Compensating developers and accutance developers : LUSENET : B&W Photo - Film & Processing : One Thread

This is one of FAQ's that I receive through email. I usually give short practical answer in direct emails, but I prefer to discuss more messy aspects here, hence this posting.

I said in recent thread that compensating formula don't work as they are supposed to be. I meant it makes only a small dent in creating a shoulder. I'm a post Panatomic-X generation who can't talk much about old emulsions except guessing from some "fossil films" that I have access to. But I feel a need for a discussion of what makes modern films modern, that is, making compensating formula ineffective to supress highlights.

When they say "improved emulsion" it usually means smaller silver halide particles with improved sensitizers and possibly other tricks on the composition and surface of the AgX's. This usually allows grain to be smaller and use less silver. It seems to develop effectively in more dilute developer possibly through better adsorption of developing agents on the AgX surface. (one usually double or triple the developping time when dilute 1+3 or 4x. this phenomenon has been explained by adsorption of developing agents, and as far as I know there is no dispute about this working hypothesis. this effect, of course, depends on the charge of the agent. HQ and PPD would have very different behavior here.)

It would be very interesting if John Hicks and others don't mind sharing their "rotary compensating factors" for many films. This is because if rotary agitation does not shorten processing time (to the same contrast) by a large factor, it indicates that the development is the limiting factor, meaning that enough chemicals are "there" and making the road wider doesn't help much. Modern films use different treatment for gelatin compared to old ones, like hardening. I don't know the details of the trick and how they affect the diffusion acrss it, but it is certainly worthwhile reverse engineering. For example, Delta 100 might use different trick on gelatin because it requires a large reduction in time with rotary agitation (John Hicks data), implying that diffusion is a bigger factor in overall rate compared to Delta 3200.

So why these films respond rather well to accutance formulae? Exhaustion of chemical agents are unlikely in many cases because compensation formulae don't do. One possible explanation is that halides released from highlight development diffuse out to adjacent midtone and shadows where development is inhibited. This is consistent with a phenomenon that increased sulfite diminish accutance effect, and that highlight is not inhibited.

A dichotomy: HP5 Plus v.s. TMX

HP5+ requires longer processing time when the developer is diluted compared to TMX. In other words, TMX is less sensitive to dilution error on the contrary to what is commonly believed. Incidentally, Tri-X is less dilution sensitive compared to HP5+. Different companies publish their data obtained for different criteria, and different developers act differently, but you can see a general trend by comparing them.

Having looked at curves by Clay Harmon (posted at Ed Buffaloe's site), pyrocatechin doesn't seem to make a big dent at least in Pyrocat-HD formula. This is partly because the film is developed to relatively lower density compared to conventional non-tanning formulae. Dilute pyrocatechin formulae and HP5+ or Delta 100 might do something. Besides that, there doesn't seem to be a lot of hope for compensating effect there.

If you want compensating effect, don't seek a trick in developer. Use films with intrinsic shoulder. If you want adjacency effect use combinations that are susceptible to bromide drag, preferrably containing little sulfite, at a great risk :-)

NOTE: What's stated above are mostly hypothetical models that I think are consistent with what's in the literature and what I observed, as well as what's previously discussed here. However, I don't know how far they can stretch to films I haven't used. I'm very interested in hearing contrasting observations.

-- Ryuji Suzuki (, May 16, 2002


No answers, but I can share some experience. The only ways I could put a shoulder on any modern film (and by this I mean FP4, HP5 kind of stuff - basically not old style emulsions like Super XX) was to 1) drastically reduce agitation with hugely increased development times (and I mean drastically reduced agitation - like once every 3-5 mins) and 2) use a water bath. Obviously both methods come with inherent risks of uneven development. The mechanism is also physical rather than chemical i.e., starve the highlights of developer. Rotary development (with about 25% reduced times) provided identical curve shapes to those I got with my usual agitation pattern. I agree with your central thesis - if you want a shoulder, use a film with an inherent shoulder. There does not seem to be any chemical way to induce a shoulder (unless it is combined with a physical way like reduced agitation). I think it is almost easier to go with masking techniques of some kind to alter gradation.

With regards to adjacency effects, Richard Henry reported that adjacency effects seemed to be primarily due to bromide diffusion within the emulsion rather than on the surface. I have found pyro, Rodinal and D23 to give very differnet adjacency effects with rotary processing, which would seem consistent with Henry's findings. In other words, you don't need to run the risk of uneven development for adjacency effects.

Cheers, DJ

-- N Dhananjay (, May 16, 2002.

"Richard Henry reported that adjacency effects seemed to be primarily due to bromide diffusion within the emulsion rather than on the surface" I agree. I meant more like "for maximum adjacency effect" instead of plain normal adjacency effect. If you used things like Microphen or XTOL, you don't get as much adjacency effect as you would in original Gainer formula. (all at normal intermittent agitation)

-- Ryuji Suzuki (, May 16, 2002.

The recent pyrocatHD tests I posted on Ed's site seem to indicate that the Fortepan400 may be as close as you'll get to a 'modern' film with a shoulder. It seemed to have a slight shoulder on both the pyrocatHD test and a subsequent Divided D-23 test I ran on some leftover film that I had exposed to a step wedge. I print mostly platinum, and need a lot of density range, so the Fortepan will be difficult to use unless I have a really high contrast scene. But for silver printers, it might be useful. The manufacturers claim it is an 'old style, silver -rich' emulsion, whatever that is supposed to mean.

