acutance, resolution, grain, scanning

There was a post awhile back about whether folks develop for acutance

or minimal grain when the purpose is to scan the film. That got me

thinking (which can be dangerous).

 

My understanding is that acutance is not the same thing as

resolution. That is, resolution is how many lines per whatever the

film can resolve, and acutance is about how sharp those lines

appear. But low acutance does not translate into lower resultion, at

least with the vast majority of developers. This is because

resolution has to do with the grain structure, and acutance has to do

with the contrast of the grain. I could be wrong about all of that,

so someone correct me if I am.

 

But if all of that is correct, and one is worried about balancing

sharpness and grain when scanning, wouldn't it make the most sense in

developing to get as fine a grain as possible even a the expense of

acutance. This is because acutance will have nothing to do with what

the scanner can actually resolve. In other words, lower acutance will

not mean lower resolution, and thus no "loss" of info when scanning.

But lower acutance will (I'm assuming again) lessen the affect of

grain. Acutance can be digitally altered with unsharp mask or some

other sharpnennign tool, which of course adds nothing in the way of

resolution. But that's no problem, because acutance doesn't affect

resolution, and hence doesn't affect hwat the scanner can resolve.

 

So, if one is scanning and worried about grain (this might not apply

to wet work, or those who like the look of grain), why WOULD one

develop for acutance/sharpness if that doesn't affect resolution?

I can only speak from exerience. I know that when I develop in Rodinal 1+50 then scan I do get a sharp image. It's also an image that can be sharpened further in PS without unpleasent side effects. When I develop in D76 then scan, I get richer tonalities. More or less what you would expect if you printed.

Here's an article that may interest you:

 

http://www.shutterbug.net/features/0703sb_six/

I find that Rodinal 1:50 is entertaining in wet prints, less so in scans. This is probably because wet materials evolved over the eons in a way that happened to be Rodinal-friendly. Darwinism, not Tammy Faye-intelligenceism.

 

I like Tetenal Emofin (a split, compensating developer) with fast films. It produces a flat negative that I wouldn't want for silver (would require grade 4). I push a stop or two (800 or 1600), scan with "slight grain reduction," Epson print "black-only," get well-defined grain with very high resolution. Looks sharper and more attractive (to me) than solvent-developed silver film (eg D76), wet printed.

 

Velvia (no grain) arguably looks as good, desaturated and printed with

Quadtone driver (no-grain looks better with a tone IMO than it does in neutral blacks), but I don't like being dependent on E6 labs and I like ei around 1000.

I think your reasoning is quite valid. In my experience the best film/developer combos for scanning are those that give little or no grain. And I find it hard to scan well grainy negatives.

 

Acutance developers may be relevant if you use the negative for wet prints. As you say, acutance may be easily added after scanning with unsharp masking.

I think you're on the right track. Think of accutance as amplitude of MTF, resolution as length. It's easier to alter the amplitude by digital postprocessing, within limits, as length is fixed. (I prefer the wet darkroom for BW but digital for color)

I think you are refering to a thread I started a few months ago. Your reasoning mirrors my own. Latelely, I've been trundling along by habit (HP5+ and Delta 100 in HC-110, APX100 in Rodinal, etc.), but I intend for my near future more intentional development experiments to center around solvent developers like XTOL and Clayton F76+ and films like Delta 100, Efke 25 and so forth.

 

Another factor to strongly take into consideration - negative density. My own datapoint here is HP5+ @ EI 800 in HC-110 (H). The first time I did this, and ever since, I got remarkably smooth scans. Sometimes, shadow detail would be a bit shy, which is understandable. But the scans would be quite smooth. I can only conclude that the low negative density is responsible for the nice scans.

 

A 2nd datapoint was my own errors in temperature control. For whatever reasons, my kitchen is quite warm. In July and August, I discovered that the developer temperature was quickly skyrocketing from an initial 68F to 72-74F while in the tank. The denser negatives were very grainy. Cooling the tank and using an ice water bath, in addition to my normal habits of rather minimal agitation for lower contrast negs, resulted in thinner negs that scanned wonderfully.

 

In conclusion, I imagine that a slower film, a slightly solvent developer, low contrast negs and more significantly, a thinner negative will all yield nice scans that I hope will support digital printing at 11x17" with low grain and smooth tonal transitions (the latter is something that mostly eludes me to date at these size enlargements).

 

Now, I just have to get off my duff and see if my theory is right :-)

 

Scott

I'd be interested to see what your results are Scott. I'm fairly new at BW development, and I just developed my first roll in stock D76. I'd used it diluted 1:1 before. I doubt there'll be a lot of difference, but stock is supposed to yield somewhat finer grain. I'd be curious to see your results in scanning with finer grained developers. I'll scan my roll tonight and see if I can tell a difference from the negs (same emulsion) developed in 1:1