It's just words, words, words: Describing and interpreting film performance.

[Ian Grant]

I said something similar to the OP in his other large thread, posting these curves WITHOUT other data is not very useful, we need to know about grain and sharpness as well.

It's interesting that John Blakemore, probably one of the finest exponents of the Zone System, makes Adams look like a rank amateur. No Densitometry, rather practical experience. You have to see his work first hand to realise why . . . . . . .

But when testing films and developers these graphs may often be useless if gain and sharpness is compromised. We need the whole, not the half.

Ian

 

[Craig]

I said something similar to the OP in his other large thread, posting these curves WITHOUT other data is not very useful, we need to know about grain and sharpness as well.

The curves are for being able to determine exposure, development and contrast. Once those are correct, then you can evaluate other film characteristics. I see these curves as being the foundation to build on to evaluate if a film is suitable for it's intended purpose.

 

[runswithsizzers]

Thank you. Yes, that is the idea. I'd like to get closer to understanding how to marry purely quantitative descriptions of exposure, film (and paper), development with the purely descriptive accounts. I am willing to bet that each discerning photographer already has such a system in place. It may be intuitive, it may be based on prior experience, it may be the result of reading data sheets, it may be based on Zone System testing, etc., but such accounts only make perfect sense to that photographer, making generalizable descriptions difficult. There are countless examples of such descriptions in photographic publications that probably make perfect sense to the author, but much less to the reader.

For example, the idea of "lumpiness" came up in this thread and the one on tabular grain films. We can see how this "lumpiness" shows up in the characteristic curve, but we do not know why this happens. Even the very term "lumpy" probably has a different meaning to different people. The lumpiness can be described as both local and global non-linearities with the film's response to exposure and development. Perhaps it happens because of the multi-layered structure of the emulsion. It would be reasonable to assume that each light-sensitive layer has its own transfer function, and the overall response is some sort of aggregate, hence displaying lumpiness where the individual layers' responses cross over? This is pure speculation. It would be nice if a chemist could chime in, especially someone with experience with tabular grain films. The Delta films do not seem to have this characteristic, at least not to the same degree.

With simple emulsions, such as the Ferrania P30, it's possible to create a model that fits the data pretty well. With such a model, we can synthesize an entire family of characteristic curves (based on a set of criteria, such as development time, CI, Gamma, etc.) and have them look similar to the actual data (see the example below). However, with KODAK T-MAX emulsions, especially the P3200, it is harder to do.

ferraniaP30 by Nick Mazur, on Flickr
ferraniaP30Table by Nick Mazur, on Flickr

ferraniaP30S by Nick Mazur, on Flickr
ferraniaP30STable by Nick Mazur, on Flickr

Just looking at the bottom table, and ignoring the rest for a minute. For those of us still trying to figure out this kind of data display...

I pulled out my Henry Horenstein "Beyond Basic Photography..." and re-read his discussion of Basic Sensitometry. He defines Gamma, Subject Brightness Range (SBR), and Contrast Index (CI), and shows how each of those are determined.

But what what do "EFS" "G" and "LSLR" stand for? And what units are associated with the numbers in those columns? I could not find anything in Horenstein to define these cryptic terms. Google gives plenty of definitions, but they are all wrong.

I realize, every body of knowledge, especially if it is the least bit technical, soon develops its own specialized vocabulary and abbreviations. And some abbreviations are used quite often, so we can assume most people who are interested in the subject will be familiar with them. Even so, one cannot assume too much; I've noticed "MF Cameras" as mentioned on this forum, are almost always Medium Format -- but on another forum, it usually means Manual Focus.

So before I can even get to the topic of this thread, "...words, words, words..." -- I need to get past, "letters, letters, letters" - aka TLA hell.

 

[Craig]

EFS: Effective film speed, call it an EI (exposure index) rather than ISO, as there is a particular way to determine ISO speed.

G should actually be G with a bar on top, and is average gradient. The average gradient line can be shown to form the hypotenuse of a right triangle of which the base is the exposure range and the height is the density range. The value of the gradient is obtained by dividing the density range by the exposure range. It's basically the slope of the lines, ignoring the curved bit at the toe.

