Michael, there is no magic developer. You have the triad of speed, grain and sharpness and you can only improve on 2 at one time. So, you are technically barred from reaching the peak of this pyramid.
As we worked on imaging even earlier than the '60s when I arrived, it was apparent that the knife edge was self limiting as it only gave information on the boundary of an image at infinite width on either side. But pictures are made up of bounded areas and thus the slits came into vogue. They gave far more information and still allowed the examination of the edge effects. Kriss published his first work on this in the '70s and work has since sped up. The reason? Total image content is a very important method of comparison with digital. With aliasing, and R/G/B separations it is the content of the image which is important.
Kriss' work is based on the work of Higgins, Perrin and others, but is the next step forward from the early knife edge exposures. I cannot hope to explain this unless I scan in the whole article and post it. It is packed with a lot of calculus. All I can say is that we used the methods I described and out the other end came figures relating to overall image quality based on these mathematical geniuses.
As Dr Henry pointed out the word acutance (A1) was originally as defined in the early Kodak work but is now also used (A2) to mean perceived sharpness which combines this with adjacency effects (A3).
To define (A3) would need another 4 numbers to show the height and slope of the fringe and border effects.
(A2) for each film developer combination would then involve 5 numbers. IMO nobody would use such a complex label.
PE: Interesting that this work continued as a result of digital.
Absolutely agree there is no magic developer (I'm actually trying to simplify the choice a little). However many things are said, not only by laypeople but by people formulating these things, and there are some claims I just don't think hold much water based on the mechanisms at work. This seems particularly to be the case concerning "sharp" developers such as Rodinal, Crawley's FXs and tanning/staining formulas. Not a new story I know, but I wanted to approach a few image structure characteristics from a critical perspective. Sharpness (or definition) is complex, but as far as the influence of developer/development goes, I'd like to question the concept of traditional acutance - ie the idea that solvent action per se has any significant influence on sharpness.
I agree introducing a lot of calculus is probably beyond what many people will find interesting so let's leave that for now (unless people want to get into it).
Alan: Why would A3 be so complicated? It is basically the idea I proposed (simplistic) but with more "factors" to account for the 4 gradients of the edge effects individually and separately from the gradient G2x in A2. Actually this is helpful since it points out a flaw in my simplified formula (ie no consideration of the edge effect gradients).
Michael, just FYI, Crawley's early developers with iodide worked because the iodide adsorbed to the early bromide emulsions and then was released during development to cause edge effects. In modern, high iodide (up to 10%) emulsions, the Crawley developers have little or no effect.
I have been discussing this with Bill Troop. I am not sure he agrees but he was on board with a positive reaction last we discussed this.
That would be pretty important given how much there is about Crawley's developers in the FDC. On the other hand Iodide was only in the FX1 formula as far as I know (although it may have been in some of his Paterson FX and/or Acu developers).
What is your view on current films (including tabular-grain emulsions) vs older style films when it comes to how readily they produce edge effects in general? Some people have said films like TMax for example "resist" the formation of edge effects even in highly dilute developers. Is there any truth to this?
Edge effects are one thing, "grain etching" by sulfite is another. That's the effect I question when it comes to the relative acutance (ie excluding edge effects) of low pH solvent vs high pH non-solvent developers. Simply put, I don't see how the grain etching argument makes much sense when it comes to image structure. For one thing the scale is too small if we're talking about individual grains. Then there is the fact the shape of a grain is only barely maintained under typical development circumstances. Then, even where the overall shape of the grain is maintained, most development results in the formation of filamentary silver, so the edges of what was the silver halide grain have a rough appearance anyway.
Crawley called the old days the acutance era, adjacency effects provided higher perceived sharpness.Films that showed the most effect are now discontinued though new Adox IDK.
Newer tabular grain films have higher acutance according to the early Kodak definition,adjacency effects are less needed to give perceived sharpness.
I have seen so many comments on these two post's subject matter that I would have to say: "run some experiments". Each film responds to each developer in a different manner. Each built in feature is described in the press by each company in a different manner.
So, lets just take the case of a film with very coarse grains but with very large edge effects. Is it very sharp or very grainy or is it to be described in another fashion? Each person here on APUG would probably evaluate the film in their own special brew and then "name" the film's features after what they find with their ever so precise (subjective or objective - which do you think is done more often?) tests.
