MP Resolution of 35mm Film?

Started 4 months ago | Discussions
grsnovi
grsnovi Regular Member • Posts: 471
MP Resolution of 35mm Film?

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

 grsnovi's gear list:grsnovi's gear list
Panasonic LX100 Panasonic Lumix DMC-ZS40 (TZ60) Olympus Tough TG-1 iHS Sony SLT-A65 Panasonic Lumix DMC-GX85 +9 more
Erik Kaffehr
Erik Kaffehr Senior Member • Posts: 3,311
Re: MP Resolution of 35mm Film?
4

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

Hi,

Back in the old time, when digital was around, many experts estimated 24x36 mm to be around 6MP.

In the old time, when Modern Photography tested lenses they often could achieve like 50 lp/mm on film. 50 lp/mm correspond by and large to 100 pixels/mm.

That give 24 * 100 * 36 * 100 -> 8.6 MP

But, films are different. Slide film does not have very high resolution while B&W negative films can have very high resolution.

Personally, I have found that 24 MP on 24x36 gave me better results than Velvia on 6x7 cm.

Keep in mind that resolution is one thing, it just means that two high contrast structures can be separated. Film tend to be noisy, and that noise covers low contrast detail. Digital sensors generally yield pretty low noise, so they may be able to show fine low contrast detail which is lost on film.

Many 4"x5" film users switched to the 39 MP Phase One P45 back as they considered it to deliver image quality like Velvia on 4x5. Now, 4"x5" has 15X the area of 24x36 mm, so 39MP / 15 -> 2.6MP.

I don't think there is a single number.

Best regards

Erik

-- hide signature --

Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles

Mark Scott Abeln
Mark Scott Abeln Forum Pro • Posts: 14,156
Re: MP Resolution of 35mm Film?
1

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

That’s a common question! And there is no one answer, because it depends on what film you use, along with what processing and how the film is printed. There is a big difference between fine grained film and Tri-X pushed to ISO 1600, which is what I liked to do, but even my oldest five megapixel sensor easily beat that.

When you look closely at film, you reach a point where you can’t be quite sure what is real detail and what is just random clumping of grain (at least with black and white film) and the trouble is that *all* film detail is grain clumping.

 Mark Scott Abeln's gear list:Mark Scott Abeln's gear list
Nikon D200 Nikon D7000 Nikon D750 Nikon AF Nikkor 50mm f/1.8D Nikon AF-S DX Nikkor 35mm F1.8G +2 more
alanr0 Senior Member • Posts: 2,237
Resolution of 35mm Film & MTF 30?
2

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

As Erik and Mark have already pointed out, the characteristics of film and digital sensors are very different.

You might find this document by Tim Vitale of interest: http://www.tmax100.com/../film.pdf. It discusses differences between halide particle size and grain size, cloud size in colour emulsions, and has tables comparing the resolutions of different emulsions.

Resolution, measured via Modulation Transfer Function (MTF) at 30% contrast varies from 35 lp/mm for Ektachrome 160 up to 160 lp/mm for T-MAX 100.

Kodak Ektachrome EKT 160 35 lp/mm
Kodak PORTRA 160VC 65 lp/mm
Fuji Velvia RVP 80 lp/mm
Kodak VR100 Color Neg 100 lp/mm
Kodak Plus-X B&W negative 125 lp/mm
Kodak T-MAX 100 B&W negative 160 lp/mm

From http://wwwru.kodak.com/RU/ru/professional/support/techPubs/f8/f8.jhtml

Kodak T-MAX 100: https://imaging.kodakalaris.com/sites/prod/files/files/resources/f4016_TMax_100.pdf

From the data sheet for Fuji Velvia RVP100, effective resolution depends on the subject contrast:

Resolution at Chart Contrast 1.6:1 .......................... 80 lines/mm
Resolution at Chart Contrast 1000:1 .........................160 lines/mm (80 lp/mm ?)

https://www.fujifilm.com/products/professional_films/pdf/velvia_100_datasheet.pdf

If we take the rather simplistic view that we need 2 pixels for each line-pair, then equivalent pixel counts at MTF30 range from 4 - 22 Mp for slide film to 88 Mp for high resolution B&W film.