G(bar), CI and Gamma and all methods of determining the contrast of the film, but measured in different ways. It's unitless, since it's slope of the curve.

SLR: Subject luminescence range. Different from subject brightness range (SBR).

 

[runswithsizzers]

EFS: Effective film speed, call it an EI (exposure index) rather than ISO, as there is a particular way to determine ISO speed.

G should actually be G with a bar on top, and is average gradient. The average gradient line can be shown to form the hypotenuse of a right triangle of which the base is the exposure range and the height is the density range. The value of the gradient is obtained by dividing the density range by the exposure range. It's basically the slope of the lines, ignoring the curved bit at the toe.

G(bar), CI and Gamma and all methods of determining the contrast of the film, but measured in different ways. It's unitless, since it's slope of the curve.

Interesting. Thank you.

SLR: Subject luminescence range. Different from subject brightness range (SBR).

Are the SLR and SBR numbers shown here measured values or calculated? I know how to determine SBR in the field, but I'm guessing, here, it might a theoretical, calculated from the other data? And I assume the units are "stops"?

 

[Craig]

Interesting. Thank you.

Are SLR and SBR measured values or calculated? And I assume the units are "stops"?

Measured. SBR is stops, I think (but am not sure) that SLR is log exposure.

 

[jp498]

90% of what is verbalized about a film (such as what we see on youtube) in simply anecdotal to be kind, or opinion.

After processing, scanning/printing makes a huge difference that makes any anecdotal observations yet another step removed from reality.

My favorite film... tmax400, I've processed in
D76 1:1, 1:2, stock
Xtol 1:1
PMK 1:2:100
Pyrocat HD (normal and semi-stand)
Pyrocat HDC
Dektol
Caffenol
Add different temperatures and times for each, tray or drum/tank, different agitation options, and one film works a million different ways defying classification.
Add exposure/development changes for zone system if ones wants (I typically don't bother with tmax400)

Spectral response does make a difference in describing films and how they work. I shoot a lot of landscape with granite, and it behaves like Caucasian skin tones. If I use a filter or different film, I get different results... eg. more red sensitive films / less blue sensitive makes a big difference.

The HD graphs are partially useful. They can be somewhat changed by agitation and developer choices.

Grain I could care less.. I rarely print big and rarely use tiny film. Mostly I shoot 6x6cm and print no bigger than 10" square, or shoot large format and contact print. When I did shoot 35mm, it was typically printed 5x7 or 8x10.

 

[Sirius Glass]

Interesting. Thank you.

Are the SLR and SBR numbers shown here measured values or calculated? I know how to determine SBR in the field, but I'm guessing, here, it might a theoretical, calculated from the other data? And I assume the units are "stops"?

I use a spot meter when I want to know the SLR/SBR.

 

[Ian Grant]

The curves are for being able to determine exposure, development and contrast. Once those are correct, then you can evaluate other film characteristics. I see these curves as being the foundation to build on to evaluate if a film is suitable for it's intended purpose.

It's a bit like putting the chicken before the egg, and increasing development times can affect the sharpness and grain. I don't disagree about curves being useful rather that it's important to have evaluation of grain and resolution, as this is also determined by the film stock and then the choice of developer.

As @Sirius Glass indicated by posting the Kodak comparison chart it's the balance of speed, sharpness, and graininess that is often the first factors in choosing a film/developer combination, then you fine tune by altering the curve to suit your own approach.

Ian

 

[aparat]

Some of you have noted that sensitometric testing cannot reveal all of the important aspect of film (and paper) performance. I agree with @Ian Grant that a family of curves, in isolation, is of limited use to most photographers. What I also gather from this thread (and others on this forum) is that it is important to strike a good balance between testing and doing actual photography. For me, personally, it is not easy. I feel very uneasy about jumping into the deep end, at least not without ample preparation. I remember when the new CatLABS film came out, one of the posters said they were going to buy 200 rolls of the film and figure out the best way to work with it by taking lots of photographs. I could never work this way, even if I could afford 200 rolls of film, but I do understand that such an approach works for some people and I have no problem that, whatsoever.

