Dennis Wood
DVXuser Sponsor
I know this info is all over the web, however after a post in another thread, I thought I'd fire this up in the hope that it may "stick". Light loss and adapters are an area that is constantly confused, and in some cases, misrepresented. A basic understanding of f/stops is required, and perhaps a methodogy for testing. F stops just describe the ratio of front glass diameter to rear aperture diameter. If you were to measure the diameter of the front glass of a 35mm lens, and divide this by the rear aperture diameter, you will get the f value. Using a 50mm lens as an example:
A 50mm f1 lens would have rear lens diameter of 50mm and an aperture area equal to ~2000 sq. mm (50/50 = 1) Thereoretically, this lens would have 0 light loss. You won't find an f1 lens for sale anywhere (EDIT: actually Dan points out there are lots, they're not cheap. Also these measurements are for discussion purposes only as manufacturers will vary.)...but that's another discussion.
A 50mm f1.4 lens would have a rear aperture diameter of 35.7mm and an aperture area equal to ~1000 sq. mm (50/35.7 = 1.4)
Given that the aperture area of the f1.4 lens is 1/2 that of the f1 lens, it requires twice as much light relative to the f1 lens. That is one stop. Each of these represents 1 stop, or a reduction in light of 50% from it's previous value, therefore the rear aperture is halved with respect to area for each stop.
f1 - f1.4 - f2 - f2.8 - f4 - f5.6 - f8 - f11 - f16 - f22
To simply read off the numeric decrease indicated on the camera (with and wthout an adapter + lens) is really not correct. A decrease from f5.6 to f4 is one stop, not 1.6 stops. Of course when you attach the adapter and the exposure drops, we tend to think linearly. There are two considerations. We need to look at light lost by the 35mm lens, and also light lost (or gained) by the adapter itself. The sum of these we'll call the total adapter "system" loss. Here's a practical example based on my own observations with the Brevis.
Let's use an f1.4 lens with the aperture fully open (f1.4).
1. An f1.4 lens has a rear area 50% that of an f1 lens. By definition then, we lose 1 stop of light with this lens at max aperture. If we were using an f2 lens, we lose two stops, and would require four times the light over an f1 lens!
2. If you frame an identical scene with and without the adapter affixed (video camera stays at a fixed distance), the camara can be used as a quick tool to evaluate loss. In my testing a typically drop is from F4 to F3.4. Most cameras have unreported f/stops in between the standards, so you need to be carefull the camera is not reporting one of the intermediates. f3.4 is halfway between f2.8 and f4.0 and therefore represents .5 stops total system loss. There are many techniques in the digital realm to measure, but most people will use their cameras.
3. Adapter loss = Total system loss - lens loss.
Adapter loss = .5 (system loss) - 1 (lost by f1.4 lens)
Adapter loss= -.5 f/stops (with most adapters this will not be a negative number)
In this case, the adapter loses negative .5 stops, (translated, this is actually a gain of .5 stops) then loses 1 stop from the lens to give a net loss of .5 f/stops. This is counter-intuitive, however the numbers are what they are. Typically, you'll see .5 to .1 f stops adapter loss for a total of f 1.5 to 2 stops (assuming f1.4 lens used) with most adapters. If you are lighing a scene, a system loss of two stops means you must quadruple the light levels if you wish to retain the same exposure. Hopefully this helps clear up some confusion.
A 50mm f1 lens would have rear lens diameter of 50mm and an aperture area equal to ~2000 sq. mm (50/50 = 1) Thereoretically, this lens would have 0 light loss. You won't find an f1 lens for sale anywhere (EDIT: actually Dan points out there are lots, they're not cheap. Also these measurements are for discussion purposes only as manufacturers will vary.)...but that's another discussion.
A 50mm f1.4 lens would have a rear aperture diameter of 35.7mm and an aperture area equal to ~1000 sq. mm (50/35.7 = 1.4)
Given that the aperture area of the f1.4 lens is 1/2 that of the f1 lens, it requires twice as much light relative to the f1 lens. That is one stop. Each of these represents 1 stop, or a reduction in light of 50% from it's previous value, therefore the rear aperture is halved with respect to area for each stop.
f1 - f1.4 - f2 - f2.8 - f4 - f5.6 - f8 - f11 - f16 - f22
To simply read off the numeric decrease indicated on the camera (with and wthout an adapter + lens) is really not correct. A decrease from f5.6 to f4 is one stop, not 1.6 stops. Of course when you attach the adapter and the exposure drops, we tend to think linearly. There are two considerations. We need to look at light lost by the 35mm lens, and also light lost (or gained) by the adapter itself. The sum of these we'll call the total adapter "system" loss. Here's a practical example based on my own observations with the Brevis.
Let's use an f1.4 lens with the aperture fully open (f1.4).
1. An f1.4 lens has a rear area 50% that of an f1 lens. By definition then, we lose 1 stop of light with this lens at max aperture. If we were using an f2 lens, we lose two stops, and would require four times the light over an f1 lens!
2. If you frame an identical scene with and without the adapter affixed (video camera stays at a fixed distance), the camara can be used as a quick tool to evaluate loss. In my testing a typically drop is from F4 to F3.4. Most cameras have unreported f/stops in between the standards, so you need to be carefull the camera is not reporting one of the intermediates. f3.4 is halfway between f2.8 and f4.0 and therefore represents .5 stops total system loss. There are many techniques in the digital realm to measure, but most people will use their cameras.
3. Adapter loss = Total system loss - lens loss.
Adapter loss = .5 (system loss) - 1 (lost by f1.4 lens)
Adapter loss= -.5 f/stops (with most adapters this will not be a negative number)
In this case, the adapter loses negative .5 stops, (translated, this is actually a gain of .5 stops) then loses 1 stop from the lens to give a net loss of .5 f/stops. This is counter-intuitive, however the numbers are what they are. Typically, you'll see .5 to .1 f stops adapter loss for a total of f 1.5 to 2 stops (assuming f1.4 lens used) with most adapters. If you are lighing a scene, a system loss of two stops means you must quadruple the light levels if you wish to retain the same exposure. Hopefully this helps clear up some confusion.
Last edited:
You've taken "a few" of these apart. Not only do the measurements vary quite a bit, but given coatings, elements etc, so do emissivity levels. Factor in different aperture shapes, (as the calcs are based on perfect circles) and you get all kinds of variables to throw "perfect" calcs. I'm curious to know if there's a standard that is used i.e. measureing light levels vs physical calcs.