-- hide signature --

Alan Robinson

grsnovi
OP grsnovi Regular Member • Posts: 471
Re: MP Resolution of 35mm Film?

Thanks guys! I thought it might have been a common question and I knew that various films had differing characteristics. I too shot a lot of Tri-X pushed. I also had my own darkroom although I never did any color in it. I have to say that I never looked as closely at a negative as I have looked at any digital image. Thanks for indulging me...

 grsnovi's gear list:grsnovi's gear list
Panasonic LX100 Panasonic Lumix DMC-ZS40 (TZ60) Olympus Tough TG-1 iHS Sony SLT-A65 Panasonic Lumix DMC-GX85 +9 more
D Cox Forum Pro • Posts: 24,393
Re: MP Resolution of 35mm Film?
1

grsnovi wrote:

Thanks guys! I thought it might have been a common question and I knew that various films had differing characteristics. I too shot a lot of Tri-X pushed. I also had my own darkroom although I never did any color in it. I have to say that I never looked as closely at a negative as I have looked at any digital image. Thanks for indulging me...

Didn't you have a grain magnifier for focussing your enlarger ?

Entropy512 Senior Member • Posts: 4,669
Re: MP Resolution of 35mm Film?

Erik Kaffehr wrote:

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

Hi,

Back in the old time, when digital was around, many experts estimated 24x36 mm to be around 6MP.

In the old time, when Modern Photography tested lenses they often could achieve like 50 lp/mm on film. 50 lp/mm correspond by and large to 100 pixels/mm.

That give 24 * 100 * 36 * 100 -> 8.6 MP

But, films are different. Slide film does not have very high resolution while B&W negative films can have very high resolution.

Personally, I have found that 24 MP on 24x36 gave me better results than Velvia on 6x7 cm.

Keep in mind that resolution is one thing, it just means that two high contrast structures can be separated. Film tend to be noisy, and that noise covers low contrast detail. Digital sensors generally yield pretty low noise, so they may be able to show fine low contrast detail which is lost on film.

Many 4"x5" film users switched to the 39 MP Phase One P45 back as they considered it to deliver image quality like Velvia on 4x5. Now, 4"x5" has 15X the area of 24x36 mm, so 39MP / 15 -> 2.6MP.

I don't think there is a single number.

Best regards

Erik

Also keep in mind that as far as sensitivity/"film speed":

Digital sensors fix resolution but increase noise per pixel as they increase in sensitivity ratings.  E.g. EDR decreases with noise, but resolution stays the same.

Film, instead, effectively keeps the per-grain "EDR" (I use quotes because it's almost surely not quite the formal definition of EDR, but it performs in a somewhat similar fashion) the same but reduces resolution by increasing the size of the film grains.

High-sensitivity film has much larger grains and hence reduced resolution.

-- hide signature --

Context is key. If I have quoted someone else's post when replying, please do not reply to something I say without reading text that I have quoted, and understanding the reason the quote function exists.

 Entropy512's gear list:Entropy512's gear list
Sony a6000 Pentax K-5 Pentax K-01 Sony a6300 Canon EF 85mm f/1.8 USM +5 more
Erik Kaffehr
Erik Kaffehr Senior Member • Posts: 3,311
Re: Resolution of 35mm Film & MTF 30?
1

alanr0 wrote:

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

As Erik and Mark have already pointed out, the characteristics of film and digital sensors are very different.

You might find this document by Tim Vitale of interest: http://www.tmax100.com/../film.pdf. It discusses differences between halide particle size and grain size, cloud size in colour emulsions, and has tables comparing the resolutions of different emulsions.