As an aside, I noticed an old post asserting that Bergger film is the same thing as Fortepan. I have boxes of both, and it IS suspicious. They are in the same odd-sized box, have the same film notch, the same interleaving papers, and most tellingly, behave the same in every film test I have run. I know what I think...


-- clay harmon (, May 16, 2002.

> if rotary agitation does not shorten processing time

The only factors I came up with are the ones I posted on Ed's website . What I learned was that there's no factor that's valid for all films.

Note that the curve shapes for HP5+ in D-76H 1:3 are essentially the same for both intermittent and rotary agitation. One could say that a D-76 type developer doesn't become a compensating developer simply via dilution _presuming sufficient agitation for good evenness_ is given even if it's intermittent.

Since I wrote that article I've worked up a development time for Delta 100 in D-76H 1:3; again the curve shape is virtually identical for both types of agitation.

In working with various "standard" developers and a couple of so-called divided developers I've never come up with signficant differences in curve shape due to intermittent vs. rotary agitation, or, in the case of divided developers, any useful compensation again presuming sufficient agitation in A or B for good evenness.

It's frustrating. I believe I used to get compensating effects with old films in dilute developers but I wasn't plotting curves back then so I don't really _know_ I was; it may have entirely been wishful seeing.

In looking at some of my old curve plots, the only ones that show a really significant difference in curve shape are HP5+ in Rollo Pyro (or PMK+, both for rotary agitation). Generally speaking, there's a really strong shoulder and, depending on dilutions, a greater or lesser divergence between visual and blue-filter readings. I think it'd be fine for 8x10 and I can see why the alt-printers really like it; otoh the negs were really grainy and when printed on VC paper the stain made the lighter tones way too flat.

Anyway, I agree with you; assuming sufficient agitation is given for acceptable evenness, compensation isn't going to happen with a standard-type developer even if it's highly diluted. Acceptable evenness is of course a personal judgment call.

-- John Hicks (, May 16, 2002.

I routinely process modern films (TMX, Delta 100 & 400 and Acros) in my catechol/glycin formula, often requiring extreme compensation for my NYC night shots. By diluting the developer to 1/4 its strength and drastically reducing agitation, I've achieved N-6 (really!) with modern films and without streaks or mottling. I typically agitate (4x5 in trays or slosher-type holder) for one minute initially, then increase my interval as processing time goes on; a typical sequence is 15 sec. agitation @ 1, 2, 2, 3, 3, 4, 4, 5, 5 .... As you might imagine, this requires conserable patience, since my developing times for N-4 to N-6 range from 30 to 45 minutes.

I use a rotary processor for N-1, N and N+ processing, and I have not seen any significant difference in acutance, grain or compensation when I use a more dilute developer. I believe the only way to achieve significant compensation is by reduced agitation in dilute developers.

Obviously, some films and developers are more prone to streaking than others. Catechol and glycin formulas, whether combined or separate, seem to be quite resistant to streaking. But I've also used Gainer's vit-C formula in similar fashion for N-2 to N-3 and the results were very good. I don't have data to back this up, but my guess is since both my fomula and Gainer's have little or no sulfite, low sulfite formulas may be a factor in limiting streaking with severely reduced agitation.

-- Ted Kaufman (, May 17, 2002.

Ted: Would you feel comfortable posting your catechol / glycin formula? It seems to me that it would be a pretty good combination, and it would save a lot of experimentation if you would post it or send it via email. I'd appreciate it greatly. Thanks. Clay

-- clay harmon (, May 17, 2002.

Great thread guys! I wish Ryuji could refine and clarify his original post a bit, everyone else could add their comments, and we could put together an article for Unblinking Eye. If each contributor ran one test on his favorite film and developer combo, we could plot them all in Excel and combine the data on a single web page.

This thread illustrates the synergy that can result when intelligent people have the chance to network.

-- Ed Buffaloe (, May 17, 2002.

Ed B's suggestion gets my vote. I use a small tank, and often want to get a compensating effect. The article and researched times would be enormously helpful.

Thanks to you skilled and intelligent people. We all benefit from your hard work and generosity. It's the sort of positive use that sites like this should be used for!

-- Ed Hurst (, May 17, 2002.

> the negs were really grainy and when printed on VC paper the stain made the lighter tones way too flat.

"Lighter tones way too flat" implies that the developer does more than enough compensation action... If you get excessive compensation then it is probably easier to tweak the formula to reverse some. Same on granularity.

> low sulfite formulas may be a factor in limiting streaking with severely reduced agitation.

I don't think so. Sulfite-poor developers have nothing to balance out bromide ions adsorbed onto silver halide by disturbing it by mild solvent action. When bromide adsorbs, other negatively charged molecules like hydroquinone and catechol can't get close to the silver halide particles unless they are already partially developed. para-phenylenediamine is not charged so it's not disturbed by bromide barrier. On the other hand, agents like metol like to adsorb there. In this case dilution does not help much, because no matter how you dilute it, metol sticks to the silver halide particles. Then you'll lose something before getting compensation.