I think that this quote from Todd (Elements of Black and White Photography, Amphoto Books, 2001) is very interesting in the context of this thread: "Other workers seem to devote so much time to testing different developers and techniques that they make me anxious, giving me the feeling that perhaps I am missing something. On the other hand, at my age, time disappears in a logarithmic manner, each day becoming a smaller percentage of the total. I would rather spend these precious moments capturing new images or learning to print better."

So I made a commitment to myself and started printing again. I made my first prints in five years this past weekend. It gave me such joy! I have been away from my darkroom due to various health issues, but I think I can slowly return to it. Having tested the materials beforehand made the process simple and strangely satisfying. It gave the numbers all the context I needed. I enjoyed that a lot.

 

[Chuck_P]

Congrats!

 

[warden]

So I made a commitment to myself and started printing again. I made my first prints in five years this past weekend. It gave me such joy!

Great to hear!

 

[aparat]

Here's another example of Ilford Delta 100. This time, in home-brew Pyrocat HD (in glycol). I used my usual sixty-minute semi-stand development that gives me very nice negatives shot in dull, overcast, winter light. Obviously, for greater subject luminance range (as in bright sunny conditions), one would need to cut development time. This is going to be nothing new to those of you who routinely develop in Pyrocat HD, but here it is, for what it's worth.

My semi-stand development is also very typical: continuous agitation during the first two minutes, followed by two gentle inversions every ten minutes thereafter, for the total of sixty minutes in developer (1+1+200). Pyrocat HD is known to produce full box speed, at least with Delta 100, and really nice-looking grain and some adjacency effect to boot. Even development is assured with this technique, at least in my experience.

Please, keep in mind that to get a more precise estimate of film speed (and other characteristics), I would need a few more curves of different contrast, so this is only a very rough estimate.

delta100pHD60min by Nick Mazur, on Flickr

delta100pHD60minScan by Nick Mazur, on Flickr

 

[aparat]

Rodinal and Pyrocat HD are used by a lot of photographers, particularly with stand and semi-stand development. There is a common belief in the photographic community that Rodinal, in a stand (or semi-stand) development, has a strong compensating effect, preventing highlights from becoming unprintable. The same holds for Pyrocat HD. In addition, both developers are believed to give pronounced adjacency effects. What we can see from the tests below is that Rodinal and Pyrocat HD have a very different response to development with Ilford Delta 100, and, hence, produce very different tonalities. Rodinal (Rollei R09 One shot) has no compensating action, and, instead, presents significant expansion in the highlights, i.e., contrast builds up really fast. Pyrocat HD, however, shows some highlight compression, even in the twenty-minute development. Both developers give very good film speed and excellent acutance, with nicely controlled graininess.

Given the different response to development, I would not consider these developers interchangeable. I can see why Pyrocat HD is so popular among landscape photographers, producing a smooth, nicely extended tonality, with good highlight protection. I enjoy using it from time to time, myself. I use Rodinal semi-stand exclusively for negatives shot in dull, flat light, to give them more sparkle, so to speak.

My Recipe for Pyrocat HD: continuous agitation for the first two minutes, with two gentle inversions every ten minutes. For Rodinal, I use continuous agitation for the first two minutes, with two gentle inversions at the half-way point. I dilute Pyrocat HD 1+1+200, and Rodinal 1+100.

delta100_pyrocathd60 by Nick Mazur, on Flickr
Ilford Delta 100 in Pyrocat HD, 60 min. semi-stand by Nick Mazur, on Flickr

delta100_Rodinal100 by Nick Mazur, on Flickr
delta100_Rodinal100Scan by Nick Mazur, on Flickr

 

[albireo]

For me, personally, it is not easy. I feel very uneasy about jumping into the deep end, at least not without ample preparation. I remember when the new CatLABS film came out, one of the posters said they were going to buy 200 rolls of the film and figure out the best way to work with it by taking lots of photographs. I could never work this way, even if I could afford 200 rolls of film

It was me who said that. And I stand by it. And you see, the two approaches are not mutually exclusive. You are twisting the '200 film' point I made to perhaps imply I was going to go out and blindly shoot 200 rolls and try to get somewhere with what I saw. Poor form on your side. The way I see it, you should do theoretical preparation AND extensive field testing of the material in your own workflow.

