Need help using zone system with film with 19-stop dynamic range

[Helge]

One important distinction is also that film, due to the semi stochastic nature and buildup curves of exposure, result in a much more graceful decline or round-off towards the start and the end of the recording ability.
Both with regards to the temporal effects but also the spatial.
This is a distinction it shares, to various degrees, with other media.
Not with photo-sensitized semiconductors though.

The human senses vastly prefer a gradual seemingly random falloff rather than an abrupt stop.

Though that abrupt stop might in short term tests be preferred. Just like in the legendary Pepsi challenge, and the resulting New Coke the human senses react positively to overstimulation in the short term, but quickly tire/figure out “the fraud”.

Same goes with for example speakers. Where most people immediately prefer a bass reflex speaker with a high cut but loud bass hump, as opposed to the gradual roll off of a horn or transmission line that digs lower into the frequency range.
But even after a few minutes almost everyone universally prefer the gradual roll-off of the deeper bass.

 

[Helge]

I recall that Delta 100 & TMax 100 do vary a bit more in HC-110 & Rodinal curve-wise (or at least TMax does - I recollect Delta stays much the same) - but I think a lot of the difference people think they're seeing is really due to the shadow speed difference that can exist with those two films & some developers.

Which is really to a large extent what speed is. Midtones and highlights you can always push-process to smithereens, but shadows is what marks true speed.

 

[Alan Edward Klein]

I shot a roll of Tmax 400 on 35mm with built-in metering. (Nikon N6006) Camera set at box speed ISO 400. Since this is a film camera, there's no LED screen, of course, to see exposure other than lining up the exposure needle. I pick an "average" area when determining exposure. I don't develop at home. I sent it out to a lab to develop normally. I scan at home keeping the histogram limits at the edges of the captured image. There was very slight clipping on some shots which I got limited during post-scan editing and the scanner seems to pick up enough at both ends. The film seems to have a wide enough latitude to me.

I'm not an expert on BW photography but Tmax 400 seems to do a nice job with many stops being covered and very nice tones. Minimal grain although I like Tmax 100 for landscape shots when I'm shooting on a tripod.

Tmax400, 35mm: https://www.flickr.com/photos/alanklein2000/albums/72157716777378896

 

[Lachlan Young]

Certainly possible they behave a little/slightly differently in Rodinal or HC-110. When I did all this stuff I benchmarked curve shapes and ISO speed with D-76/ID-11 and XTOL at various dilutions (turns out the manufacturers are correct - surprise, surprise). I used other developers along the way for comparison, but not Rodinal specifically.

As for people seeing things... I’ll stick with sensitometry when it comes to sensitometry.

From what I remember reading years ago about Ilford's design aim for Delta 100, I think they wanted to make a more consistently behaved film across all/ almost all developers (ie everything to behave like ID-11/ D-76) - rather than TMax 100 which can swing highlights up quite hard with some specific developers like HC-110 (great for specific purposes, but not for the unwary processing their film with Ansel Adams' words taken as scripture) - however as you & the manufacturers show, under 'normal' usage in D-76/ Xtol they are sensitometrically very close. HC-110 was a brilliant solution to quite specific problems, but whether it's a truly useful general purpose developer is a different question...

 

[Lachlan Young]

When I tried it on Delta 100 it caused an anomaly I could not explain and could not get rid of. Granted that is an extreme case, but anyway.

What did it do? I'm fairly sure the Delta films all use one degree or another of epitaxial structures (and TMax doesn't), so I wonder if it was an interaction with that?

 

[Alan Edward Klein]

What did it do? I'm fairly sure the Delta films all use one degree or another of epitaxial structures (and TMax doesn't), so I wonder if it was an interaction with that?

What are epitaxial structures? Are they harder to scan than Tmax t-grain?

 

[DREW WILEY]

Helge - Great analogy! But what you forgot to mention is that when you stretch a rubber band too far it snaps!

Michael, Lachlan - ironing out the TMX curve developer-wise is something I am very skilled at, but don't want to get into all the details here. It is remarkably developer-ubiquitous, but with somewhat different results at significantly different levels of anticipated gamma. I simply don't have a comprehensive enough background with Delta to make the same range of comparisons, but have learned enough to make Delta a "Choice B" film if somehow TMax was gone, at least for general shooting. But I do know one person who substituted it for TMax100 in his custom 8X10 film-recorder for sake of color separations, wanting lower film price. But the only way he could make the two varieties of curves match is by reprogramming the whole nine yards and digitally restructuring the respective pairs of RGB curves. It would be a near-hopeless task in a darkroom. But that level of precision is way beyond what ordinary black and white photographers need. I certainly like the exact spectral response and faster speed of TMX better.

