Film exposure latitude: A practical test

Film exposure latitude is a well-understood concept among film nerds. The problem is that there are really two closely related but different concepts involved in the discussion, and sometimes they’re mixed in a single breath.

From my perspective, the two concepts are:

  1. The ability to shoot a(n entire) roll of film at a speed other than box speed — either higher speed (push), or slower speed (pull) — and then adjust the development times to ensure usable negatives. (Underexposed, “pushed” film needs more time in the soup, and overexposed, “pulled” film needs less, in order to get good results.)
  2. The ability of a single roll of film to yield usable results when your exposure isn’t quite right on any given shot — whether under- or over-exposed intentionally, or accidentally.

I could make a case that there’s a third option here as well: The ability of any given film to deal well with dramatic differences in lighting within a single shot, i.e., areas that are very bright, and areas that are very dark. But in my view, that’s part-and-parcel with the latter above.

In any case, while the underlying latitude qualities of the film may be the driver for both of these situations, they are still separate and distinct in my view, since the former customarily involves changes in the darkroom, while the latter does not.

There are reasons why you’d shoot an entire roll at a speed other than box speed, and people who do this cite differences in grain, differences in contrast, and other factors, believing that they get better results from film “X” if it’s shot at speed “Y” for example. That’s fine; I’m in no position to argue. But to be honest, I have a hard enough time juggling everything else about film photography without trying to decipher whether I know something that someone else doesn’t. If I want a different box speed, I choose a different film — but that’s me.

I do find, however, that I struggle with getting the exposure just right, despite using an exposure meter, and knowing perfectly well how to use it. Because I don’t have a spot meter, it’s difficult to know precisely what the ambient meter is “seeing” in a given situation, and sometimes incident measurement isn’t possible as an alternative — or doesn’t work well in the situation either. That means I depend on a film to sort of cover my mistakes, which in turn has made me profoundly curious about the former item above: Just how much latitude do you have to miss the exposure mark?

After reading the following passage from Foma’s data sheets for Fomapan 100, 200 and 400, I decided to do a test. Here’s the extract from the Fomapan 200 data sheet, but the language is identical (other than speed) for the others:

FOMAPAN 200 Creative has a nominal speed rating of ISO 200/24°, but due to its wide exposure latitude the film gives good results even when being overexposed by 1 EV (exposure value) (as ISO 100/21°) or underexposed by 2 EV (as ISO 800/30°) without any change in processing, i.e. without lengthening the development time or increasing the temperature of the developer used. 

That pretty much says it all, really. So to put this to the test, I decided to load my Rolleiflex Automat MX with a roll of Fomapan 200, and give it a try. At the end of the day, this is nothing more than exposure bracketing — it’s just across five full stops, from +2 to –2.

To be honest, the concept of exposure bracketing is something I’ve known about for a long time, but I’ve just never actually done it. Shooting film has a cost, my shots have never been a life-or-death sort of deal, so I take one shot. If I get it, great; if I don’t, well, I lost out. But actually doing it for this test was insightful and interesting.

The negatives were developed for 8:00 minutes in HC-110 Dilution H; I picked that time because it splits the difference between what was quoted in the Massive Dev Chart (9:00), and the doubled Dilution B times (3:30 x 2 = 7:00).

The PrintFile sheet looked like this on the light table:

(NOTE: Disregard the other images on the sheet; those were tests of something else for a future blog post.)

The settings were as follows:

  • +2: f/22 at 1/30
  • +1: f/22 at 1/60
  •   0: f/22 at 1/125 (Metered Values)
  • –1: f/22 at 1/250
  • –2: f/22 at 1/500

Now, let’s look at how these images came out of the scanner, in the same order (by exposure, not the order I shot them in), from overexposed to underexposed (and forgive the fact I was struggling with the tripod a little, so slight differences in framing here):

+2 Over

+1 Over

0 (Zero)

–1 Under

–2 Under

The results are probably about what you’d expect:

more exposure = darker negatives = lighter positives


less exposure = lighter negatives = darker positives

You can easily see the progression.

The shadowed front façades of the houses, and the tree, are the two main areas where I see a big difference, and if I’m honest, the +2 image is actually my favorite. Now, that could be down to poor metering on my part, but in truth, I’d argue that all of the images are usable. While it’s tricky to adjust the –2 image in Photoshop to fully restore the façades without adversely affecting the rest of the image, it can still be pretty easily improved and made usable.

So what’s the takeaway here? Exposure is still important, metering is still important, and if you want optimal results, you can’t just be sloppy about it. But latitude is appropriately named; you do have some latitude for mistakes, and additionally, you have some latitude to achieve different creative results when the need dictates.

All in all, it was a fun and worthwhile little test, and it’s clear that what Foma says in its data sheets appears to be accurate — knowledge that will come-in useful in the future.