" LEDs). See the Callier effect (works for scanner lighting as well as enlarger lighting): https://en.wikipedia.org/wiki/Callier_effect

_________

Below is an experimental exposure I made on TMax 400 film (6x6), of a scene with extreme brightness range. The f/numbers show which f/stop would be needed to get a "medium gray" tone for various parts of the scene - 13 stops difference. I in fact exposed by metering the darkest area (f/0.7, black curtain with white polkadots, shaded by the coffee stand), and then stopping down 3 stops from that meter reading to make that area a dark, but not pure-black, tone (in Zone System parlance, I "placed" that area in Zone II with exposure). Hand-held meter rather than internal camera meter.

I then developed with reduced ("pull") development - the developer (HC-110) was overdiluted to 1:55 instead of the normal 1:32, and in addition I cut back on the agitation - once every 2 minutes (120 seconds) instead of once every 30 seconds (Kodak's "box" technique).

I also used Vuescan, with a diffuse-illumination scanner (Epson flatbed). I used Vuescan's "Auto Levels" exposure setting - BUT I also included the black border around the actual image  (film-base-plus-fog density), so that Vuescan's autoexposure algorithm "calibrated" itself to give me pure black only in that completely unexposed area. I also made an overall camera-meter reading - which would have resulted in a higher shutter speed (influence of the white wall in the scene) and in less exposure and lost shadow detail.

Result - detail on both black cloth in deep shadow, and in white-white paint in direct sunlight.

A bit of a "party trick" just to test Kodak's 13-stop claim - but it did verify the range that can be captured.

There are more convenient ways to expose and process film (overall reflected camera metering, auto-exposure, automated lab processing machine) - but they can extract a price in dynamic-range quality.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

It is simple. At some point digital gear reviewers switched dr to the term how many iso stops they could push. 

[A miller]

3 hours ago, adan said:

I don't know about A miller...

 

 

That's "Fedoraman" to you, Andy. 🕵️‍♂️

What I was saying is that a good scanner with good software settings should produce a flat scan of "negative" film (positive film a different story) that has a very wide range of tones and EVs that will serve as a buffet for a wide range of digital dodging and burning renditions.

Edited November 14, 2019 by A miller

[Jon Warwick]

On 11/14/2019 at 1:26 PM, Ko.Fe. said:

If you need resolution, you don’t need film. Film quality is in gradations of colours and tones.
The only film resolution I impressed with is LF.

 

I personally find it very hard to get the same gradations of colours and tones from digital as I get from film. Images off film, to my eyes, simply look less “flat”.

Not sure if there’s any post processing techniques to get me closer?

In terms of resolution, I’ve got a lot of drum scanned material.

My instinct is a 45mp camera is similar in apparent resolution to a drum-scanned 6x7 E6 on c 45” wide print.

But it’s quite a jump required in megapixels to get close to the 1.6gb(!) drum scans I get done off 4x5 when printing to c 60”+ wide ..... I’ve played with files from Phase One 150mp back, and they were the first I’ve used that felt similar to the 4x5. Very impressive image quality but too $$$ for my wallet!

Edited November 16, 2019 by Jon Warwick

[Ko.Fe.]

6 hours ago, Jon Warwick said:

I personally find it very hard to get the same gradations of colours and tones from digital as I get from film. Images off film, to my eyes, simply look less “flat”.

Not sure if there’s any post processing techniques to get me closer?

In terms of resolution, I’ve got a lot of drum scanned material.

My instinct is a 45mp camera is similar in apparent resolution to a drum-scanned 6x7 E6 on c 45” wide print.

But it’s quite a jump required in megapixels to get close to the 1.6gb(!) drum scans I get done off 4x5 when printing to c 60”+ wide ..... I’ve played with files from Phase One 150mp back, and they were the first I’ve used that felt similar to the 4x5. Very impressive image quality but too $$$ for my wallet!

You can't PP something which doesn't exist...

It first came to me in nineties then I started as technician on television. I noticed how films taken on film looks much pleasing than on video. I asked ex Silicon Graphics  Instruments engineer about it and he was first person to explain it. Basically, here is not so much of the limit of DR for tones and colors for film. 

One thing which stickers me is how it is still visible then film is digitized :).

Forgot to add. Color on film is defined by people just as it is defined on digital. Yet, cheap Kodak Gold film colors are much more pleasing to me. Don't know why. Taste? Hiring wrong people? I just looked at Sony A7s II colors on its Flickr group. Colors just awful, IMO.

Edited November 16, 2019 by Ko.Fe.

[250swb]

On 11/13/2019 at 9:44 PM, John Miranda said:

In re-experimenting with film, my scanner can produce a fairly large size file (6k or more pixels horizontally).  What I do see, more than resolution, (it can be good or bad, depending on the intent) is a much narrower dynamic range than the digital counterpart 35mm file.  In B&W though, I am enjoying the contrasty images right out of the film.

Make your scans intentionally low contrast, don't try to make the final image in the scan. All you want is all the information available from the negative, the scan will look awful but then with that information you can post process it in Lightroom or Photoshop creating a proper contrast.

[John Miranda]

thank you great advice on scanning using the full gamut then processing in Photoshop.  

[Einst_Stein]

While I also feel the film has the more pleasant tonal gradation, somehow most posted film images actually looks very harsh.

The beautiful tonal gradation only exists in my old memory of pictures from the best photo labs.

Is it scanner or over post-processing? or,  bad monitor, bad color management? 

[250swb]

On 11/22/2019 at 11:16 PM, Einst_Stein said:

 

Is it scanner or over post-processing? or,  bad monitor, bad color management? 

Fashion?

Looking for a technical fault often overlooks the greater aesthetic milieu. For example the Japanese aesthetic for street photography often encompasses the things you don't like. Equally people feeling down about the current political climate around the world may express this. Personally I blame Rodinal for the ease by which using it can influence such fashions and expressions, and become an expression itself, but I also like it because nowadays photography should have gone well beyond pandering to the camera club anyway.

[Doc Henry]

....  one useful lecture

Rg H 

https://istillshootfilm.org/post/114131916747/the-real-resolution-of-film-vs-digital?fbclid=IwAR275o-bc2hhjmVJO8klASItUB7DLUyEfQMNXIJuaqsuFE1F7DM6nt-ENUs

Edited November 27, 2019 by Doc Henry

[frame-it]

and one more useful article http://imbs.uci.edu/~kjameson/ECST/Witzel_Gegenfurtner_MemoryColor.pdf

[frame-it]

On 11/11/2019 at 10:03 PM, jaapv said:

 

To measure resolution you count LP/mm for a given contrast level and take it from there.

this would help

http://www.imatest.com/micro_site/imatest-5-2/