 

[Lachlan Young]

I’ll have to show you the curves but basically it caused an unacceptable bump/shift in the curve at a certain exposure level. It looked like two curves spliced together and it just would not go away. I had a brief discussion about it with Ron at the time but I can’t remember coming to any conclusion, and since I had intended the developer to be a special purpose thing specifically for TMX and TMY-2 (mostly “optimized” for TMY-2 as far as I went) I simply considered it a failure with Delta 100 and didn’t pursue it further. It seemed almost like what one might expect from say a two-layer film that hates a higher than normal Phenidone concentration, but as far as I know Delta is a single layer emulsion, and even if it had two layers I would still be at a loss to explain the problem.

This is an extreme case though.

As far as I know, Delta 100 is single layer, but probably 2+ blended emulsions with epitaxy, whereas TMY-2 is definitely 2-layer, one layer tabular grain, the other 3D grain structures (number of emulsions in each layer unknown) - I wonder if it related to some sort of inductance period/ diffusion issue in terms of getting one of the emulsions to react at the same rate as the other in the specific developer environment (possibly in 'normal' conditions, byproducts are intended to be released to regulate one emulsion component to match the other?) - in his book, Bob Shanebrook states that TMX had to have an extra thick top coat applied to regulate diffusion rate in order to keep development times from being far too short to be controllable.

 

[Lachlan Young]

What are epitaxial structures? Are they harder to scan than Tmax t-grain?

At the simplest level it's a method by which you can grow further crystals on to the points of one you've already grown - it has a range of potential benefits, but essentially relates to the usual attempts to improve the sharpness/ granularity/ latitude/ speed relationships. There are threads on here that cover epitaxy in emulsions. Fuji and Ilford seem to be the main users of the technique in non medical/ industrial photographic materials currently - but from what I recall reading, under practical use conditions 2-electron sensitisation/ layered dye techniques offered (at the time) about equal boosts in speed/ grain etc overall - the next step would be to combine them. Finer, sharper grain will improve scans - but only up to the quality ceiling of the scanner.

 

[DREW WILEY]

Still don't get it, do you Michael? You can't match curves characteristics at box speed. It takes an entire stop more of exposure to essentially launch Delta 100 off the toe onto the straighter line portion of the curve. Therefore, for all practical purposes, I consider it an ASA 50 film. But most Ilford films are overoptimistic about their rated speed. I don't give a damn about industry conventions, especially when it involves interpolated curves rather than detailed ones. I do care about the real-world implications in terms on rendering shadow detail in the print in high contrast scenes, and specifically how the toe performs. In more subdued lighting, the rules inherently change.

I often need a film flexible enough for a very wide range of lighting conditions, because that's what I actually encounter, often on the same day. Perhaps you will be able to visit here someday and personally realize how drastically the lighting can change in a short period of time depending on atmospheric conditions, especially the effects of fog. The same applies to an even greater extent at high altitude in the mountains. At 11:59 AM the light is harsh and extreme. Fifteen minutes later the sky is black and it starts snowing, and the lighting ratio goes soft. Then the sun comes out again, if you haven't died of either hypothermia or a lightning strike in the meantime (which will be one thing lending way more than 19 stops to your film when it hits your camera too! Fantastic DMax or BBQ, whatever you want to call it).

Don't get me wrong - I love FP4, HP5, ACROS, etc etc. But for handling extremes, the longer and straighter the line the better. And now with Super-XX gone, as well as Bergger 200 (both of which were too grainy in my opinion for formats smaller than 8x10 anyway, and with Foma/Arista 200 having poor quality control and even worse long-exposure characteristics, that makes the two T-Max emulsions the conspicuous next best, and even better overall if smaller cameras are in mind. Like I said, Delta is a distant cousin, a Plan B film, for my own needs. For others, encountering different conditions or with different expectations, they are indeed welcome to play by their own rules.