Unless of course one enjoys more the desk work than the field work side of photography - which is fine of course.

 

[faberryman]

Tell me about this “extensive field testing” you plan on doing with your 200 rolls of film.

 

[aparat]

I have another quote that I think is worth thinking about in the context of this thread. As I mentioned, it's never a question of right or wrong; it's all about what works for each photographer. Roger Hicks and Frances Schultz (The Black and White Handbook, David & Charles, 1997, p. 18) say the following:

"Photography is at once an art, a craft, and a science. These three aspects can work together, or they can be at conflict. Those who follow only Art (you can often hear them pronounce the capital “A”) will all too often disdain anything so mundane as technique: you can recognize the prints they produce by their flat, muddy tonal range, liberal sprinkling of dust spots, and dog ears. On the other hand, whose who prize craft above all else may turn out prints which are technically superb, but utterly devoid of content or aesthetic merit; at best they merely repeat hackneyed pictorial formulae. And those who cleave only to science may never shoot real pictures at all; instead, they endlessly photograph test charts to test their lenses, and calibrated grey cards and grey scales to test their films and developers."​

And they offer the following resolution to this apparent conflict: "The trick lies in taking what you need from all three areas."

I cannot help but feel that the authors may be on to something here. The question is what the "what you need" part is and how we can use it to our advantage.

@albireo I sincerely apologize if you think I twisted your words. Please, see my statement in the context of this thread. We are discussing how best to evaluate exposure, film, development, etc., to suit our needs. The thread is about the richness of approaches, not about disparaging anything or anyone.

 

[faberryman]

And they offer the following resolution to this apparent conflict: "The trick lies in taking what you need from all three areas."

I cannot help but feel that the authors may be on to something here. The question is what the "what you need" part is and how we can use it to our advantage.

Sometimes the obvious bears repeating to get people back on track. Thanks.

 

[aparat]

I think I did not make my point clear when comparing Rodinal and Pyrocat HD. Here's a more detailed comparison.

Rodinal stand and semi-stand development has been made famous partly due to its alleged compensating effect. If that were true, we would see significant highlight compression, relative to the mid-tones, in Zones VII and VIII. This does not happen, in my experience. I have seen this not just with Ilford Delta, but also with Fomapan 100, FP4 Plus, and other films. Ilford Delta 100 is a good film to test these types of effects because of its near-linearity in conventional developers, such as XTOL and D-76.

In the plot below, I circled Z VII and VIII (the highlights), and you can see that their relative "width" is greater compared to the mid-tones. By the way, I used a "useful negative density range" of 1.2, which is recommended by Lambrecht and Woodhouse (Way Beyond Monochrome, 2011). I typically use the value of 1.05 because it works best with my process, but these values are meant to be flexible.

delta100_Rodinal100DensityRange by Nick Mazur, on Flickr

And here's the thirty-minute curve in Pyrocat HD. You can see highlight compression, evidence of a "compensating" effect.

delta100_pyrocathd60ensityRange by Nick Mazur, on Flickr

 

[aparat]

@faberryman I agree. However, I have been guilty of forgetting how to strike that balance.

 

[snusmumriken]

It isn’t only “what you need”, it’s also “what you can practically employ to your advantage”. I suspect most of you are LF or MF shooters, and that your subject matter doesn’t run away. If one shoots 35mm, is attracted to fleeting or ephemeral subjects, and rarely uses an entire roll on the same subject or in the same lighting conditions, the opportunity to make best use of technical knowledge isn’t there. I am deeply interested in acquiring that knowledge (thank you, @aparat!), and absolutely love a photo that has got beautiful, satisfying tonality. But in practice I have to select a film that will suit a range of likely circumstances, and process it to suit the most hopeful shots.

 

[snusmumriken]

I think I did not make my point clear when comparing Rodinal and Pyrocat HD. Here's a more detailed comparison.