Resolution, measured via Modulation Transfer Function (MTF) at 30% contrast varies from 35 lp/mm for Ektachrome 160 up to 160 lp/mm for T-MAX 100.

Kodak Ektachrome EKT 160 35 lp/mm
Kodak PORTRA 160VC 65 lp/mm
Fuji Velvia RVP 80 lp/mm
Kodak VR100 Color Neg 100 lp/mm
Kodak Plus-X B&W negative 125 lp/mm
Kodak T-MAX 100 B&W negative 160 lp/mm

From http://wwwru.kodak.com/RU/ru/professional/support/techPubs/f8/f8.jhtml

Kodak T-MAX 100: https://imaging.kodakalaris.com/sites/prod/files/files/resources/f4016_TMax_100.pdf

From the data sheet for Fuji Velvia RVP100, effective resolution depends on the subject contrast:

Resolution at Chart Contrast 1.6:1 .......................... 80 lines/mm
Resolution at Chart Contrast 1000:1 .........................160 lines/mm (80 lp/mm ?)

https://www.fujifilm.com/products/professional_films/pdf/velvia_100_datasheet.pdf

If we take the rather simplistic view that we need 2 pixels for each line-pair, then equivalent pixel counts at MTF30 range from 4 - 22 Mp for slide film to 88 Mp for high resolution B&W film.

Hi,

Quite true, but it is only a part of the story. The film has an MTF and so has the lens. Good film era lenses had perhaps 60% MTF at 40 lp/mm, measured on the optical bench. We could put a large format Velvia on the light table, but I don't think anyone would be interested in viewing a 24x36 mm slide on the light table. So, we need to either:

  • Project it
  • Print it in the wet darkroom
  • Scan it

With projection the MTF of the projection lens also comes into the equation. I would guess a bit below 0.6 at 40 lp/mm.

The Velvia 100 has an MTF around 55% at 40 lp/mm. So total MTF at projection may be:

0.6 * 0.55 * 0.6 -> 0.2 that is 20%

Printing in wet room may yield similar results, assuming top notch enlarger optics, like Apo Rodagon, used at near optimal aperture.

I don't know about MTF/aperture on drum scanners.

With the digital workflow we can use sharpening both on capture and before printing, that is possible in part because digital capture is less noisy and that depends in part on digital capture having high quantum efficiency.

So, I don't think there is a simple answer.

On the other hand, it seems that a lot of 4"x5" photographers switched from 4"x5" Velvia to 39 MP on Phase One P45+ and those photographers probably knew how to scan.

On the other hand, it seems to me that scanning may have been at 2000 PPI in that time, because it was good enough in that time. But, scanning film at say 6000 PPI may deliver better detail.

I have some 6000 PPI scans from 6x7 Velvia and I would think that lenses and photographic technique may be a more limiting factor than film.

That said, I have seen some quite impressive drum scan from Mamiya 7 on high res B&W film, but the Mamiya 67 was pretty large in format and had excellent lenses. Not least, the scanner operator, Tim Parkin, knew what he was doing.

Best regards

Erik

-- hide signature --

Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles

Trebor1 Regular Member • Posts: 194
Re: Resolution of 35mm Film & MTF 30?
2

alanr0 wrote:

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

As Erik and Mark have already pointed out, the characteristics of film and digital sensors are very different.

You might find this document by Tim Vitale of interest: http://www.tmax100.com/../film.pdf. It discusses differences between halide particle size and grain size, cloud size in colour emulsions, and has tables comparing the resolutions of different emulsions.

Resolution, measured via Modulation Transfer Function (MTF) at 30% contrast varies from 35 lp/mm for Ektachrome 160 up to 160 lp/mm for T-MAX 100.

