Shadow detail and tonality

[roger_hicks1]

Dear All,

 

Rich shadow detail is something of a Holy Grail for many

photographers -- hence, in part, re-rating films well below their ISO

speeds. But how often does it really work? All too many pictures with

much deep shadow area are murky, gloomy and dull, especially if the

film has been underdeveloped (N-, in Zone terms) to compress the

tonal range.

 

Mortensen said (entirely accurately) that photography deals best with

short tonal ranges, and indeed recommended development to gamma

infinity (maximum contrast -- N-plus-as-much-as-you-can-get, in Zone

terms). You may not care for his art and composition -- I don't,

greatly -- but the tonality of his pictures is often fascinating and

many of them meet the criterion, noted in another thread, of looking

good under any lighting against any background.

 

So here are the questions:

 

First, is it the extra shadow detail, or the change in tonality

resulting from extra exposure, that makes so many people downrate

their films? This is assuming they understand metering at all.

 

Second, how many people have tried the Mortensen approach of high

contrast development, and fitting the subject to the medium, as

opposed to the Adams approach of trying to fit the medium to subjects

for which it is not inherently well suited?

 

Cheers,

 

Roger

[stephen_benskin]

<i>First, is it the extra shadow detail, or the change in tonality resulting from extra exposure, that makes so many people downrate their films? This is assuming they understand metering at all.</i>

<p>

I think a misunderstanding of how the meter works accounts for 90% of the underrating of most film. The change in tonality resulting from extra exposure is dealt with on the Over Processed - Anecdotal vs Objective Analysis thread. There is a slight increased contrast in the shadows. Mostly, slight overexposure is only a safety factor. The increased overall density is simply printed through on the print. I have a graphed example in the wings if needed.

<p>

<i>Second, how many people have tried the Mortensen approach of high contrast development, and fitting the subject to the medium, as opposed to the Adams approach of trying to fit the medium to subjects for which it is not inherently well suited? </i>

<p>

Isn't this how commercial photographers work? Isn't that why they Make certain subject choices and bring reflectors and lights to limited the luminance range?

<p>

I also have a question about Mortensen's approach. How fast did the films of the day reach Gamma infinity? If it was reached at a fairly low contrast, then his approach turns out to be not so radical. This is a problem when someone like Mortensen doesn't objectively test. Who knows what he was really doing.

<p>

<i>All too many pictures with much deep shadow area are murky, gloomy and dull, especially if the film has been underdeveloped (N-, in Zone terms) to compress the tonal range.</i>

<p>

This idea was also dealt with on the Over Processed thread. It comes down to the tone reproduction rule that preferred midtone reproduction contrast should be greater than the original subject's midtone contrast. I believe it falls around a gradient of 1.10. If the midtones are too flat, will that influence the perception of the shadows?

<p>

Roger, this is the third perceptual question in as many days. What are you working on?

[oskar_ojala]

1. An interesting question. I do think that many photographers are far more interested in the technical aspects of their negatives (eg. grain size, capturing all the tones in the scene) rather than the artistic qualities (eg. how good does it look when printed.) A possible reason would be to be on the "safe side": compressing the tonal range and downrating slightly to get better shadow detail gives some protection against underexposure, which would indicate that "they" don't understand metering well enough.

 

2. I'm not familiar with Mortensen's approach, but "fitting the subject to the medium" sound way too limited. One appeal with digital techniques is to overcome precisely the limitations of the medium that film posesses in order to be able to shoot any subject without doing the medium choices that one does with film.

[conrad_hoffman]

It seems you have a choice. Either fit the process range to the subject, and live with whatever midtone slope you get, or adjust the lighting ratio via reflectors, fill flash, softboxes, or cloudy days, to match a process that gives you the midtone slope you want. I'm not sure gamma infinity has a place in either method, but one guess as to which one looks better most of the time.

[scott_eaton]

<I>All too many pictures with much deep shadow area are murky, gloomy and dull, especially if the film has been underdeveloped (N-, in Zone terms) to compress the tonal range. </i><P>You've just described 80% of the images uploaded to the B/W forum.<P>I agree......pull processing Tri-X and TMY to EI 200 under overcast skies results in the most bland images imaginable. We then pick on XP2 and the chromogenics for being low in contrast....go figure.<P>OK, aesthetics aside. If you at least have shadow detail you have somewhat linear processing and have something you at least have the option of throwing out. More often that not lack of shadow detail with novices is the fault of gross under-exposure. If we had a global rule about B/W developing that if you have processed less than 20rolls of B/W film, increase your developer time by 50% and don't argue about it.

