Film granularity/LPM chart

Paul Verizzo · Jul 22, 2008

I put this together for my own edification and reference.

I believe in trying to get objective information whenever I can in life. Such info may not be the whole story, but it's a good place to start. I don't put much stock in individual's observations, so many variables and skill sets are in play, to say nothing of meaningless descriptors.

Ilford does not release this type of information, although I found RMS granularity for the FP4+ movie film.

If Fuji is to believed, they have the best films for a given speed bar none. For any given film speed, they claim quality equal to everyone else's one speed lower. Frankly, I don't believe that they alone hold great film making secrets. In fact, I garnered enough information on the Acros rating to see that they used a proprietary developer - probably never seen on these shores - and then rounded down.

RobC · Jul 22, 2008

I read recently that engineering types say Lines per millimeter (l/mm or lpmm) whereas photographer types say Line Pairs per millimeter (lp/mm or lppmm).
However, when engineering types say lines per millimeter, they usually mean line pairs per millimeter so 60lpmm = 120 lines.
Confused? Well I am because you never know what the quoted numbers actually mean if the don't say lppmm. Do they mean 60lpmm or 60lppmm?

df cardwell · Jul 22, 2008

Good job, Paul. It is hard to boil it down to a spreadsheet.

EKC is correct and forthcoming, giving you the criteria for granularity tests,
they use normal developers, and not always the ones which will give the highest numbers but ones which will likely be used in normal practice. Finally, they define the contrast level for their resolution values.

Look at the Neopan 400 data. Where, exactly, was that from ? Fuji does not have that information on their data sheets.
Equal granularity, and similar resolution to TMY ? Right.

As for Acros, Fuji's data rightfully equates it to TMX; it is noteworthy that EKC got their numbers from D-76, Fuji from their Microdol X equiv. As you point out: anything to get less granularity than Kodak.

The tradition is to publish the source of the data. Without manufacturers' help it is an impossible task. The best thing to do with scanty data is to not split hairs over unsourced data, Acros & TMX being a good example. Especially if you are going to use a different developer than produced the data.

Curve shape and color sensitivity are more important to the image anyhow. Good job.

Paul Verizzo · Jul 22, 2008

Apples to apples

rob champagne said:
I read recently that engineering types say Lines per millimeter (l/mm or lpmm) whereas photographer types say Line Pairs per millimeter (lp/mm or lppmm).
However, when engineering types say lines per millimeter, they usually mean line pairs per millimeter so 60lpmm = 120 lines.
Confused? Well I am because you never know what the quoted numbers actually mean if the don't say lppmm. Do they mean 60lpmm or 60lppmm?

As long as they are all using the same standard and technique, the precise definition and terminology doesn't matter. And I do believe that they are. It's old, established science.

Paul Verizzo · Jul 22, 2008

Cardwell; good observations, all so true. I believe that I found the Fuji info on the film data sheets. I believe the inferences about Fuji "cheating" was found somewhere on the net. Since this was for my own use, I didn't keep everything footnoted. But, hey, trust me!

Good points on Kodak using real world developers. There's only one reason anyone would use a strong solvent developer on a slow T-grain film and that is to fudge the specs. No one in their right mind would do it in the real photo world. It wouldn't surpise me that the used a different developer (dilute D-76?) for the sharpness determination.

Looking at the now deceased APX films, one can only pine for their best of both worlds ratings. Unlike Fuji, their ratings are more believable, I think.

cmo · Jul 22, 2008

Let's add something to this...

aic.stanford.edu/sg/emg/library/pdf/vitale/2007-04-vitale-filmgrain_resolution.pdf

Page 11 and 14 show data.

I have some german documents. Sorry...

A while ago, I found a document by Zeiss, listing the fims that they use for testing lenses, from 2003:

http://www.zeiss.de/C12567A8003B8B6F/EmbedTitelIntern/CLN_17d/$File/cln17d.pdf

Result:
Since 1999 they use Fuji Velvia as a reference and confirm 160 LPM, which is what Fuji says.
In Black-and-White they stocked up with Agfa APX 25 before its production ended. They confirm 200 LPM, which is what Agfa said.

In the meanwhile they must have switched to a new film: Spur Orthopan.

http://www.zeiss.de/c12567a8003b0478/contents-frame/d366ef465fba5e31c125711d0019b0e8

Result:
SPUR Orthopan UR, resolution 400+, higher than Kodak Imagelink HQ with 320 LPM. They say that 400 LPM is the absolute limit of any lens at aperture 4, a final frontier caused by diffraction.

What else:

Efke 25 has 115 lpm
ROLLEI INFRARED IR400: RMS 11, 160 lpm
Rollei ATP: 300 lpm

Paul Verizzo · Jul 22, 2008

cmo, great research and thanks for the info! I'm going to add it to my chart.

cmo · Jul 22, 2008

Maybe it is a good idea to ask the manufacturers, too? They must be interested in objective information...well, as long as they are not from China and as long as we don't raise an issue like doping

After that, why not make a permanent link here and publish it on sites like digitaltruth.com?