Kodak Ektachrome EKT 160 35 lp/mm
Kodak PORTRA 160VC 65 lp/mm
Fuji Velvia RVP 80 lp/mm
Kodak VR100 Color Neg 100 lp/mm
Kodak Plus-X B&W negative 125 lp/mm
Kodak T-MAX 100 B&W negative 160 lp/mm

From http://wwwru.kodak.com/RU/ru/professional/support/techPubs/f8/f8.jhtml

Kodak T-MAX 100: https://imaging.kodakalaris.com/sites/prod/files/files/resources/f4016_TMax_100.pdf

From the data sheet for Fuji Velvia RVP100, effective resolution depends on the subject contrast:

Resolution at Chart Contrast 1.6:1 .......................... 80 lines/mm
Resolution at Chart Contrast 1000:1 .........................160 lines/mm (80 lp/mm ?)

https://www.fujifilm.com/products/professional_films/pdf/velvia_100_datasheet.pdf

If we take the rather simplistic view that we need 2 pixels for each line-pair, then equivalent pixel counts at MTF30 range from 4 - 22 Mp for slide film to 88 Mp for high resolution B&W film.

Useful additional data, regarding true film resolution, at at a specified contast level, can be found if movie film stock (post year 2000) are also considered. In the following link, the film contrast for green, at 80 lp/mm, is 50%, with blue higher and red considerably lower. When the whole camera system (Panavision), including lens, was measured by the ITU, the results were as follows:

Close to limiting resolution at MTF 6% = 106 lp/mm and at the previously considered 80 lp/mm, a contrast of 17% was achieved.

In film scanning there appears to be a semi-arbitrary threshold of 10% MTF, I've always wondered - why not 9% to match the Rayleigh Criterion? The paper suggests that a scanner of up to 11K resolution might be required to capture all the data on film but then the economics start to intrude!

https://www.dft-film.com/downloads/white-papers/DFT-SCANITY-white-paper.pdf

Yesterday, I was looking at the figures for resolution from Super 16 film and the arguments (plus downright lies) from some of the TV broadcast representatives, that were advanced . It was boldly stated back in the day, that Super 16 was only Standard Definition and not capable of HD but this appears to have been more related to deficiencies in the early MPEG2 codecs (complaints of too much moving noise/graininess in static scenes), rather than based on the actual film resolution figure.

This caused a lot of grief for nature photographers in particular, who were effectively forced to abandon this convenient format. Ironically, the early HD video cameras were initially restricted to 1440 horizontal pixels rather than the HD TV standard of 1920 but were allowed anyway.

There were a number of productions shot for the BBC and ITV on Super 16 then broadcast in Standard Definition, as per the (unnecessary) mandated standard but were then later released as HD, on Blu-Ray! These include the first 2 (3?) series of Merlin and also Doc Martin which was latterly repeated on ITV, in 1080i.

https://cinematography.com/index.php?/topic/70895-broadcast-standards-for-bbc/

Things have moved on and now it is relatively common to scan Super 16mm film at 4K, in order to deliver HD content (But still not permitted for original TV broadcast). This scanning requirement appears sensible because at 80 lp/mm, with a frame length of 11.76 mm (12.52 mm maximum gate), this would need a minimum of 1881 pixels or 2003 pixels at open-gate. Super 35 film is scanned at either 6K or 8K, for UHD/Cinema 4K output.

On a Super 35 mm film camera, an Arriflex 535B for instance, the ground-glass (safe area) when shooting at a 1.85:1 aspect ratio is 21 mm on the long dimension. Scaling to 35 mm full-frame would give a horizontal resolution of 7021 pixels (assuming that the camera lens is equally as good and that the Super 35 scan delivers Cinema 4K, or 4096 pixels. This would equate to 32.87 MP for the 24 x 36 mm format but at a relatively low contrast, because this would be at 97.5 lp/mm.

OTOH, at 50 lp/mm the Super 35 camera plus lens, shown in the first link, delivers 43% contrast, on film and the overall MTF after scanning will still be useful. This would equate to 8.64 MP, on 35 mm full-frame, when using colour film, assuming that there is not a huge difference in the lens performance between the two formats.

