The exposure math of the Holga has always been a mystery to me, because it seems to violate the reliable Sunny 16 Rule. The rule basically states that if you set your shutter speed to the ISO of your film and the aperture to f/16, you’ll get a solid exposure in direct sunlight. (For example, if you had ISO 50 film, you’d set your shutter speed to the nearest number, likely 1/60.)
Admittedly, every Holga is a bit different, but I did have one of my Holgas tested with a shutter speed tester, and it came in at a fairly reliable 1/100 sec, which is what it’s supposed to be. As I found out while I measured for The facts about Holga apertures, a stock Holga has an aperture of f/13.3, regardless of the position of the sunny/cloudy switch. So with aperture of roughly f/13 and a shutter speed of 1/100, you should be able to shoot ISO 100 film in bright sunlight and have a proper exposure. (Technically about a half stop over as we’re at f/13.3 instead of 16, but that’s well within the exposure latitude of modern print films, so you should just get a little extra shadow detail).
So why the hell do I have to shoot ISO 200 or 400 film in bright sunlight to get a decent exposure, while I get underexposed mud with ISO 100, which should work perfectly? After thinking about it for a while, I strongly suspected that the plastic "Optical Lens" must have a good deal of transmission loss and was blocking two stops of light, so I set up a test: I pointed a Canon 580EX flash fitted with a Sto-Fen Omni-Bounce diffuser straight into a Holga with a Sekonic L-558R meter (which is extremely accurate and consistent) stuck in the back. I metered the flash through the camera five times with and five times without the lens in place and was fucking shocked by what I discovered: there was effectively no transmission loss. The readings averaged out to differ by only about 1/6 of a stop.
WTF??!? It’s not the lens. It’s not the aperture. It’s probably not the shutter: while I can allow for the possibility that the speed changes over time—it was over a year ago I had it tested—I consistently have the same exposure requirements with all of the 8-ish Holgas I have. What’s left? Light-eating gnomes?
Gnomes….
[Badly wants Nicolai to do this same test with reflected light but cannot quite explain why.]
… or black holes, little ones.
I don’t expect anything will change, but I’ll give it a shot…
damn, I was hoping you had cracked this nut.
Dark matter. It’s out there – and Holgas are dark matter magnets. Don’t ask why. They just are.
falloff? Given the vignetting, I bet that it is perceptually better to have the center stop over, a ring of normal in the middle, and the outside edges under.
wirehead: That’s a good point, but I think it depends on what you’re after. Sometimes I like it like you described, and sometimes I want the center dead-on and the edges fading out to nothing.
Two things come to mind:
1) Falloff as wirehead mentions, maybe it’s a perceptual illusion caused by the darker surrounds?
2) Is the lens really 60mm, and would this affect your f-stop calculation? Maybe its not really f13?
Interesting, even though Holga exposure characteristics are still a bit murky. I didn’t know what the aperture is supposed to be, or the shutter speed, but I sure had noticed the problem with 100 ISO film.
What kind of range of wavelengths does the typical camera flash emit? One random thought that crossed through my mind is that the Holga lens does not have transmission loss at the wavelengths emitted by the flash, but does at wavelengths outside that range.
The only other thing I can think of is simply that you’ve got some as-yet-undetermined source of experimental error (e.g.., the shutter speed has changed markedly in the last year).
As far as I know, it goes from IR to slightly UV, which is the entire visible spectrum and then some. I’m looking for a spectral graph, though…
Nicolai, Did you even use the shutter in this test? I thought no.
The shutter moving at 1/200, 1/400, or faster would explain the light loss, and thus the breakdown of “sunny 16”. Does anyone think that kind of speed is possible?
If the lens focused the transmitted light onto the 558’s sensor, and the missing lens allowed light to bypass the sensor, this could mask a transmission loss.
science: That’s a good question. I don’t know how to figure out the nodal point of a meniscus lens for testing. Anyone? Speaking of which, does anyone know of a good optical primer that’s actually in print?
Captoe: I held the shutter open in bulb mode, so it effectively wasn’t in the experiment.
I’m not sure I follow your last paragraph, but if I understand it correctly, that wasn’t a factor either. The meter was stuck into the back of the Holga, with the body blocking the light from the flash. So the light path was flash -> diffusor -> Holga aperture (w/ and w/o lens) -> meter.
You addressed my last paragraph with:
“The meter was stuck into the back of the Holga”
I was thinking about the falloff effect applied the light meter. Focused light versus the same light spread evenly.
I wrote bypass the sensor because I didn’t know where you were holding the meter. Left to my own devices I might have put the meter in the film plane, which at 6×6 is larger than the sensor on the 558.
Hrm, yeah. I did hold the meter’s sensor in the middle, but farther into the camera than the film plane (I wanted to eliminate as much ambient light as possible). I don’t know enough about optics to really say one way or the other.
It’s certainly far from a perfect test; I’d ultimately like to take a lens to an optician and have them do an actual transmission loss measurement on an appropriate machine, because I’d really like to know what the hell is going on!
“So why the hell do I have to shoot ISO 200 or 400 film in bright sunlight to get a decent exposure, while I get underexposed mud with ISO 100, which should work perfectly?”
I shot all my ‘holga at the beach’ photos with 160iso negative film. Almost all prints came out nicely exposed.
At least for me…
Wow… I thought I was alone to mix science and Holga… But you guys are amazing…
Here’s my modest participation in the field of Holga sciences :
http://www.lomohomes.com/weblog/stouf
Stouf
Wow, nice work, stouf!!
I shoot 100 ISO film in sunlight, even when it’s slightly cloudy/overcast, and my photos come out fine, sometimes even overexposed in areas.
I’ve heard about people having to use higher ISO films, and I know some instructions for other cameras recommend up to 800 ISO on a cloudy day, which totally baffles me.
I love how people over engineer their holgas. I built an optical shutter speed tester that measured to the nearest millisecond. I found the shutter was particularly consistant. Bizarre.