For tanning formulae, low sulfite is really a necessity, but other than that, I see no reason to avoid some sulfite. Hydroquinone developers for continuous tone purpose tend to go out of control without sulfite, and ascorbate developers tend to lose speed and require longer induction time without sulfite. These are well documented in literature with some good explanations. Basically, sulfite is there for a number of reasons, regenerating developer oxdation products, minimizing gelatin swelling and making sure developers go through gelatin easily (because gelatin tends to be positively charged at pH above 5), preservative, mild solvent, etc. People think solvent effect loses resolution, but this is not so in the case of moderate sulfite. Solvent effect is not as important as others, and sulfite is not a strong solvent anyway.

Ted's N-6 sounds interesting, but I'd have to think about a way that can be applied to roll films because most of films I use are 120/220.

> I wish Ryuji could refine and clarify his original post a bit,

Refining part takes some off-like editing because I hate editing in this tiny box, but if there are further questions I can try my best. (Most stuff, you can just go to library and read though.)

-- Ryuji Suzuki (, May 17, 2002.

"Lighter tones way too flat" implies that the developer does more than enough compensation action...

Actually, in this case, I would suspect there might have been no compensating action at all. The greenish pyro stain acts as a variable contrast mask with variable contrast paper which is sensitive to green light. So typically, the shadows print with very good local contrast and separation and the highlights print with increasingly less local contrast. Functionally, thought, you are right - the effect is the same as putting an artificial shoulder on the curve of the film. Whether one likes it is a personal call, I guess. I have increasingly come to prefer straight line characterisitcs - -just like the aesthetics of it, no abrupt changes in tonality in different parts of the scale. One other reason I have come to like straight line characteristics is that printing is very easy. Moving the print exposure around will keep local contrast the same in all areas, especially if your printing paper also provides reasonably straight line characteristics - with more marked shoulders and toes, when print exposure is varied, local contrast will change also. But then, keep in mind I do not do the kind of work Ted talked about above. The problem with night shots is that local contrast is already low in the shadows and high in the highlights to begin with - so in these situations we really do need a way to change the distribution of those tonalities - simply developing to a lower contrast index will make shadows very murky.

Cheers, DJ

-- N Dhananjay (, May 17, 2002.

> Actually, in this case, I would suspect there might have been no compensating action at all. The greenish pyro stain acts as a variable contrast mask

Right, that's what I was getting at. While there was definitely fairly strong highlight compression the stain color made it _way_ too much for printing on VC paper. It might have been fine on graded paper or the stain could have been decreased or eliminated but the graininess was so ugly I didn't pursue it further.

-- John Hicks (, May 17, 2002.

> I would suspect there might have been no compensating action at all.

Clay Harmon's data show no shoulder, but John Hicks said he got highlight rolloff in addition to the increasing stain further lowering contrast... Don't you call it compensation?

I guess we should start define what "compensation" is before discussing... ;-)

-- Ryuji Suzuki (, May 17, 2002.

When I was young, about 30 years ago, someone told me about the magical compensating powers of divided D-23. For a while, I thought it was the next best thing to sliced bread. Then I built a densitometer and started comparing curve shapes of a lot of film-developer combos. Much to my surprise, there was no discernible difference between D-23 and divided D-23. In those days of Super XX and the like, we wanted as much shadow speed as we could get. Compensation was our name for that. No one thst I can recall thought much of trying to produce a shoulder. I still do not. We have both Kodak and Ilford procucing long-toed printing paper to counteract the upsweep in some of the higher film tones of modern films, and people trying to get a shoulder and then cursing the manufacturers for the long-toed paper when they succeed. I have found that in two radically diferent developers, D-76 1:1 and my own phenidone-vitamin C stuff, when you find the normal developing time for periodic agitation, say 5 sec/0.5 min, you can multiply that time by 1.42 to get the time for no agitation except the first 30 seconds. I get somewhat better shadow contrast with the same highlight density. As an old available light people photographer, I am just tickled pink with that knowledge. This works for a normally 5 min developer, my PC, and for D-76 1:1 which I did for 7 min at 70 degrees with agitation and 10 minutes with no agitation. You worry about lack of agitation. I have done 35 mm and 120 in reels and 4x5 in tray both face up and face down, and standing on edge in a Combi Plan tank without agitation and have seen no sign of bromide drag ot other smearing or uneven development. I think the problem is that once you make up your mind not to agitate, stick with it. Continuous agitation can be worse than none if it causes eddy currents, which is very likely to happen when developing film in reels in a rotating tank. I plan to do more work on this subject. It is very tedious because one thing at a time must be varied in order to make sense out of the mess. Too many of us change too many things at once in our experiments. The final answer may be simpler than changing dilution and accelerator and method and degree of agitation and developing agents name it. Maybe in the end we'll find that a simple change in agitation procedure will be all we need. Wouldn't that be great?

-- Patrick A. Gainer (, May 17, 2002.

I've used Dixactol a bit but never really got what I wanted. It will compress a remarkable scene range into the negative, but getting it out can be a chore. I don't know if it makes a shoulder or just compresses the range but in any event it will capture sharp highlights.

I want(ed) a developer to make conventional 35mm films behave like XP2. I've tried various 2 bath methods, changes in agitation, dilution, etc. and I've never found the magic potion.

-- Henry Ambrose (, May 17, 2002.