[pat_wilson1]

As I understand it, and it seems to work in practice for me at least, film is underrated because film tests are greatly linked to the equipment they are done on. I shoot TriX sheet film and it is rated, according to Kodak, at 320. I rate it at 200 because when I performed a film test using my equipment, my developing method (nothing wacky, just my hands in the trays intead of someone elses) it was at 200 when I got a density on the film that was 0.1 above the film-base + fog. I did development tests to see when a Zone VIII exposure was 1.3 above fb+f and that was my correct development time. Does this make my pictures asthetically better, well no, but when I meter I know what kind of information I'm going to get on the negative.

[stephen_benskin]

There are three reasons why the majority of people who do a ZS type testing get results such as Pat's.

 

One: Way too many variables in a single test. That whole testing for one's personal equipment, not so much a valid argument.

 

Two: Speed point is only 3 1/3 stops from the meter's exposure point and not four stops.

 

Three: Flare which isn't a factor in the test, but is calculated into the ISO film speed. This may not show so much in the testing, but will shift the placement of the shadow higher even than the 2/3 stop extra given by the meter misinterpretation.<div></div>

[arne_strasser]

If you expose a film at the rated speed with your favorite developer, you are quite likely to be on the "mortensen-side"- you underexposed and overdeveloped your negative. Mortensen simply found this to often look better when printed.

There is an interesting read here http://unblinkingeye.com/Articles/Mortensen/mortensen.html

[stephen_benskin]

After doing the Zone System test and finding a EI at 2/3 stop less than the ISO film speed rating and correcting for this, it's time to use it in a real world situation. In the real world, there exists flare. While this was ignored in the ZS testing, it will influence the shadow placement on the film's characteristic curve.<div></div>

[pat_wilson1]

I would still argue that doing testing is better than shooting blindly. I, in no way, advocate extensive testing. Photographers make photographs, not line graphs. I still get better results using tested data than just using manufacturers iso/dev time standards. I didn't test out to perfection, for example, I did 3 tests for development time, recomended, +33%, -33% and just took a SWAG for my normal development time based on that. Then I can tweek my system as I shoot real shots instead of grey cards.

 

As with everything in photography, how you get to the goal really isn't as important as getting there. A lot of the photographers I admire the most did nothing of this sort, it was all internalized and they did what felt right, and I respect and admire their work. I just report the way I've learned to do it, and can say that I feel my negatives and prints have come out way better than the stuff I produced when I started. So I'll borrow an acronym from PERL...TIMTOWTDI. (There Is More Than One Way To Do It)

[stephen_benskin]

Hold on there Pat. Don't get defensive. Any testing is better than no testing. The fact that you test places you ahead of the majority of photographers. I'm just showing how the ZS test, which many people use, can result in a higher exposure than people think they are really getting. It is a very common misconception bordering, so common it is almost in the realm of myth. The graphs are a simple way of expressing my idea. The ZS testing, in effect, is only bringing back the safety factor from the pre 1960 standards.

 

On the muddy issue, there is an old adage, shoot and process for the overall contrast, print for the local contrast. Just because a negative can print the entire luminance range of a scene without burning doesn't guarantee a quality print. Yes, the tones all fit the paper, but in what relationship. I don't factor in most skies when I meter because the relationship between the clouds and the ground scene is greater than where I want the relationship in the print.

 

Incorporate the clouds into the metering and the ground values will shift down. They become dark. In the negative many tones are placed into the lower portion of the curve and into the toe compressing values. In the print, the ground scene is darker and has less life. My approach is to meter for the ground with the intention of having to print down the clouds. The print is keyed to the ground values and not the highlights.

[patrick_gainer]

I was principal oboist of the Norfolk Symphony in Virginia and took many photos of guest artists at dress rehearsals where stage lighting was harsh. My camera's built in auto exposure told me the exposure should be different every way I turned. I found out that was not true. Light reflected from music and not reflected from other things did not average out to the average of spot readings, nor did any one spot reading suffice if I just used the camera in point-and-shoot mode. I had to take a shadow reading, set the exposure setting a couple of stops down from that reading, and disable the auto exposure mode. That was good wherever I turned.

 

Changing the film speed and using the auto mode would not have done the job.

 

My complaint against those who say "I rate so-and-so film at such-and-such EI" is that I never have seen any one describe their method od using that EI, let alone what it was that caused them not to use the ISO number. I think if you find out what the ISO means, you can adjust the meter reading to suit the job. That means not saying that the box speed is wrong absolutely but that I will take into account the circumstances when I use it.