Rodinal stand and semi-stand development has been made famous partly due to its alleged compensating effect. If that were true, we would see significant highlight compression, relative to the mid-tones, in Zones VII and VIII. This does not happen, in my experience. I have seen this not just with Ilford Delta, but also with Fomapan 100, FP4 Plus, and other films. Ilford Delta 100 is a good film to test these types of effects because of its near-linearity in conventional developers, such as XTOL and D-76.

In the plot below, I circled Z VII and VIII (the highlights), and you can see that their relative "width" is greater compared to the mid-tones. By the way, I used a "useful negative density range" of 1.2, which is recommended by Lambrecht and Woodhouse (Way Beyond Monochrome, 2011). I typically use the value of 1.05 because it works best with my process, but these values are meant to be flexible.

delta100_Rodinal100DensityRange by Nick Mazur, on Flickr

And here's the thirty-minute curve in Pyrocat HD. You can see highlight compression, evidence of a "compensating" effect.

delta100_pyrocathd60ensityRange by Nick Mazur, on Flickr

Worse, the Rodinal has severely compressed shadow and lower mid-tones as a result. Presumably it would appear either ‘muddy’ or starkly contrasty, depending how it was printed?

But isn’t it essential to incorporate the characteristic curve of the printing paper in its developer? Otherwise it’s impossible to state what the ideal film curve should look like.

 

[aparat]

Worse, the Rodinal has severely compressed shadow and lower mid-tones as a result. Presumably it would appear either ‘muddy’ or starkly contrasty, depending how it was printed?

But isn’t it essential to incorporate the characteristic curve of the printing paper in its developer? Otherwise it’s impossible to state what the ideal film curve should look like.

Yes, you are correct. Thank you for pointing it out. I am a bit disappointed with Rodinal's performance. I was not expecting this type and degree of curve distortion. Last summer, I tested Rollei R09 One Shot and Adox Rodinal to see if they produce similar results, and, in this type of semi-stand development, they basically do. I'd have to test this further, with other films, to be sure. My previous bottle of Rollei went bad after about three years, if I remember correctly so I bought a bottle of Adox Rodinal when it was back in stock hoping that it has the same shelf life as Agfa Rodinal.

adox_rodinal_rollei_R09PFomapan100 by Nick Mazur, on Flickr

Here's something kind of interesting. Fomapan 100, in a conventional developer, such as ID-68 (a.k.a., Microphen), has an s-shape curve, but Rodinal manages to make it straight, so to speak, presumably, due to its highlight expansion.

fomapan100_id68 by Nick Mazur, on Flickr

I also agree with you that it would be ideal to have the paper curves as well. Unfortunately, I did my most recent paper test a while back, before the current Ilford MG emulsion revision, so this plot is not useful, but I think it can be used to show what you found, namely, how the shadows tone compression shows up in the print, or I should say, a print. This is just one of several possible ways of plotting this data set. Here, I chose to focus on fitting the shadows onto the paper, but the data could also be aligned to the highlights or mid-tones, reflecting possible creative decisions of the print maker. I am still not happy how I am plotting this, so I need to revise the code and my approach somewhat, but, crudely, I think it does show how the film and paper data can be matched.

toneReproductionIlfordDeltaRodinal by Nick Mazur, on Flickr

 

[aparat]

Here are more words of wisdom. These are from Ansel Adams himself (The Negative, New York Graphics Society, 1948, p.51):

"There is nothing more destructive to a creative approach than the domination of 'the perfect negative' or 'the standard negative' - unless such standard be determined by the individual in relation to his own concept. The time-gamma charts have a definitive comparative value in relation to my approach, chiefly in regard to temperature variations of the developer."​

This one struck me as a bit odd, since the entire book seems devoted primarily to creating the perfect or the standard negative. Perhaps I am just taking this quote out of context?

Ansel Adams has an interesting take on the whole concept of describing film and paper performance. To him, a simple description such as "7 minutes at 68F in Ansco 47 for Isopan represents normal to me" contains everything he needs to know. Interesting. I tried something like that a long time ago, but my target for normal kept changing, leaving such descriptions somewhat inadequate, long-term.

 

[Melvin J Bramley]

Perhaps the whole idea is produce a good negative, not necessarily a perfect negative.
The pursuit of graphs is neither artistic nor practical for roll film users.



TB