Paul Verizzo · Jul 22, 2008

cmo said:
Maybe it is a good idea to ask the manufacturers, too? They must be interested in objective information...well, as long as they are not from China and as long as we don't raise an issue like doping

After that, why not make a permanent link here and publish it on sites like digitaltruth.com?

All but a few of the specs I listed came from manuafacturer's info sheets. A few were second generation, I presume they found something once upon a time somewhere.

Ilford, as noted, refuses to supply such info. I understand, and yet I don't. I'm sure that they have nothing to be ashamed of, and most people accessing this type of information understand that it's only a starting point.

cmo · Jul 22, 2008

Paul Verizzo said:
I'm sure that they have nothing to be ashamed of, and most people accessing this type of information understand that it's only a starting point.

My dream is a database of all films, showing a standardized example, the film exposed and developed according to the manufacturers' manuals, scanned in a standardized way on a high-end drum scanner and showing small crops of the images on a website. To make it really complex, varieties of the setup using other developers are a must So, everyone could get a good impression instead of listening to repeated advertising BS... but it's the work of a lifetime.

df cardwell · Jul 22, 2008

As much as I am a recovering tech-geek,
the photographic process buffers the granularity of any film and developer combination.
Handled well, there is not much difference between any ISO class of film, certainly not enough to
be reason to choose, for instance, FP4 over Plus X.

The REAL differences are three-fold:

1. Color response. There are subtle differences between FP4 and PLus-X. And a vast difference between them and the pseudo-ortho signature of APX 100. If you are shooting outdoors, FP4 will just about spare you a light yellow filter to record a toned sky. APX 100 is very sensitive to blue, and getting shadow and sky at once is... well, that's why I have so much experience flashing skies ! But for one film that flatteringly records every complexion of our planet, APX 100.

2. Film curve. FP4 and TMY have never-ending straight lines. Plus X does not. Pick your subject and your desired image, then pick the film that does the dirty work FOR you.

3. Availability. If I can't get it, or worse, if I think I can get it but CANNOT, it is no use to me. Even if it grainless 1000 speed film.

Alan Johnson · Jul 22, 2008

Thomas Wollstein (I believe he is a well known writer on photography) gives some data from Ilford Germany:
http://www.photo.net/black-and-white-photo-film-processing-forum/002Z9E

Paul Verizzo · Jul 22, 2008

Don't forget...

cmo said:
My dream is a database of all films, showing a standardized example, the film exposed and developed according to the manufacturers' manuals, scanned in a standardized way on a high-end drum scanner and showing small crops of the images on a website. To make it really complex, varieties of the setup using other developers are a must So, everyone could get a good impression instead of listening to repeated advertising BS... but it's the work of a lifetime.

Different developer temperatures, right hand vs. left hand twists of rotation, stainless vs. plastic .......

Paul Verizzo · Jul 22, 2008

Thanks, more enlightened observations

df cardwell said:
As much as I am a recovering tech-geek,
the photographic process buffers the granularity of any film and developer combination.
Handled well, there is not much difference between any ISO class of film, certainly not enough to
be reason to choose, for instance, FP4 over Plus X.

The REAL differences are three-fold:

1. Color response. There are subtle differences between FP4 and PLus-X. And a vast difference between them and the pseudo-ortho signature of APX 100. If you are shooting outdoors, FP4 will just about spare you a light yellow filter to record a toned sky. APX 100 is very sensitive to blue, and getting shadow and sky at once is... well, that's why I have so much experience flashing skies ! But for one film that flatteringly records every complexion of our planet, APX 100.

2. Film curve. FP4 and TMY have never-ending straight lines. Plus X does not. Pick your subject and your desired image, then pick the film that does the dirty work FOR you.

3. Availability. If I can't get it, or worse, if I think I can get it but CANNOT, it is no use to me. Even if it grainless 1000 speed film.

Yes, within a speed family and grain type, and probable fudging by Fuji (how's that for alliterration?), there is little difference. One exception is Fomapan 100, noticably more grainy than most of the others. Still, a nice film and at a rock bottom price, especially as the Arista variant.

I took some shots on the new TMY, including the various B&W filters I have. Didn't do any manual compensation, just let the camera do its thing. I have to tell you, the one through the blue filter made the most stunning negative! I haven't tried to print any yet, but it just jumped out at me, beautiful contrasts and tones.

cmo · Jul 22, 2008

Paul Verizzo said:
Different developer temperatures, right hand vs. left hand twists of rotation, stainless vs. plastic .......

Tab water vs. Perrier, sure

Well, not really, just one standardized method, of course.

Philippe Grunchec · Jul 23, 2008

Paul, which blue filter did you use? 47?