What is the low contrast threshold, below which high spatial frequency image detail is not perceptually significant: 20%, 15%?

rich_cx139
rich_cx139 Regular Member • Posts: 192
Re: MP Resolution of 35mm Film?

Have you had a look at Roger Clark's site: there is a wealth of stuff there including thus article on film:

https://clarkvision.com/articles/film.vs.digital.summary1/

Since I only used and use consumer film ( ie not velvia or portra etc) in my case almost any digital camera  i use has a higher effective MPx than any of my film shots - obvious anyway when i scan negs.  Its not about resolution for me - its very cliched but its the 'whole experience'

richard

ps talking 35mm here NOT  large/ medium format film where its still a different ballgame

A Marcus Forum Member • Posts: 58
Re: MP Resolution of 35mm Film?
1

I don’t think the world will see future films with higher and higher resolution. I do think we will see imaging chips with higher and higher resolution.

I think this now boils down to optics. In other words, the limiting factor is how sharp an image the lens can produce. Lenses are plagued by the twin demons of diffraction and interference which limit their sharpness.

This was well studied by John Strutt, Nobel Prize Physics (3rd Baron Rayleigh & Astronomer Royal).

The Rayleigh Criterion --- The theoretical resolving power of a lens is stated in lines resolved per millimeter. This value varies with wavelength but it is generally stated for green wavelength 589 millimicrons as Resolving power = 1392 ÷ f-number.

Thus a table is generated (has not yet been exceeded):
f/1 =1392 lines/mm
f/1.4 =994 lines/mm
f/2 = 696 lines/mm
f/2.8 = 497 lines/mm
f/4 = 348 lines/ mm
f/5.6 = 249 lines/mm
f/8 = 174 lines/mm
f/11 = 126 lines/mm
f/16 = 87 lines/mm
f/22 = 63 lines/mm

SmilerGrogan Contributing Member • Posts: 767
And While We're On The Subject…

This is a great opportunity for you guys to recollect what microcontrast really means.

alanr0 Senior Member • Posts: 2,237
Lens resolution - reference wavelength
1

A Marcus wrote:

I don’t think the world will see future films with higher and higher resolution. I do think we will see imaging chips with higher and higher resolution.

I think this now boils down to optics. In other words, the limiting factor is how sharp an image the lens can produce. Lenses are plagued by the twin demons of diffraction and interference which limit their sharpness.

This was well studied by John Strutt, Nobel Prize Physics (3rd Baron Rayleigh & Astronomer Royal).

The Rayleigh Criterion --- The theoretical resolving power of a lens is stated in lines resolved per millimeter. This value varies with wavelength but it is generally stated for green wavelength 589 millimicrons as Resolving power = 1392 ÷ f-number.

Have you any evidence that lens resolution is generally stated for 589 nm?

Low pressure sodium discharge lamps were widely used in school and undergraduate physics laboratories in the 1960's and 1970's. The emission spectrum is a doublet (589.0 & 589.6 nm). This may explain why the sodium D line was discarded from the International Commission of Optics (ICO) list of recommended emission lines in 1962.

The peak sensitivity of human vision is for 555 nm green light, but 589 nm is yellow.

A 550 nm green wavelength is commonly used when discussing microscope resolution (Zeiss , Nikon, Olympus).

Norman Koren mentions 550 nm, but favours 500 nm for daylight illumination.

Edmund optics quote resolution for 520 nm green light.

I would be interested to hear if there is a more widely recognized or de-facto standard, but I doubt it is 589 nm.

Thus a table is generated (has not yet been exceeded):
f/1 =1392 lines/mm
f/1.4 =994 lines/mm
f/2 = 696 lines/mm
f/2.8 = 497 lines/mm
f/4 = 348 lines/ mm
f/5.6 = 249 lines/mm
f/8 = 174 lines/mm
f/11 = 126 lines/mm
f/16 = 87 lines/mm
f/22 = 63 lines/mm

I assume these are line-pairs per mm at 589 nm for the Rayleigh diffraction limit, not to be confused with the line-widths (for MTF50) used by sites such as OpticalLimits.

