I have not presented pictures of it (yet? we’ll see), but it is evident that I have problems with the amount of exposure I get with the large format camera. Often the negatives came out on the thinny side. Sometimes almost unusably so. And some were way too dark. All in a haphazard way, very confusing.
My first test of this phenomenon was through a stillebein which was photographed with varying shutter speeds, from 1s up to 16s. The longest time, in particular, would make the negative dark I thought. But not so. They were almost the same density despite the different exposure times. The phenomenon of reciprocity reared its ugly face here: Beyond about 1 sec, film reacts very slowly to further exposure, and just to complicate matters in a highly non-linear fashion (different films also have different reciprocity behaviour). Suffice it to say that all negatives were about as thin as the first one. I learned to keep exposures at below 1 second, always!
At faster speeds the cause of these exposure problems was that the speeds of the Press Compur shutter were, ahem, “varying” – some slower speeds were in fact fast, and vice versa! A thorough disassembly and cleaning helped a great deal and the speeds are now at least in the correct order. However the speeds are still generally a bit off compared to the markings on the speed ring. This was solved in a (c)rude way: New markings on the speed scale… It turned out that by moving the speed cam a few millimetres past the mark, the actual speed was achieved. So the shutter now has small marks for the actual times of 1s, 1/2s, 1/5s, 1/10s, 1/25s and 1/50s. By luck the 1/100s mark was more or less exact and the 1/200s was almost 1/100s too (at 1/200s the extra accelerator spring must be getting old). With the shutter speeds in order, the exposures were more consistent.
The other day I decided to do a film exposure test, to get a feel for film and its allowable exposure tolerances. To keep it simple I did this test with a GW690II and Fomapan 100 film. I do not have access to a densitometer, for measuring the amount of black density of the film, so the result has to be analyzed by eye only.
A start value of the exposure needed was made with my Canon 5D2 set at ISO 200. I took a reading of the entire scene with highlights just below saturating the digital sensor. This is called “+/-0” in this text. From this start value a series of exposures were taken at +2.5, +1.5, +0.5, -0.5, -1.5, -2.5 and -3.5 stops. The reason for the “half-stops” is that the shutter speed setting on the GW690 happened to have these numbers.
The photos in the test run are presented below, with exposure compensation numbers under each photo (all at f/8):
In the dark area, to the right of the chin in the picture, there should be some detail if the exposure is just about right. This happens at an exposure compensation of +1.5 (on my screen anyway). Below this value the area is pitch dark. The brightest part of the picture, the reflexions in the apron, always come through, no matter the exposure. With an exposure compensation of about 1 stop the scene’s dynamic range fits well onto the film. (A digital camera’s histogram conveys exactly this information but in a graphical manner.)
Since the film needs some overexposure a conclusion to draw is that the film acts slower than box speed. Or is it? If I set the digital cameras ISO to 100 instead of 200 the exposure reading from the camera could be used immediately. Why does this not come as a surprise? Well, it is because the digital camera used was set at ISO 200, and the measurements were of course 1 stop off… The film had a ISO 100 sensitivity! Bummer!
To conclude: Exposure estimation of a scene with the digital camera to just below clipping is exactly what to use for the GW690II loaded with Fomapan 100. This camera’s shutter speeds are spot on, I have tested them. The result comes as no real surprise, but is good to know.
Another conclusion is that the film is tolerant to overexposure. Look at the +2.5 picture, that has an overexposure of 1-1.5 stops and is still perfectly usable! The film does not take underexposuresgracefully though, there is substantial difference already between the underexposures rendering these dark negs unusable!
At a closer look it is also evident that a light negative is more prone to dust, the slightest dust particle (they always scan as black stains) shines out like a beacon from the black background in the resulting picture. I wouldn’t be afraid to allow another stop just for good measure. These results are not automatically transferrable to the large format camera I started off with, for a number of reasons:
- The LF camera uses another film (Forte 400 now)
- The LF cameras shutter speeds are (somewhat) different
- Depending on the bellows extension some extra compensation may be needed
It was however an interesting experiment to see how film reacts to different exposures. With a densitometer one should also be able to sketch the film’s response in the form of an S-curve. While that would be a scientifically interesting measurement, just looking at the pictures is good enough for now.
PS. If you know your GW690II, you know there should be one more photo before the roll is complete (a 120 film has eight shots in the 6×9 cm format). And you are right. I also did an eighth shot, a self-portrait, of which I cropped this:
PS/2. I can’t help it but the gnome’s half-witted smile scares the shit out of me… Gnomes are EVIL!