I've been thinking about what the goal is here. To my way of thinking, the whole point in 'compensating' developers is to get great shadow contrast without the highlights blowing out on you. I have it easier than a lot of printers since I print in palladium which has the long toe of all long toes. But shadow contrast can be a challenge. I am now wondering about one other factor that no one has talked about : reciprocity failure and its' effect on contrast. In looking at some of my better negatives of subjects with a high brightness range (SBR of 10-14), it seems that the ones that required exposures that measured in the minutes to almost and hour due to reciprocity departure seem to have better shadow contrast than the ones that required only a half second or less The increase in contrast due to reciprocity departure is well known, thus the recommendation for reduced development times in the Kodak and Ilford data sheets. Is it possible to get the higher shadow contrast through reciprocity departure, yet keep the highlights from developing off the chart through stand development or using a divided developer that exhausts (in theory) in the highlights first. I don't know. I am thinking out loud here. I have an idea on how to test this idea, and will communicate back to this thread if anything comes of it. Any thoughts about reciprocity failure's role in the scheme of things? Ted's fantastic N-6 example has great shadow separation in the foreground area. Is it possible that some of the shadow contrast is caused by reciprocity departure? BTW Ted, what was the exposure time on that image? Food for thought.


-- clay harmon (, May 17, 2002.

Hi Clay, I think you are onto something here. The failure of reciprocity basically means that the sensitivity of films drops at low light levels. The reason for the increase in contrast is that the shadow end of the scale is being hit by much dimmer light than the highlight end of the scale i.e., the shadows suffer from greater departure from reciprocity than the highlights (thus the increase in contrast and recommendation for reduced devlopment). However, you are right, this would mean that contrast in the shadows should INCREASE. This would be because the shadows would suffer from much greater failure from reciprocity i.e., their densities would drop much more dramatically than the decrease in densities of the higher zones. Very elegant solution to the problem of getting better contrast in the shadows without affecting the contrast in the highlights. It's probably worth plotting some curves.... food for thought. Cheers, DJ

-- N Dhananjay (, May 17, 2002.

> highlight rolloff in addition to the increasing stain

To clarify; I got highlight rolloff in both silver and stain density; it was the combination that caused too-low highlight contrast. Since it was a pretty abrupt rolloff it may have been more a case of developer death (pyro in a Jobo) than anything else. Anyway, I filed away the results for future reference, such as when I get lured into platinum etc.

What I'd like to achieve is a broad shoulder rather than abrupt flattening; Delta 3200 and Delta 400 come closer to this than anything else I've tried. Reading between the lines of Ilford's promos etc leads me to think that Ilford will tend this way with future film versions.

-- John Hicks (, May 17, 2002.

There were some interesting suggestions here. I strongly prefer repeatable procedure avoiding infrequent agitation. Even if the solution is simple, if it requires a magic hand to make it, it doesn't seem to be a good solution to me.

If some pyro or pyrocatechin developer can make usable shoulder, I suppose it is easy to reduce or remove the stain after that point to avoid exceedingly lowered highlight contrast. One problem is grain, as John Hicks suggested. If grain can be tamed, it might be a good soup for TMY :-) If grain is unavoidable, I might just go back to Plus-X instead, depending on how much modification the new version receives.

I might have said a little about this in the original posting, but the most likely reason staining developers can help achieving compensation is through local hardening of gelatin. The oxidation products of those agents increase cross linking of gelatin, making it shrunk, harder, as well as stained. However, literature data suggest local hardening requires less oxidation products than staining. A caveat here is that those data were obtained with old emusions and relatively high pH, low salt environment, letting the soft gelatin swollen as much as it liked until development progressed.

"Is it possible to get the higher shadow contrast through reciprocity departure [...]?"

This is an interesting question. Something I never wanted before! Non tabular grain films of high speed gives you great depearture from reciprocity law, but you want to chemically create it. I suggest trying various restrainers. But alkali metal salts of halides are probably poor choice because they will turn into silver halide solvent at high concentration, possibly before you get desired restraining effect. (This may sound strange violation of solubility product principle, but the solvent effect is because of formation of soluble silver halide complex ions, and is subject to fog and solution physical development.) I suggest those restraining dyes should be tried. This, obviously, I have never tried and I don't know how workable the idea is.

-- Ryuji Suzuki (, May 17, 2002.

Right after I made my previous post, I went and developed a test strip of 35 mm Tri-X 400 in PMK (pyro) developing it for 20 minutes at 70 degrees without agitation. This is supposed to be a no-no. However, it worked just fine, giving me full scale negs with plenty of stain without a sign of uneven development. I'll send some stuff to Ed when I can get to the post office. I don't think I could show what I want to show by e-mail.

-- Patrick A. Gainer (, May 18, 2002.

A couple of notes:

N Dhananjay wrote: "However, you are right, this would mean that contrast in the shadows should INCREASE. This would be because the shadows would suffer from much greater failure from reciprocity i.e., their densities would drop much more dramatically than the decrease in densities of the higher zones."

DJ, I think the increasing is not obvious; lets assume the normal curve is stright. Then, when we start experience the failure of reciprocity, the shadow part of the curve may become convex or it may become concave, -- it depends how fast drops the film sensitivity. -- Am I wrong? -- Undoubtedly is it worth plotting some curves. Very interesting idea.

