Color Separation Test
I've been thinking about this project for a while. I find that often, after figuring something out either as an in-camera technique or as a printing technique, I get excited about figuring the same thing out, but in the opposite situation. Sometimes this is a natural progression, like when we moved in class from learning about matting in camera to matting using an optical printer. Other times it moves backwards, as in this instance.
I'd been thinking a lot about color replacement in black and white images, and color blending through multiple exposure - all on the optical printer - when I realized that I could very easily, and using the same filters, do something similar in-camera. It would work like this: make sure you have a tripod, close your f-stop by 1.5 stops for a triple exposure, place a color filter over the front of the lens, (and compensate for that as well,) shoot, rewind, place a second filter over the lens, shoot, rewind, third filter, shoot. If I wanted the colors to combine realistically on top of one another, I would have to use either red/green/blue or cyan/magenta/yellow combinations (I use both variants in the film above). Hypothetically, this would leave you with an image that looked completely normal wherever the subject matter was standing still, and ghostly and psychedelic wherever it wasn't. If there were objects in a different place during each exposure, they would appear multiple times in different colors.
That's pretty much what I did to make this test film. It's 100ft of 250D, shot on my Bolex Rex-5. Some things that didn't go as planned:
-First of all, and this is glaringly obvious if you watch the footage, the tripods I was using totally sucked. I was home for the holidays and borrowing tripods from very generous friends, but the tripods were meant for small digital cameras and couldn't handle my fat fatty Bolex. That's what causes the rampant misregistration. I was aware during the shooting process that the image probably wouldn't register very well, and I'm glad I did it anyway, because in certain shots (the chairs I like especially, and the goats) it works very nicely and actually adds some interesting compositional qualities - like the multiple arms of the chairs on their sides.
-Secondly, the filters I was shooting through were filthy. That's what causes the ghosty blurs you see moving lumpily around. I borrowed filters from the optical printing room before break (shhh, don't tell) and they are full of crap from grubby fingers and warped from the heat of the bulb.
-Thirdly, the lenses I was using were all old, some in more questionable condition than others, and all non-reflex. As of now, I'm educated as to what to do in such a situation, but at the time I still believed the myth that non-reflex lenses will work on a reflex camera as long as you compensate for the light loss by opening up 1/3 of a stop. So, I was compensating for this myth, the filters, and my triple exposure (which actually ended up working out nicely: in the end I had to open 1/2 a stop from the standard reading for each exposure) and the result was a roll with wildly varying exposure. I feel like I'm usually really good at proper exposures, so that kind of stings.
-Lastly, and this I didn't anticipate: 250D is a relatively low-grain stock, especially Vision 3. But my film turned out exceptionally grainy. I assume it's because every time I exposed, I was underexposing, and the slight grain increase this causes was multiplied by three. You can't really see it in the YouTube upload (sorry) but it's there.
In summary: I am planning on doing this again, but with a really, really sturdy tripod. I'm also planning on trying the same experiment with hand cranking (on a really, really sturdy tripod). The first exposure would be cranked forward, the second back, and the third forward again. Hypothetically, this would lead to really interesting light wavering - not only would the exposure waver with the speed of the crank, it would waver different colors depending on the speeds of the combined cranks - and if I were shooting cars, for example, one color of cars would be driving backwards. The frameline would get fucked up, too. How exciting!