 

Did I get off topic? Sorry.

[stephen_benskin]

Found something that kind of deals with this. They have a simulation.

 

Simulation of Human Visual Response at Varying Levels of Illumination

 

http://www.cs.berkeley.edu/~efros/java/vision/vision.html

[lex_jenkins]

Roger, you and Stephen really keeping our brains on their toes, so to speak, lately.

 

First, I'll respond directly to your puzzler without reading the other replies. I'll come back, read them and perhaps add something later.

 

I must confess to being sympathetic to Mortensen's approach, if only because he was so vilified by the Adams cultists. And, in fact, Mortensen's approach has worked well for me *within those boundaries* you've aptly described. Over the past couple of years I've taken many photos around our lakefront property, mostly of scattered planting pots, galvanized steel buckets, that sort of thing. Very limited tonal range and I usually prefer hazy light or late afternoon/early evening light filtered through our thick stand of trees.

 

Exposed and developed normally these often would not turn out so well. But after a few experiments and failures I've figured out which combinations of films, developers, EIs and development times will work well so that these subjects don't appear as merely murky midtones, lacking any highlights (because there are none) or significant shadow detail because of the diffuse lighting.

 

In retrospect, I may have applied his gamma infinity approach to my photos taken under full moons, at least when stand developing in very dilute Rodinal (I don't think Diafine quite applies due to the fixed nature of its development). If I'm interpreting his meaning of gamma infinity and its use correctly (it's been a while since I read anything by Mortensen) some of my goat skull or wine glass on dock over lake photos under full moonlight might qualify as examples in support of the theory. At any rate, I certainly wasn't thinking in zonie terms when I made those photos.

 

Finally, I don't wish to dismiss the Adams approach to fitting the medium to the subject, regardless of the obstacles. In fact, most of his work was an incredible testament to doing things the hard way and making it work.

 

But besides the fact that I work only in 35mm and 120, generally making the zone system irrelevant or at least impractical for me (tho' let's please not rehash that old argument), I'm not interested in recreating the type of work Adams did, either in subject matter, exposure, development or printing. He pioneered an approach and a "look" to prints that have influenced countless photographers. Sometimes, tho', I look at another recently made photo of craggy canyons in that style and all I can do is sigh. Nice pix. Seen it.

 

I guess what interests me is trying to find subject matter and lighting that cooperate with the medium, without being boring (that's the tricky bit), and seeing what a little conjuring in the darkroom can produce. Maybe I'm lazy, but I would just hate looking at a high contrast midday scene and trying to figure out methodically how to squish it into the impossible confines of the medium. It's admirable when others can do that, tho'.

[patrick_gainer]

With regard to Mortenson's approach, gamma infinity for HP5+ is only about 0.6 in Rodinal 1+50. Not that he used that film, but that is an example of a film that will not get out of the normal range under some circumstances. He could not have used Tech Pan with that method. Also, if he was using stand development, gamma infinity might have been quite a bit lower than with agitation.

 

No big point, just a comment.

[rob_gruber]

I've never seen a photo and had my first (or second or third) thought be "wow, great

shadow detail". It literallly doesn't impress me, that is make much of an impression.

Overall tonality, yes, very much so. But shadow detail, not at all.

 

<p>Some of the my favorite photography, Robert Frank, Bresson and other street

photographers, is technically pretty imperfect and tends toward over exposure.

William Gibson a more atistic photographer uses overexposure and over-

development very deliberately. And for artistic effect it works. Same with Ellen Von

Unwerth. Tri-X in Rodinal over exposed and over developed.</p>

 

<p>I've never really understood the obsession with shadow detail and minimal grain.

But then while appreciating what went in to Ansel Adams photos, I don't particulalry

<i>like</i> them artistically. I think if photographers are honest with themselves,

the obsession with shadow detail, the zone system et al can all really be traced back

to him.

 

<p>Interesting thread.

[james_megargee]

It seems that whenever this discussion comes up (the benefits of over

exposure/under development) there never seems to be any mention of the

printing technique (paper selection, paper grade, developer, enlarger type,

etc.) ? When I discuss this with other photographers it generally results with

them telling me that they usually print these type of negatives on higher than

"normal" grade papers? This logic escapes me. You create a negative that is

flatter than "normal" to add additional information to that neqative, only to

have to print on a higher grade of paper that will eliminate that additional

information?

[eduardo_benavidez1]

The printing process, properly made, wont necesarily eliminate that additional information but even preserve and enhance it. Eduardo.