Ryuji Suzuki wrote: "but the most likely reason staining developers can help achieving compensation is through local hardening of gelatin. The oxidation products of those agents increase cross linking of gelatin, making it shrunk, harder, as well as stained".

Ryuji, may be it makes sense to split the development into 2 baths: 1st does mostly sa elective hardening, the 2nd does mostly the very development? I think it is important to find a "sweat point" when the hightlights are hardened enough to impede the developer penertation, and the shadows are still penetrable enough. In case of success it can be goos means to tame the contrast in rotary processors.

And the 3rd possible way: Pre-bleach and overdevelop. D.Kachel's method of pre-development bleaching (applied to film) alows to "eat" the highlights, leaving the shados almost untouched. In other words in introduces a shoulder. If we then extend the development time I suspect we can essentially raise the contrast in shadow while keeping the contrast in midtones & highlights tamed (how much tamed --it probably depends on pre-bleach intensity). I never tried it, just an idea.

Great topic.

-- Andrey Vorobyov (, May 18, 2002.

Sorry, shoud be: "..1st does mostly A SELECTIVE hardening.."

-- Andrey Vorobyov (, May 18, 2002.

A couple of you have written and asked about compensation development using roll film in tanks vs. sheets in trays. The procedure is exactly the same and the results are equally good. Even better is the fact you can do it with the lights on.

A good starting point would be 1/2 strength developer, using 10 seconds agitation every 3 minutes; increase your development time by double--yes, twice as long. This will yield N-2 to N-3 with most films and developers and it will not reduce EI.

What films work best? I haven't found any film that doesn't work. (A word of caution for those using "old" style emulsions such as Bergger or Forte: I have not tested those films, so I do not know how they would respond to long exposures, either by film speed loss or contrast buildup.) Mostly I use HP5+, Delta 100, TMX and Acros in 4x5, but I've also used it with Delta 400, as well as those afore mentioned films, in 35mm tanks. Acros has become my first choice both for its tonality and its freedom from reciprocity failure (no compensation up to 2 minutes, then 1/2 stop compensation thereafter) and total lack of contrast increase with long exposures.

-- Ted Kaufman (, May 18, 2002.

Clay, the N-6 example of the GW Bridge shot was a 7 minutes exposure on Acros 4x5. I allowed for only 1/2 stop additional exposure. As for shadow contrast increase, I do not think that was a factor. My experience with this film using compensation processing indicates there is little, if any, contrast increase with long exposures.

-- Ted Kaufman (, May 18, 2002.

> when we start experience the failure of reciprocity, the shadow part of the curve may become convex or it may become concave, -- it depends how fast drops the film sensitivity. -- Am I wrong?

If the film is ideal straight line, it is concave and convex. However, if the film loses sensitivity in low exposure region, the actual curve lies below the straight line, making that neighborhood concave. When you look at reciprocity failure at the other extreme, at high exposure, the curve again lies below the ideal straight line, again making it concave. What I think you are confused is that contrast is related to first derivative while concavity is related to second derivative. But let us not talk much about mathematical analysis here.

> may be it makes sense to split the development into 2 baths: 1st does mostly sa elective hardening, the 2nd does mostly the very development?

It makes sense, but we have to make sure that the second developer does not soften the gelatin again.

Even with modern films, highlight areas require increased intake of chemicals and build up more reaction products. It's probably that the absolute amount of chemical needed has decreased compared to the era of Super XX, or gelatin's diffusion became so efficient, or gelatin can hold enough chemicals to survive between somewhat infrequent agitation. A simpler approach, perhaps with lesser effect, is to further slow down the diffusion of chemical across the gelatin. I am more comfortable to let statistics of kinetics control this instead of letting a magic hand controlling fluid turbulence. One rough indication for whether the rate of diffusion is mainly limiting the rate of development is as follows: Take a developer X (1+0), X(1+1), X(1+3). Let t(n) be the time required to maintain certain contrast criterion at 1+n dilution. If t(3) \approx 2 t(1) \approx 4 t(0), development is nearly perfectly diffusion rate limited. When the concentration-time curve is plotted, the closeer to linear with slope 1, the larger the contribution of diffusion rate to the overall rate. (This does not take initial gelatin wetting time into account, so be aware of that.) What does it tell you? Microphen combined with HP5+ is largely diffusion limited. This (HP5+ in Microphen 1+4 or something) might be worthwhile testing with somewhat infrequent agitation Ted suggested, possibly with pre-bleach as Andrey suggested, which I think is a good idea. Microphen should be pretty resistant to bromide drag (because of phenidone at low pH and added bromide). Incidentally, I have been trying a similar idea with different approach for different purposes. I'll report it at another thread, because I haven't considered compensation in that formula.

If there are other suggestions, please post them here. When the thread dies out, I'll summarize, edit and give it to Ed B.

-- Ryuji Suzuki (, May 18, 2002.

The question I raised about the influence of reciprocity departure intrigued me enough that I decided to run an 'end points' test yesterday. I have a perfect SBR 13+ stairwell in my house with a window that has even open shade illlumination from outside. I used some neutral density filters to allow me to make two different lengths of exposure on Tmax400. Exposure #1 was 16 seconds, just barely into the reciprocity failure zone with this film, and # 2 used a 5 stop combination of N.D. filters to make another exposure at a calculated 'correct' time of 25 minutes. My thought is to develop the two sheets of film together in my jobo to the same theoretically correct gradient of .38 to achieve my normal density range of 1.4 that I use with pyrocatHD and my palladium printing process, and see if I can detect any differences in shadow contrast. The negs are 5x7, so I will be able to use a densitometer to pull out some quantitative data from the shadow areas to compare between the two. Does this sound like a reasonable sort of test to run to chase down this idea?