 

[Craig75]

if you are just recording on the straight part of the film then then you are printing a "distorted" reproduction.

nothing wrong in that but there is no need to get off the toe unless you want the shadows forced apart more than highlights

 

[Stephen Benskin]

if you are just recording on the straight part of the film then then you are printing a "distorted" reproduction.

nothing wrong in that but there is no need to get off the toe unless you want the shadows forced apart more than highlights

All photographs are a distortion. According to tone reproduction theory, a preferred photograph can tolerate compression of the highlights and shadows as long as the midtones of the reproduction curve is greater than a 1.0 gradient (usually 1.1 - 1.2) or slightly higher. The reproduction curve is derived from comparing the original scene to the print. Film curve shape isn't the only influence. There's camera and enlarger flare and the paper curve. Plus viewing conditions such as illuminance, the surrounding tones around and within the image and their influence with simultaneous contrast, local inhibition and local adaption.

If you had a subject luminance range of 5.7, the flare factor would be rather large reducing the image luminance range on the film by compressing the shadows.

 

[Stephen Benskin]

Not that this is going to help, but below is one comparison of TMX with D100 (XTOL developer). Virtually identical.

I don't know Michael, proof usually works for me.

 

[Stephen Benskin]

Me and Drew have been arguing about certain films for years. It’s tradition

But as Lachlan pointed out earlier, there are circumstances under which two films with otherwise similar curves might indeed behave a little differently. Drew’s special (classified) developers might do this.

That's one of the main reasons for the change to the 1993 b&w film speed standard. The older standard's developer didn't reflect real world use of T-Max films.

I remember one of the two emulsions of the old version of TMY had a very different sensitivity to, I believe, red light. When shooting with a red filter, the emulsions would spit creating a flat middle but overall good overall contrast to somewhat contrasty overall. Images tended to look flat and contrasty at the same time.

 

[DREW WILEY]

I don't have any experience with Xtol. But I do know what people tend to perceive as very small differences in the toe can have a very significant difference indeed in real world applications. Want proof -Stephen - I've got literally hundreds of prints that wouldn't be what they are if I had chosen Delta, or FP4, or even ACROS instead, and certainy not HP5. How do I know that? - because I've done many many prints with those too. So which kind of proof is more substantial - a crude graph a few inches wide, or the end result? And yes, I do indeed know how to use a densitometer; yes indeed.

Now why do I emphasize what constitutes the toe threshold, especially when Zone System gurus like Barnbaum thinks the kneecap is the toe? Well, it's helluva lot easier to rein in a bit of shoulder than retrieve very subtle gradations on the toe. Worse, you can't recover density or texture or content that's not even there to begin with due to underexposure. Now the Zonies would preach, elevate your toes clear up to the belly button and then put the whole nine yards through a pasta roller to get it flat enough to print. No wonder so many of those prints themselves look somewhat flat with anemic sparkle. Why not do like me, and if there's significant contrast in the scene, choose a film with the longest straight line possible (all other things being equal, which they never are; but one has to prioritize).

I can even see the difference in the print way way down in the shadows between old school "straight line" films like Bergger 200 and the present contender, TMY or its brother, TMX. TMY just might be the "best" overall film ever for we LF junkies - fast speed, very fine grained combined with excellent edge acutance, superb quality control, long straight line with a short toe, great expanded high-gamma performance (or conversely, especially good low-contrast performance if you know a few tricks), especially cooperative spectral sensitivity, and realistically usable in small formats as well. TMX is similar except for obviously slower speed and finer grain; but it has disappointing edge acutance in most developers, but thankfully, not all. Yet neither of these films will dig as deep down into the shadows as the good ole 200 straight-line ones. Everybody knew that once. Now they don't go to the horse's mouth, but to the horse's arse - web opinions.

But Michael, I will admit if the results I got looked like the curves you posted, I'd use "none of the above" films in your charts. If this does indeed reflect real results, I'm certainly not going to be converted to XTOL anytime soon. Worse sag than D76 results. But it is hard to tell because, by trying to show the full range way over onto the shoulder (for legitimate reasons), you're scrunching all the vital "fine print" down in the toe to the point unreadability. I'd rather see the vertical bar separated into .05 density increments or even less. I personally use .03. Then the point I'm making would be a lot more obvious. It's like using a magnifying glass for the most critical part of the curve, transition-wise, just like the most important transition in an airline flight is the takeoff and landing, the "toe". (One more reason I was the cynic over on the DIY densitometer thread; if one wants to fell a tree, don't use a fingernail file.)

 

[Helge]

Helge - Great analogy! But what you forgot to mention is that when you stretch a rubber band too far it snaps!