-- hide signature --

Alan Robinson

A Marcus Forum Member • Posts: 58
Re: Lens resolution - reference wavelength

@alano --- As you know, the Rayleigh Criterion is universally accepted. The lines per millimeter limit is a variable based on wavelength. Photographically 1392 millimicrons is based on the visual focus of person working with a ground-glass image and a loupe. Films have a different chemical focus based on type.

alanr0 Senior Member • Posts: 2,237
Re: Lens resolution - reference wavelength
1

A Marcus wrote:

@alano --- As you know, the Rayleigh Criterion is universally accepted. The lines per millimeter limit is a variable based on wavelength.

Sure. It is not the only resolution criterion, but it is very widely used.

Photographically 1392 millimicrons is based on the visual focus of person working with a ground-glass image and a loupe. Films have a different chemical focus based on type.

I am not clear what you mean, or how this relates to lens resolution.

For a diffraction-limited lens, the Rayleigh criterion at wavelength λ and f-number N corresponds to a resolution at 9% contrast of a spatial frequency of 1/(1.22 λ N)

1392 cycles per mm is equal to 1/(1.22 x 0.000589) = 1/(1.22 λ) when the wavelength, λ, is 589 nm.

This has nothing to do with ground glass.

I questioned why you chose 589 nm, which is yellow/orange and not green. Presumably either the sodium D doublet, or perhaps the zinc 589.44 nm emission line?

As far as I can tell, it is far more common to specify lens resolution at a green wavelength near 550 nm, where the eye is most sensitive, or at a slightly shorter wavelength such as 500 nm or 520 nm.

Not a big deal. I was simply curious why you thought [resolution] "varies with wavelength but it is generally stated for green wavelength 589 millimicrons".  589 nm seemed an odd choice.

Cheers.

-- hide signature --

Alan Robinson

Trebor1 Regular Member • Posts: 194
Re: Lens resolution - reference wavelength
1

alanr0 wrote:

A Marcus wrote:

@alano --- As you know, the Rayleigh Criterion is universally accepted. The lines per millimeter limit is a variable based on wavelength.

Sure. It is not the only resolution criterion, but it is very widely used.

Photographically 1392 millimicrons is based on the visual focus of person working with a ground-glass image and a loupe. Films have a different chemical focus based on type.

I am not clear what you mean, or how this relates to lens resolution.

For a diffraction-limited lens, the Rayleigh criterion at wavelength λ and f-number N corresponds to a resolution at 9% contrast of a spatial frequency of 1/(1.22 λ N)

1392 cycles per mm is equal to 1/(1.22 x 0.000589) = 1/(1.22 λ) when the wavelength, λ, is 589 nm.

This has nothing to do with ground glass.

I questioned why you chose 589 nm, which is yellow/orange and not green. Presumably either the sodium D doublet, or perhaps the zinc 589.44 nm emission line?

As far as I can tell, it is far more common to specify lens resolution at a green wavelength near 550 nm, where the eye is most sensitive, or at a slightly shorter wavelength such as 500 nm or 520 nm.

Not a big deal. I was simply curious why you thought [resolution] "varies with wavelength but it is generally stated for green wavelength 589 millimicrons". 589 nm seemed an odd choice.