One other compensating development approach that I have read about but have yet to try is the one that Mark Citret espouses for his high SBR images. He has a full description in an article on his web page, but in brief, he uses the extreme dilution approach just suggested by Ryuji with much extended times. His preference is Rodinal at 1:149. His theory is that highlight development needs 'energy' (i.e. chemical concentration) and shadow development just needs time. Check out his web page at He has some great images to browse through. Wanted to emphasize that last part, since all this is great fun, but if it doesn't result in some interesting images, then its sort of pointless.

Great thread.

A totally unrelated question to someone on this thread. What is the solubility of glycin in water? I'm trying to mix up a glycin/ catechol soup and the glycin is very difficult to dissolve. And should the glycin be mixed first, or after the solution has a higher pH?


-- clay harmon (, May 19, 2002.

In my limited experience glycin is slow to dissolve in water - it takes quite a while. OTOH when mixing 130 paper developer I get a wonderful effervescence and very fast dissolving of the glycin - probably all that carbonate that goes in just before the glycin. I can't comment about mixing order since I don't know what else you're mixing.

-- Henry Ambrose (, May 19, 2002.

The few things I've read about glycin indicate that it's very slow to dissolve in water but dissolves readily in an alkaline solution.

-- John Hicks (, May 19, 2002.

I just have a very brief comment to make inspired by something Pat said (and by the way, hello Pat! Glad to hear you're still tooting the old oboe--I miss our phone chats since I left the magazine). With all this stuff, before making ANY assumptions, try a controlled experiment.

I can't tell you how many times I've gotten carried away by one or another of the old pearls of received wisdom, only to find when I gt around to running an experiment with a control that it didn't really make a damned bit of difference. In all my years at _Photo Techniques_ I never saw a single article come over the transom that proved a compensating effect with current emulsions, yet there were plenty of times that I talked authors out of their enthusiasm for it by simply suggesting they do a controlled experiment.

Compensating effects, water-bath development, physical development, even the old saw about Rodinal being "high acutance" (D-76 1:1 has literally the same acutance as Rodinal), not to mention split-filter printing all WORKS, which combines with peoples' enthisasm to create phantom efects...until you simply do the experiment with a strict control. Phil Davis has even proved to my satisfaction that with exceedingly careful sensitometric matching and gradation control, he can make identical prints from D-76 negs and pyro negs. (I couldn't get him to write this up as an article only because he didn't want to take all sorts of shit from the potion-and-magick crowd! )

If Pat Gainer says that stand development makes a difference, by golly I'll try it (though I'd feel a lot more comfortable trying it with T-Max Developer or DDX than with D-76, I'll tell you that! All that bromide...). But remember what Grant Haist said--I'm going on my faulty memory, but it was something like, the large number of development options only multiply the manner in which identical results can be achieved. (That always makes me smile.)

(Haist was the author of perhaps the most comprehensive reference on B&W processing ever published, for those of you who may not know the name.)

-- Mike Johnston (, May 19, 2002.

After reading Mike Johnston's article (http://www.luminous-,

I would like to amend my previous post by substituting the word "photograph" for the word "image" in the two or three places I used it. Mea culpa. It is an unfortunate linguistic 'tic' that I've picked up in the same unconscious manner that I've found myself using my teenage daughters' "Uh, like" construction on occasion. I've also sworn off the word "giclee" as well, and now use the more descriptive "freaking inkjet print". Thanks for your dose of curmudgeon-like clarity, Mike.


-- clay harlmon (, May 19, 2002.

> Compensating effects, water-bath development, physical development, even the old saw about Rodinal being "high acutance" (D-76 1:1 has literally the same acutance as Rodinal), not to mention split-filter printing etc....

You are mixing up different things. Water bath is water bath even if you get no effect and there is no question about it. A transistor radio is a transistor radio, an 8x10 is an 8x10 no matter how crappy it is. Physical development is a physical-chemical process that occurs in certain conditions, and it may or may not be good, and may or may not make visible difference in negative. Compensating effect is the one we are questioning here:

> In all my years at _Photo Techniques_ I never saw a single article come over the transom that proved a compensating effect with current emulsions...

There is compensating effect with current films. It's just that the effect is small and not worth bothering especially if it requires unrepeatably tricky agitation technique, etc. This is particularly so since we now get Delta 400.

Coming back to the first quote, about accutance effect, I would speak more carefully. There are some people who give strictly defined meanings to that term as well as "definition" "shartpness" etc., while some people say "high accutance" whenever they see sharp-looking (probably big) grains. Eastman Kodak Lab proposed a measurement method for accutance, but does that coincide with what we really perceive as accutance? We don't know. Sure, based on EKC method, Rodinal, D-76 1+1, etc., are within experimental error. But I wouldn't assume EKC method is THE accutance or anything. Same for granularity. Some say XTOL is grainer than D-76 and others say opposite. I think they are talking about different aspects of same grain. EKC's granularity measurement method gives you one way to measure, but very different kinds of grains can result in an identical number. I somehow think Phil Davis spoke more carefully - two prints with identical tonal reproduction curves are not necessarily identical prints. They may differ in other aspects, whether the difference is important or not.