Michael, Lachlan - ironing out the TMX curve developer-wise is something I am very skilled at, but don't want to get into all the details here. It is remarkably developer-ubiquitous, but with somewhat different results at significantly different levels of anticipated gamma.

Please do go into detail. I have a high contrast TMax 100 roll I’d love to “iron out”.
Is compensation involved?

 

[Stephen Benskin]

Drew, I think your point would be better made without all the ad hominem attacks. Why not say giving film sufficient exposure to record shadow detail is a good idea.

If the idea is to make a comparison between the way two films respond to exposure and development, I pick the curves. They are a graphical representation of reality without extraneous and unknown influences. You can then apply the film data to a multitude of scenarios using a reproduction diagram.

I once got a call from the Herb Ritts studio. Their printer was complaining about the quality of the negatives she was given to print and was blaming the camera crew. At the studio they showed me test negatives of a very well done still life with color charts and grey scales. I decided to make some sensitometric exposures instead. Sent them to the labs they used, and after reading the results, the situation was resolved. One of the labs was using replenished D-76 and couldn't keep it under control. What they were calling normal development was really around +2.

Most of the argument between you and Michael appears to be about personal taste.

 

[Stephen Benskin]

Even a little bit of flare effectively changes the toe densities and gradient, and you can’t control it.

And there is no way to determine exactly how much flare exists in any one scene. Luminance range and Luminance distribution are both factors.

Something that can be evaluated using the film data and a reproduction diagram.

 

[DREW WILEY]

No. Compensation is precisely what I'm preaching against. Over 95% of my shots are all developed "normal". With LF sheet film it's easy to develop separate sheets for different amount of time for sake of different degrees of negative contrast or gamma. But unless you're MF camera has interchangeable backs, there is a probability that a shot or two on any roll will be out of sync in terms of original scene contrast than the others. Modern VC papers can salvage much of that, but not unless there is sufficient threshold on each frame to begin with. That's why I put emphasis on very careful spot metering of the shadows and making use of the entire straight-line portion of the curve. The reason Zonies overexpose the film is that they don't trust the fact that different films have different characteristic curves.

Back when my older brother was in a photo academy in the 60's, you either understood that in actual practice or you didn't graduate. They to learn three distinct films - Super XX with its very long straight line and wide development range (commercial and landscape applications), Plus X pan sheet film for it's extremely long gradually upswept toe or "all-toe" for sake of studio high-key Caucasian portraiture or wedding dress photos etc (not to be confused with the roll film version), and somewhere in-between, Tri-X as the most popular journalistic film (which back then might have meant a Graflex 4X5). But in so much Zone dogma, it's as if the mid-range Tri-X examples are the only ones that count. Hence compression, compensation, minus development, whatever you want to call it. I don't care much for Tri-X in general, especially not its buckshot grain; but that's not the kind of issue I'm describing here. If a different species of film has way more real estate in terms of lighting ratio range, why not take advantage of it?

Now the downside. The people who teach basic non-digital film & darkroom classes hate TMax because they don't understand it themselves. The lack of a long toe means that you might fall off a cliff if you don't carefully meter for your desired shadow threshold. It's not a forgiving film at all for shoot-from-the-hip wing-it OK Corral type photographers, at least in high contrast situations where this particular species of film really counts. In lower contrast situations, you have a lot more latitude. But I don't even like the term "latitude" in relation to a film like this. It works best for careful snipers rather than machine-gunners. And it has a somewhat longer learning curve to master, and not just a longer characteristic sensitometric curve.

To get the most out of TMax I prefer staining pyro developers, but all kinds of common developers work reasonably well. Pyro gives much better control over the high values, giving better tonality as you near the shoulder. But that's a more extended topic in its own right.

I realize that most of what I've said so far is axiomatic rather than specific; but it's important to get to first base first. Feel free to ask more questions; but I'll be away a few days, so might not respond immediately.

 

[DREW WILEY]

Flare is super easy to control. And Stephen, even mfg spec sheets show the significant difference in all these curves if one looks closely. I've plotted hundreds of curves TMax vs FP4, for instance, in far more detail than either tiny published ones or these web type things. Why? - for specialized lab applications like color separations or matching complex masking work, where it really counts. I'm not guessing, nor is it mere opinion. It's hard-earned fact. I've got binders full of densitometer plots, all done the old fashioned way.