I have not been involved in direct tests on an actual film for years. Remember, I did see emulsion tests which in my later 15 years at EK but for the first 15 years, I did a lot of film tests. These differ as the raw emulsion lacks many of the refinements that the final product does. So, when I tested a film, it had everything, but when I tested an emulsion, it lacked the acutance dyes and many other addenda used in a real product. We made something close to 2000 emulsion of different types every year in KRL, and they could not all be tested in a product environment. In fact, some were designed to be just components of a product and not the whole thing.
However, one raw emulsion can still be tested against another raw emulsion and the results can be fair in their assessment.
That said, if an emulsion was earmarked for greater things it was given the battery of tests that I outlined above involving slit light and Xray exposures and a lot more.
I guess that's the problem. Even in the thread Alan linked to, it's about 99% nonsense, especially once you get Rodinal and staining/tanning people involved. But I suppose I'm differentiating actual sharpness from characteristics that can enhance the impression of sharpness. Edge effects, for me, fall under the latter category. With a given resolution, acutance (excluding edge effects) has mostly to do with the film itself and exposure, so turbidity, emulsion layers and thickness, homogeneity, acutance dyes, anti-halation and other things. Then there is the exposure itself, which of course causes some degree of irradiation etc. depending on the film characteristics. These things determine the actual acutance. I don't think developers can do much with that. I think they mostly influence graininess (as evidenced in the Altman/Henn study), and to some/varying extent edge effects. And I also think when people see pronounced grain they see sharpness.
Originally Posted by Photo Engineer
I also agree in the end generalizations are tricky. The Altman/Henn study and Henry's tests showed this. Who would ever expect stock D-25 to be sharp? Well it certainly seemed to do just that with Panatomic-X.
Another question (ultimately in relation to the sharpness/grain masking claims made by people formulating tanning/staining developers). I think I remember reading somewhere in Haist (can't seem to find it now though) about additives in colour emulsions to prevent dye spreading - in other words to keep the dye from forming too large a cloud around the silver. Am I dreaming that up or is it real?
Yes, additives are present in color films (and sometimes in color developers) to prevent image spread. If they are not present, the dye clouds can be too big. In fact, in the absence of these chemicals, you can have dye contamination between layers. You can get a magenta dye cloud become so big that you get cyan and yellow dye clouds in adjacent parts of the coating. So, color is a more complex best. A common developer additive is Citrazinic Acid (E6) and Sulfite (All CDs).
Now, back to your observations about B&W. The example I chose was done on purpose. You see, grain can interfere with sharpness. So, you can prove on paper that a film is sharp, but when you see the grain it is similar to aliasing in digital.. Yep, quite similar in some aspects, but not overall. So, this is why Kriss came up with Image Content. Or, how good does a given image look when you combine gran with edge effects and vary their contributions to the quality of what you see. This can be reduced to a mathematical formula which is now in use at EK. This treatment will single out the optimum combination of multiple effects such as halation and light scatter and add them to the mix of grain and edge effects.
But, there is more to this than the above. You see, modern B&W films are made up of 2 or 3 components and modern color films have up to 9 imaging emulsions. Each of those will respond differently, and thus we must look at the edge effects of the slow, medium and fast components for example. And that is folded into the grain of each. And these are folded into the overall result of the full multilayer. This is a lifetime of learning when you build a film.
A lifetime of learning indeed. I wish I could tour Kodak or Ilford and see how they actually test these things.
On the subject of the additives to prevent image spread, what are the implications when it comes to the acutance of stained/tanned negatives? This has always confused me. I've mentioned this before but would be interested in your thoughts. Formulators of modern staining formulas make certain claims:
1. Less graininess than non-staining high acutance developers due to lower silver densities. Makes sense to me.
2. Less graininess than non-staining high acutance developers and "smoother" tonality, both due to the dye filling in the spaces between developed silver particles (ie grain masking). Perhaps correct - but this is image spreading, so if this is the case to any significant degree, resolution and acutance must be degraded.
3. Tanning results in a somehow more precise development and formation of silver, limits the effects of irradiation, and limits diffusion. If tanning limits diffusion, does it limit diffusion of the developer, bromides, oxidation products/dye, all of these or some of these?
There appear to be some inconsistencies in the common descriptions of how these developers actually work - sort of like having your cake and eating it. For example, the developer gives high resolution and acutance, and more precise high density development due to low diffusion, but also encourages pronounced edge effects and masks graininess due to dye spreading (implying diffusion). And in colour systems it would seem attempts are made to prevent dye spreading.
Unfortunately these mechanisms are not discussed in great detail in volume I of Haist. Tanning developers are discussed mostly from the point of view of relief images.
As an aside, I hadn't noticed this before but there is at least one reference to one of your patents in volume II of Haist. :)