If there is a way to do control experiment taking all these into account without making any assumption, I would very much love to know it!

-- Ryuji Suzuki (, May 19, 2002.

Mike, what about Howard Bond's article called "Lowering B&W Contrast" reprinted (I don't know the original publication date.) in Photo Techniques, "Mastering B&W Photography," Vol I. In it he compares six methods of compensation processing.

-- Ted Kaufman (, May 20, 2002.

DJ, of course you're right, the contrast increases, the curve gets concave. I had to grasp it immediately, my thanks to Ryuji.

One more way to introduce the shoulder: bleach the overdeveloped (and fixed) in a superproportional reducer. "Super" means that it affects highlights much more than shadows. Since shadows will probably be also somewhat affected, some loss of film speed seems to be unavoidable, but, unlike the pre-development bleaching, this process is under full visual control, thus no more the Black Art, only the White One.

Ted, if you have the famous N-6 image online, could you please thell the address? Just curious...

Thanks and best regards

-- Andrey Vorobyov (, May 22, 2002.

I just worked up some tests of three films in Pyrocat-HD (metol version) and got some interesting results, some useful, some maybe no so useful.

That "standard" is D-76H 1:1 or 1:3, rotary processing, which yields virtually dead-straight curve shapes with the tested films.

HP5+ takes on a distinct S-shaped curve, showing (for normal midrange contrast) decreased shadow contrast and decreased highlight contrast. Although the shadow compression isn't really bad, it may require more exposure for good shadow contrast. Shouldering begins down in the useful range but is mild and rounded. Speed for .10 DU above fb&f is EI 200.

TMX is straight out to about Zone VIII, after which it rolls off into a rounded shoulder, much more so than HP5+. Speed is EI 64.

Acros is the odd one, straight out to about Zone XI, then abruptly going _upward_ as it does in other developers. Not the slightest hint of a shoulder; in fact, quite the opposite. Speed is EI 50.

The neg color is brown but the stain isn't near as heavy as, say, PMK, and may not have a tremendous effect when printing on VC paper although I do expect at least a little reduction in highlight contrast by the stain alone.

So...a very significant shoulder appears in two of the films by choice of developer alone.

-- John Hicks (, May 24, 2002.

It seems to me that the first step in philosophical discourse has not been observed. Define your terms. I don't yet know what "compensation" means or what compensating development hopes to achieve. If it only refers to producing a "shoulder" on the H&D curve, define where that shoulder should appear. All films must produce a shoulder somewhere along the curve. Density cannot build to infinity.

-- Patrick A. Gainer (, May 25, 2002.

Patrick, what makes you say something very contradictory to what you said earlier?

Also, I would very much like to know if you can define where the shoulder appears in your limited agitation method.

-- Ryuji Suzuki (, May 25, 2002.

I define compensation/compensating development as that which reduces or limits highlight density and contrast via local developer exhaustion, whether by divided development or reduced agitation, and the compensating effect is dependent on developer exhaustion.

This is imho distinct from limiting highlight density/contrast through other means.

For me a "useful" shoulder begins within the range that could normally be printed in a straight print, eg. around an exposure that would normally produce a zone VII or VII tone in a print. A shoulder that begins five stops higher wouldn't be useful to me, nor would a shoulder that's so abrupt that d-max is reached too low.

Actually I don't usually encounter scenes of such a high SBR that such shenanigans are needed, but they do occur every now and then. The object is to get relatively normal shadow and midtone contrast while retaining some tone and contrast in non-specular highlights without having to resort to intricate burning.

-- John Hicks (, May 25, 2002.

John, since you usually prefer D-76H over PMK on the basis of grain, I like to hear your comments on this aspect of Pyrocat-HD. Thanks.

-- Ryuji Suzuki (, May 25, 2002.

In just looking at the HP5+ negs with a loupe they were grainier but didn't look bad. Printing will tell, of course, once I get around to shooting some "real" test photos.

-- John Hicks (, May 25, 2002.

I should have said I do not know what compensation means to you gentlemen. Stand development is not my technique by any means. Perhaps I apply it, when I do, in unorthodox ways. I have no desire to produce a shoulder within what might be called the normal log exposure range plus an allowance for possible overexposure. I would like to straighten the curve of film developed to less than normal contrast so as to keep the speed point as close as possible to that for each of the curves for other contrasts. This does not happen with Rodinal and most films when it is used as the manufacturer expects. I apply stand development to undiluted developers mostly to show that it can be done and also that it has the effect I want. I don't think I contradicted myself. I did contradict just about everybodt else, I suppose. I have no desire to peoduce a premature shoulder, and in the days when I first heard the term, neither did anyone else. We were all concerned with raising the toe.

-- Patrick A. Gainer (, May 25, 2002.

To any or all, I did a test on Tri-X 400, stand developing it in D-76 1:1, and noticed that the emulsion side felt rough. Thinking I had caused reticulation somehow, I looked more closely and saw the bas relief of tanning development. I don't remember this happening before with D-76, at least not to the extent that it felt rough. Is this a common experience? There is no sihn of reticulation in the prints.