Personal taste is always a factor. It's the name of the whole game. But that very fact is dependent on using the correct tools one needs to achieve a desired end. A pallet knife doesn't substitute for 00 sable hair brush.

 

[Stephen Benskin]

Flare is super easy to control.

Actually it's not when shooting. The majority of flare comes from the scene. You can mask off flare from the sun hitting the lens but not the highlights of the scene influencing the shadows. Two scenes with the same luminance range can produce different amounts of flare depending on the percentage of light to dark values. The best that can be done is to use an average when calculating development for a given Luminance range.

Two other points from your post. One of the reasons TXP was considered a portrait film is because the long curve would produce the same results for the lower flare conditions of a studio as TX did in exterior shoots.

Nothing against hand held meters. I use one myself, but they are separate from the camera's optical system and therefore have to average / assume certain factors. Also the spectral sensitivity of the photo cell and the color temperature of the calibrating device plays a factor as well as how linear it's response is to different Luminance ranges.

And a discussion over who has the most refined taste isn't worth my time. What is worth my time is for you to define what you consider to be a normal negative. From Luminance range to aim LER to CI. Numbers please. Another question. After your careful metering of the shadows, then what?

 

[DREW WILEY]

I aim for a versatile neg, yes. "Normal" is a plastic term differing from one person to another, depending on how they do things afterwards and their specific intentions. I'm not about to be tethered to the ball and chain of classic ZS dogma. Flare within details or aspects of the scene itself, rather than an optical artifact in the lens or bounced off the bellows, is part of the magic of the light itself, just like reflections and sparkle. You won't find a polarizer in my kit. On a studio copystand, well, that's a different story.

 

[DREW WILEY]

Michael - don't pretend to assume what I do or don't get in the field. The proof is in the pudding. I like to use the full skating rink, right to the edges. If someone else feels more comfortable without that risk, that's their choice. I wouldn't get the prints I do without understanding where the toe really does. It's not just another "Zone". I'm actually have in my head the specific shape of the toe I need with any particular film, by now largely subconsciously. So yes, at least down there I'm thinking in terms of a subtle continuum where minor differences really count. That's how I avoid burning up quality real estate way up near the shoulder, and then be forced into minus dev like Zone theory teaches. I'd rather mask if push comes to shove. All the sparkle and fine gradation gets preserved.

 

[removed account4]

Actually it's not when shooting. The majority of flare comes from the scene. You can mask off flare from the sun hitting the lens but not the highlights of the scene influencing the shadows. Two scenes with the same luminance range can produce different amounts of flare depending on the percentage of light to dark values. The best that can be done is to use an average when calculating development for a given Luminance range.

Two other points from your post. One of the reasons TXP was considered a portrait film is because the long curve would produce the same results for the lower flare conditions of a studio as TX did in exterior shoots.

Nothing against hand held meters. I use one myself, but they are separate from the camera's optical system and therefore have to average / assume certain factors. Also the spectral sensitivity of the photo cell and the color temperature of the calibrating device plays a factor as well as how linear it's response is to different Luminance ranges.

And a discussion over who has the most refined taste isn't worth my time. What is worth my time is for you to define what you consider to be a normal negative. From Luminance range to aim LER to CI. Numbers please. Another question. After your careful metering of the shadows, then what?

Stephen
I always look forward to your posts and sharing
Thanks !
John

 

[Stephen Benskin]

Over 95% of my shots are all developed "normal".

So basically a virtually meaningless statement. Sure maybe you were generalizing or using imprecise language. I have no problem with that, but then there shouldn't be claims about precision and disparaging others for their lack of it.

Let me be more precise on flare. Veiling flare is an overall non-imaging forming exposure that compresses the shadows and comes primarily from the luminance of the scene interacting with the optical system of the camera. The average flare for a statistically average scene is around one stop. It can be higher or lower. A stop flare adds a stop exposure to the shadows, changing the placement of the shadow exposure and disproportionately compressing the shadow values. It effectively takes a 7 stop scene and reduces it to a 6 stop illuminance range at the film plane. This is a part of defining "Normal" development.

I did a flare test where a "black box" target was created. A hole was cut into a black velvet lined box. A hood surrounded the opening to reduce extraneous light from entering. Cards of single tones and mixed tones were used to surround the opening and introduce flare. It was shot on 35mm film. An exposure was determined. After the test a sensitometric exposure was place at the end of the roll. The resulting values from the test are placed onto the film curve.

Now, this just shows the change in shadow placement caused by flare and not the compression.