View Full Version : A Q for Bill re: Pulling Film
We've all heard the maxim " expose for the shadows and develop for the highlights." In simplest form, this should mean "pulling" your film i.e. overexposing and reducing development times.
Given this, why don't b&w photographers simply rate films at half the box speed and develop accordingly? What, if anything, do we lose in doing this? does this compromise something else e.g. dynamic range? Tonal seperation? that films are optimized for at box speed?
I ask because this 'seems' to follow logically from such advice, but yet pulling film is much less prevelant than is pushing it.
I ask you because you taught me everything I know back in the 70s when the Leica Manual was my Bible. Thanks BTW, Your knowledge was/is an incredible resource to many of us.
Not Bill replying, but no, "expose for the shadows and develop for the highlights" does not mean overexposing and reducing development times.
It means pretty much what it says: base your exposure on the shadows (ideally using a spot meter, because that is the ONLY way to know what the shadow brightness is) and then develop for the tonal range. In other words, with a short tonal range (highlights not much brighter than shadows) you develop for longer, while with a long tonal range (highlights a lot brighter than shadows) you develop for less time.
If you use a spot meter (and the appropriate shadow index) you will almost invariably find that films can be rated at box speed. Overexposure is normally a 'fudge' to take care of the fact that most metering systems don't accord sufficient weighting to the shadows, especially in subjects with a long brightness range. I read somewhere that Ansel Adams reckoned his exposures increased on average by a stop when he got a spot meter.
The penalties for overexposure are coarser grain and reduced sharpness, though both are slightly ameliorated by the fact that reducing development time will slightly reduce grain and increase sharpness. It will also lower contrast, so unless you're dealing with high-contrast subjects, you may find your negatives very flat and hard to print.
And, of course, halving the film speed gives you less depth of field (= wider apertures) or less action-stopping (= lower shutter speeds) or both.
Most of us are shooting 35mm or roll film. That means that there will often be a variety of shots of different tonal ranges, contrasts, whatever on a single roll of film. It’s also the nature of small format photography that we are often making quick grab shots, not doing a very good job of metering because the picture, the moment we want, is about to disappear. So, we have to come up with a development time that is a compromise for all those other compromises.
In general, that’s going to be less developing time than a manufacturer thinks produces the best negative for enlargement, presumed on a diffusion enlarger. If your negatives are destined for scanning instead of conventional wet darkroom printing, that may mean you want a negative with even less density range. There are two relatively straightforward solutions to these problems.
(1) A development time for silver film that is usually under the manufacturer’s recommendation combined with generous, shadow oriented exposure of scenes with a long tonal range. Probably the best way to determine the actual numbers on this method is trial and error over a lot of rolls of film with a large variety of scene types. We’ll call this the Roger System, presuming Roger is stuck with a variety of brightness ranges all photographed on one roll of film.
(2) Shoot one of the CN films like Ilford XP2. This is automatically going to give a negative with limited density range. It’s good for scanning, not only because of the density range, but because non silver films respond to some of the scanner programs that eliminate dust spots while silver doesn’t. You can, of course, shoot color negative, scan and computer print, deciding when you print whether you want black-and-white or color. Regardless of how you print it, how you meter (incident, reflected, TTL auto and prayer) it, what your EI is, e.t.c., is something you will work out over time. It will be different for different folks. But, in general, it too will lean towards a generous, shadow and dark tones oriented exposure in contrasty situations. We’ll call this the lazy boy or Pierce System.
Obviously, different strokes for different folks is an understatement. Everybody works in different ways with different gear and different opinions of what the final image should look like. The real key to any system is realize its weaknesses and strengths by shooting a lot of film. In other words, make it your system.
Your point about providing a 'buffer' against underexposure is of course spot on, but I took the original query slightly differently, with the assumption that there was time to meter, i.e. you can afford to 'expose for the shadows' rather than just 'expose and hope', which is, as you say, all we may have time for with many shots.
'Exposing for the shadows' also explains why I'll set different speeds on different meters for different metering techniques, while still rating the film at exactly the same speed. On a sunny day, shooting HP5+ developed in DD-X, I might use EI 500 on the spot meter and EI 250 on a broad-area reflected light meter.
And, like you, I am a very great fan of XP2 Super, and I'll add another advantage you didn't mention: no Callier effect, i.e. no silver grains to scatter the (highly collimated) light from a scanner, so far less risk of blown highlights. The main objection to C41-developed XP2 Super, as far as I can see, is that it doesn't allow obsessive-compulsives to piddle around with silly developers.
One of the ways I would work with the different contrast ranges produced by different light sources was to use an incident meter and take two readings. The first reading would be conventional; the second I would cast a shadow on the meter’s hemisphere with my hand. I would use an exposure in the middle of the meter’s two recommendations. Under flat lighting there would be little difference in the two readings. Under contrasty light like bright sunlight, there could be a 4 stop difference, and I would end up giving 2 stops more exposure than a conventional incident reading would recommend. Obviously, the film speed set on the meter would be higher than the manufacturer’s recommendation. It was something determined by experimentation, but Tri and HP5 might be rated at 650. It was a good rating for flat scenes, but the double reading in harsh sunlight could give an exposure of 1/500 at f/8, considerably more exposure than you would get with almost any conventional metering technique.
Does that make sense?
vBulletin® v3.6.8, Copyright ©2000-2013, Jelsoft Enterprises Ltd.