[stephen_benskin]

James, �Under processing� the film and printing the negative on a higher than normal grade of paper will not change the tonal extremes (black and white points), but it will change the tones in between. It will increase the local contrast. This is a very useful creative tool.

 

Roger, I�m not sure if this falls under visual blending or adaptation. You see this in phenomenon in movies all the time especially in night scenes. It�s acceptable to have a higher lighting ratio on the actors face at greater distances than with close-ups. When someone walks toward the camera at night, their face can be in shadow until they come within a certain distance and then they lighten up. This is of course a lighting trick, but it�s also valid in the real world. From a distance a night, you cannot make out the features of a person walking toward you. They appear too dark. As the person approaches, his features become recognizable yet the illuminance level hasn�t changed. This appears to have something to do with distance and perception.

 

That�s why I like to print dark areas larger than just accents to what I call �perceived black.� If you look at the print from a distance, the dark area will appear black or close to black, and as you approach the print, detail in this area begins to become apparent. This approach also tends to give the impression of dark without being harsh and lifeless. An additional benefit is that it tends to maintain a perception of print quality under a variety of print viewing conditions. My final step in evaluating my prints is to view the finished print under different lighting conditions. If it achieves the criteria that Roger has explained then it�s acceptable.

[stephen_benskin]

Now that I think about it, blending is making more and more sense. It�s just in all the examples on blending I�ve read have to do with tones blending to create the perception of a midtone. There�s the classic oak tree example. From a distance, the oak tree�s trunk appears gray but as you approach, you can see the dark crags in the bark. Then there�s the one about distant mountains. Negating atmospheric influences, the dark shadows cast by the trees blend with the highlights to create a very flat middle gray looking scene. Then there�s lithography. Little black dots against white paper appear gray from a distance.

 

Perhaps if the area was at a certain uniform darkness, the blending will shift downward and not toward middle gray.

[james_megargee]

Stephen - in practical application it is true that the higher in contrast you print

the less information you are starting with. Of course the highest and lowest

points will not change but the mid range and (more importantly) the subtlier

lower and higher values will no longer be available to begin with reguardless

of how accomplished the printer. My point is that in my experience most

photographers do overexpose their shadow areas (for whatever reason) and

then have to go to a higher than normal paper to get rid of that extra density

in order to create a sense of contrast in those areas. I have always seen this

as a questionable practice especially with the newer emulsions available

today.

[stephen_benskin]

Sorry James, I meant to say Rob. I think I know what situation you are trying to explain in your response to my misdirected post; however, if I'm interpreting it correctly, it isn't supported by tone reproduction theory. This is a bit off topic, so if you want to start a new thread, I think it could be an interesting one.

[chris_waller]

I once tried downrating FP4 to 20 ASA and reducing development accordingly. Certainly I recorded all the tones in the subject within the density range of the emusion, but the negative and resultant prints were as flat a pancake. I am currently running trials of D23, a soft-working developer, and good as it is, I still find that split-grade printing is necessary to get the best out of some negs.

[stephen_benskin]

<i>I once tried downrating FP4 to 20 ASA and reducing development accordingly. Certainly I recorded all the tones in the subject within the density range of the emusion, but the negative and resultant prints were as flat a pancake.</i>

<p>

I don't have the exposure data to allow me to confirm the exposure situation stated above. I do have enough data to confirm a one stop over exposure, and I have found that the increase in density range of FP4P with an increase of exposure levels off within a stop. Once the shadows are off the toe of the curve, any additional exposure will not increase the negative's density range. The only question remaining is the point in the upper portion of the curve where additional exposure will begin to decrease the density range. The resulting density range increase with the increase of exposure at a CI 0.61 was only 0.04 or about 1/4 of a paper grade. This suggests that it is counter-indicated to pull the processing with any over exposure.

[patrick_gainer]

Exactly so. Overexposure does not require underdevelopment unless the high end of the scene is above the shoulder of normally developed film. The shoulder of underdeveloped film is lower than the shoulder of normally developed film, or at best no higher. Overexposure is not the way to save a wide range scene. Minimum correct exposure is the best way to go for any scene. Development as for a normal scene range preserves middle range contrasts on grade 2 paper. If the actual scene range was greater than normal, you can burn or use softer paper. If the actual scene range was lower than normal, you can use harder paper. There is no way to get around burning a wide range scene. If you underdevelop the negative, important parts of the scene will be flat. Then you must use a harder paper and burn.

 

When you use roll film, unless you are using one roll for each scene brightness range, you must make a choice or develop the whole roll the same. Even A. A. recognized this fact.