Peak sensitivity in the human visual system is at 555 nm, in normal to bright light but this shifts to 550 nm in lower but not Scotopic levels, such as found in an inspection booth for testing materials, using fluorescent penetrative inspection, at this wavelength.
https://www.nde-ed.org/EducationResources/CommunityCollege/PenetrantTest/Introduction/lightresponse.htm

Regarding the 589 nm wavelength used by A Marcus, for lens resolution: This is a fluorescence peak wavelength, in microscopy/biological applications: DyLight 549 555 nm (excitation) 569 nm emission but there are several other fluorescent dyes also in the 567-570 nm region but otherwise not particularly significant?
https://www.micro-shop.zeiss.com/en/de/shop/filterAssistant/dyes/

There are considerably greater number of the Long wavelength 'Red' sensitive cone cells present , in the human visual system, in the central fovea 2:1 relative to the Medium 'Green' with only 2% sensitive to short wavelengths 'Blue'. There are also rods (Purple-Blue-Green sensitive) in the 1.2 mm diameter fovea but not within the central high-resolution 0.3 mm foveola.

http://hyperphysics.phy-astr.gsu.edu/hbase/vision/colcon.html
http://hyperphysics.phy-astr.gsu.edu/hbase/vision/rodcone.html#c4

looking at the cone responses in the link above and particularly the ratio of Long to Medium cones, would it be reasonable to slightly shift the wavelength, used for lens diffraction/resolution calculations, towards the Long wavelength cone peak of 575 nm, for Human vision at least, although digital cameras using various Bayer CFAs are a different matter?

Going a bit off-topic: it is interesting that most of the short-wavelength receptive 'Blue' cones are actually not in the fovea and that resolution, in humans, for this region of the visual spectrum, is relatively poor ie. visual acuity falls by 75% at only 6 degrees off-centre! This matches well with the improved performance of the 10-degree 1964 CIE obsever, with colour matching, relative to the 2-degree case. OTOH, it doesn't seem to be well represented by the Bayer CFA, which more sparsely samples both blue and red, whereas the HVS is significantly lower resolution, only in the blue?
https://www.ncbi.nlm.nih.gov/books/NBK10848/

The eye also has a different type of photoreceptor, in addition to cones and rods, in the form of intrinsically-photosensitive retinal-ganglion cells. These are Blue sensitive and may play a role in human vision, including adaptation to light levels and possibly chromatic adaptation? They are most sensitive to light at around 480 nm (Berson, 2002; Dacey, 2005; Tu, 2005), significantly different (≥20 nm) from the best wavelengths for stimulating rods and cones

" In addition, there is emerging evidence that melanopsin contributes to visual perception, which is probably mediated by ipRGC projection to the LGN and/or SC (Dacey, 2005; Brown, 2010; Ecker, 2010; Estevez, 2012; Zhao, 2014). Some blind patients with severe outer retinal degeneration but relatively normal ipRGCs possess a rudimentary ability to detect the presence of intense blue light (Zaidi, 2007), and fully sighted humans as well as mice appear to depend partly on melanopsin for brightness discrimination (Brown, 2012). Mice lacking rod/cone photoreception (but with intact melanopsin photoreception) were able to distinguish a computer screen displaying black and white stripes from a uniform gray screen of an equal mean intensity, suggesting melanopsin is sufficient for a certain degree of pattern vision (Ecker, 2010). There is also preliminary psychophysical evidence that melanopsin directly contributes to color vision in humans, challenging the trichromatic theory (Horiguchi, 2013). Most recently, mice lacking melanopsin were shown to have behavioral deficits in contrast sensitivity (Schmidt, 2014).

IpRGCs are likely to perform additional image-forming visual functions because under conditions preserving synaptic input, primate ipRGCs receive color-opponent (blue OFF, yellow ON) input from cones (Figure 21B), implicating a capacity for color discrimination (Dacey, 2005)."

https://webvision.med.utah.edu/book/part-ii-anatomy-and-physiology-of-the-retina/melanopsin-expressing-intrinsically-photosensitive-retinal-ganglion-cells/

So perhaps digital cameras should have a 5 colour CFA, in order to better match the human visual system?

Cheers.