-- Patrick A. Gainer (, May 26, 2002.

OK I understand where you're going with it. I'd look towards a phenidone developer to lessen the speed loss.

Phil Davis concocted something called DI #13 which I believe is intended to do what you want. I've never tried it. Anyone?

-- John Hicks (, May 26, 2002.

Disregard my post about the tanning. Yes, there was some, but no more than one would expect from a developer with hydroquinone in it. Now if I improve my drying procedure, I won't get excited over specks of dust like the ones I saw when I looked more closely. My main purpose in this excercise was to find out if stand development really had to be done over periods of hours with developers that were more drinkable than some tap water I have encountered. I think the answer is no, it can be done with any developer including PMK pyro. The effect I see was described in "Principles of Optics" by Hardy and Perrin in the 1932 edition as the Eberhard Effect.

-- Patrick A. Gainer (, May 26, 2002.

I developed the film from the totally unscientific SBR 13+ (N- 6)test last weekend in which I used ND filters to create 2 vastly different exposures on the same film to assess whether reciprocity effects helped to increase shadow density in an actual pictorial situation. The bad news is that I ran the test on Tmax 400, which apparently is one of the only films NOT to exhibit contrast increase with longer exposures according to my research from digging through some old Phil Davis articles. Had I known this, I would have done the test on FP4 or something similar. At any rate, I made 2 exposures at each of the exposure time endponts, 2 at 16 seconds and 2 at 24 minutes (with the ND filters) and processed one set of different exposures in pyrocat for 8 minutesand one set in Rodinal 1:150 for 14 minutes at 75 degrees, both in Jobo expert drums. Both yielded printable negatives, but the Rodinal clearly had better shadow contrast, for what its worth. Has anybody on this thread done any analytical tests as to whether extended development in very dilute rodinal does what we apparently are after here, ie good shadow contrast with highlight 'compensation' or shouldering?

-- clay harmon (, May 26, 2002.

I noticed an error in the previous post. I meant to say ' shadow contrast'', not shadow density in the beginning of the post. Need more caffeine.

-- clay harmon (, May 26, 2002.

Caffeine doesn't help my mental alertness...

I haven't done experimental analysis, but it makes sense that you get higher shadow contrast with p-aminophenol than with phenidone in dilute solution with very little sulfite. p-aminophenol will receive more restraining effect of halides whereas phenidone is less susceptible to restraining effect. I think this is another reason why phenidone developers tend to be less sharp than metol counterparts. If you make pyrocatechin alone developer, or pyrocatechin-metol developer, my guess expects to see higher shadow contrast.

-- Ryuji Suzuki (, May 27, 2002.

I have a War Department Technical Manual of Photography dated July 1, 1941 that has this to say about Para-aminophenol hydrochloride. "The rapidity with which a Para-aminophenol developer will develop an image decreases rapidly as the solution is being used. This can be prevented by adding, at intervals, small amounts of accelerator to the solution." Interesting.

-- Patrick A. Gainer (, May 27, 2002.

I hear this often, that phenidone is less sensitive to bromide than others. I read once that in fact, the activity of a phenidone developer increased with bromide concentration up to about 1 gram/liter. Does this same rule apply to superadditive combinations of phenidone with other agents like hydroquinone, ascorbate, etc.? I seem to get restraining action in PQ print developers with KBr.

-- Patrick A. Gainer (, May 27, 2002.

Patrick, your War Department Technical Manual isn't kind enough to say why. I think the most likely explanation why p-aminophenol's rate of development declines is because its oxidation product is acidic, and theby requiring external source of electrons.

The low restraining effect of KBr would apply to superadditive combinations with hydroquinone and some other reductants. It really depends on the affinity of the developing agent at the working pH to silver halides, water, and KBr. Bromide ion likes to adsorb on silver halide particles. This slows down working of HQ, ascorbate, etc. but doesn't bother phenidone. HQ at pH of 9 is mostly undissociated and won't reduce silver halides directly to any appreciable degree. Ascorbate almost entirely dissociate one -OH and is active, but for some reason will not develop silver quickly. However, ascorbate reduces things dissolved in water effectively. The reason I think is because ascorbate has high affinity to water and doesn't want to bump into silver halides, and even if it does, it won't adsorb onto it. This is very different from phenidone.

However, in print developers, condition can be different. Silver halide compositions and its physical structure can be different; sulfite is usually moderate; higher pH; etc. Plus, you usually add more bromide to paper developers. Some visible restraining action is not too surprising.

-- Ryuji Suzuki (, May 27, 2002.

Ryuji, Now it's even more interesting. Perhaps that makes p-aminophenol HCl doubly good for the kind of compensation I like. It is sensitive to pH and bromide and, beside the lowering of pH in the fashion of other developers its own product of oxidation is acidic. I have been playing with it in a simple combination with ascorbic acid in a stock solution, activated at time of use by carbonate. I can get a 5-6 minute working solution at 1+50 dilution. No Rodinal is involved.

-- Patrick A. Gainer (, May 27, 2002.

> very dilute rodinal

My curve shapes for several films in Rodinal 1:100 are pretty much dead straight, no compensating effect at all. Note that this is for rotary agitation; I'd expect reduced intermittent agitation, especially with higher dilution, to induce some compensating effect.

-- John Hicks (, May 27, 2002.

Moderation questions? read the FAQ