Paul Verizzo · Jul 23, 2008

Philippe Grunchec said:
Paul, which blue filter did you use? 47?

An 80A.

Photo Engineer · Jul 23, 2008

Paul;

After due consideration I have to add that Kodak gets Lines per by analyzing positive and negative images on a film to guage bloom and fill in, two characteristics that are typical of resolution. In bloom, a dense black spreads into a white surround, and in fill in, a dense black background expands into highlight areas. Lines per is some function of both neg and pos lines.

These two factors must be known to determine response from LPM.

For RMSG, the diameter of the reading spot must be known so that there is no aliasing of the grain. For example, if you pick a diameter of the spot larger than your grain, the grain appears lower than it actually is, because the spot is averaging two grains. You must have a diameter much smaller than the grain. If you do not, then you have offset your data (aliased it) from the true value.

Also, with grain, there is instrument noise. We determined this by casting pure dye with no grain. It was a uniform background, but we picked up a rather unexpectedly large grain value due to the noise of the instrument reading such a small size and at the density levels required.

So, LPM are not really LPM depending on how you do it, and RMSG is not RMSG depending on how you do it. One of the raging debates in dye stability is whether to use 100 FC or 500 FC to fade color images. You get totally different answers depending on test method.

PE

cmo · Jul 24, 2008

Now, how could we make a common measuring unit? If every manufacturer has its own method we compare (green) apples and (yellow) bananas.

RobC · Jul 24, 2008

Paul Verizzo said:
As long as they are all using the same standard and technique, the precise definition and terminology doesn't matter. And I do believe that they are. It's old, established science.

Just to confirm, where you say for example, Fuji Across 200 lines per millimeter, thats means 200 line pairs per millimeter. Yes?

Paul Verizzo · Jul 24, 2008

I GUESS SO!

rob champagne said:
Just to confirm, where you say for example, Fuji Across 200 lines per millimeter, thats means 200 line pairs per millimeter. Yes?

Further research is up to you.

RobC · Jul 24, 2008

Paul Verizzo said:
Further research is up to you.

I'll take that as a don't know then.

df cardwell · Jul 24, 2008

On Manufacturer’s Documentation and The Photographic System

Kodak has a tradition of publishing accurate and copious data to enable a photographer to get the most from EKC materials, and in the process, learn a great deal about photography.

Ilford's documentation has never been directly comparable to Kodak.

Fuji does everything possible to look better than Kodak.

...................................

Film classes give similar performance.
If you can make a picture with Tri X, you can make it with HP5, or Neopan 400.

There are differences in curve shape, and how the films render color,
but granularity and resolution are essentially the same.

In this regard,

TMY = Delta 400

TMX = Delta 100 = ACROS.

If you see a difference between films’ grain,
review your processing; you probably induced the difference.

Remember that Photography is a System.
Using a perfect camera, a perfect lens at its perfect aperture,
and printing perfectly through your perfect enlarger,
increasing the resolution of your film from 100 to 200 lines/mm gives a net gain, on the PRINT, of 17% ! ( More or less).

For example, the differences between using TMY instead of ACROS
come down to an imperceptible resolution loss,
a small increase in granularity,
and a significant reduction in potential image blur because of a shorter exposure.

Without the systematic context, no true comparisons can be made.

sanking · Jul 24, 2008

rob champagne said:
Just to confirm, where you say for example, Fuji Across 200 lines per millimeter, thats means 200 line pairs per millimeter. Yes?

This can be very confusing. One of the issues is that there are at least two charts commonly used to test resolution, the PIMA/ISO resolution target, and the USAF 1951 target. The first expresses resolution in line pairs per millimeter (lp/mm, or LPPM), the second in lines per millimeter (l/mm, or LMM). Just understand that numerically lines per millimeter (L/mm) is double line pairs per millimeter (LP/mm). 100 lines per millimeter (l/mm) is 50 line pairs per millimeter (lp/mm).

Kodak uses the term "lines per millimeter" when discussing resolution.

Sandy King

RobC · Jul 24, 2008

sanking said:
This can be very confusing. One of the issues is that there are at least two charts commonly used to test resolution, the PIMA/ISO resolution target, and the USAF 1951 target. The first expresses resolution in line pairs per millimeter (lp/mm, or LPPM), the second in lines per millimeter (l/mm, or LMM). Just understand that numerically lines per millimeter (L/mm) is double line pairs per millimeter (LP/mm). 100 lines per millimeter (l/mm) is 50 line pairs per millimeter (lp/mm).

Kodak uses the term "lines per millimeter" when discussing resolution.

Sandy King

Exactly, so when Fuji says Acros is 200 l/mm it means 100 line pairs per millimeter.( I think )

Luminous landscape has some info on this but I don't know if its correct.

http://www.luminous-landscape.com/tutorials/understanding-series/understanding-mtf.shtml

Film granularity/LPM chart

Attachments

Privacy & Transparency

Privacy & Transparency