A Marcus Forum Member • Posts: 58
Re: Lens resolution - reference wavelength
2

Green near the chemical focus of panchromatic films. When focusing a film camera, their is a chemical focus and a visual focus. The film holder or position of film plane compensates.

alanr0 Senior Member • Posts: 2,237
Re: Lens resolution - reference wavelength
1

A Marcus wrote:

Green near the chemical focus of panchromatic films. When focusing a film camera, their is a chemical focus and a visual focus. The film holder or position of film plane compensates.

I agree that a green wavelength would be a sensible choice. I thought I made that explicit in my initial response and here.

My point is that 589 nm is NOT GREEN.

https://en.wikipedia.org/wiki/Color#Physics_of_color

There may be good reasons for choosing yellow/orange. I wondered why this might be, and whether it was common practice when measuring or specifying photographic lenses.

A cursory search found 500-560 nm green wavelengths quoted in analysis of diffraction limits on photography, for microscope objectives and by suppliers of technical lenses.

Many photographic lenses are limited by a combination of geometric aberrations and longitudinal chromatic aberration at their maximum aperture.  In such cases, performance with monochromatic light will not be representative, giving over-optimistic predictions.  This may explain why a single wavelength is not routinely specified.

-- hide signature --

Alan Robinson

J C Brown Senior Member • Posts: 1,620
Re: Resolution of 35mm Film & MTF 30?
2

alanr0 wrote:

grsnovi wrote:

I don't generally read this forum but was wondering (with all of the general interest in the other forums about sensor resolution in MP) if there is a generally accepted number representing the resolution equivalent of a 35mm film frame? B/W? v. color (slide). TIA. Sorry if this has been beaten to death previously - I admit to not attempting a search.

As Erik and Mark have already pointed out, the characteristics of film and digital sensors are very different.

You might find this document by Tim Vitale of interest: http://www.tmax100.com/../film.pdf. It discusses differences between halide particle size and grain size, cloud size in colour emulsions, and has tables comparing the resolutions of different emulsions.

Resolution, measured via Modulation Transfer Function (MTF) at 30% contrast varies from 35 lp/mm for Ektachrome 160 up to 160 lp/mm for T-MAX 100.

Kodak Ektachrome EKT 160 35 lp/mm
Kodak PORTRA 160VC 65 lp/mm
Fuji Velvia RVP 80 lp/mm
Kodak VR100 Color Neg 100 lp/mm
Kodak Plus-X B&W negative 125 lp/mm
Kodak T-MAX 100 B&W negative 160 lp/mm

If we take the rather simplistic view that we need 2 pixels for each line-pair, then equivalent pixel counts at MTF30 range from 4 - 22 Mp for slide film to 88 Mp for high resolution B&W film.

The number of sensor pixels required to resolve a line pair depends on the position of their edges in relation to the pixel boundaries. To be sure of resolving a line pair at any position on a sensor a value closer to 3 pixels per line pair is required.

For the examples quoted above the equivalent pixel counts at MTF30 range from 9.5 to 50 MP for slide film to 199 MP for high resolution B&W film.

Jimmy

-- hide signature --

J C Brown
Though it may not have been discovered yet there is a rational explanation for everything.

Mark Scott Abeln
Mark Scott Abeln Forum Pro • Posts: 14,156
It gets worse

J C Brown wrote:

To be sure of resolving a line pair at any position on a sensor a value closer to 3 pixels per line pair is required.

It gets worse!

For line pairs of any color combination and orientation, more like six pixels width per line pair are needed. And this is only in the case where the colors are otherwise rather distinguishable by the sensor and eye. Some raw converters routinely blur colors, so the line pair widths needed might be larger in some circumstances.

 Mark Scott Abeln's gear list:Mark Scott Abeln's gear list
Nikon D200 Nikon D7000 Nikon D750 Nikon AF Nikkor 50mm f/1.8D Nikon AF-S DX Nikkor 35mm F1.8G +2 more
Keyboard shortcuts:
FForum MMy threads