The Internet is a very good at propagating hoaxes and transforming these hoaxes into so-called facts (let’s not go into tweets and “fake-news”). The APS-C and the equivalency factor is just one of these hoaxes floating around the Internet.
Some people claim that a 23mm f/2 lens on a APS-C camera is in fact a 34.5mm with an f/stop of 3.2.
So at great expense and a lot of time, I brought in the debunking unit!
The definition of an f/stop is:
The f-number of an optical system such as a camera lens is the ratio of the system’s focal length to the diameter of the entrance pupil.
Smith, Warren Modern Optical Engineering, 4th Ed. 2007 McGraw-Hill Professional
Nowhere does it mention the type or size of the camera.
For a Fujifilm XF 23mm f/2:
The focal length is: 23mm
The diameter of the entrance pupil is: 11.5mm
23/11.5 = 2 :: the f/stop is f/2
For a Canon 24mm f/2.8:
The focal length is: 24mm
The diameter of the entrance pupil is: 8.5mm
24/8.5 = 2.8 :: the f/stop is f/2.8
Nowhere, on the Canon box or on the Fujifilm box containing the lens, does it state that the aperture is dependent on the format of the camera. If this was true and since they do not put as an ‘*’, they would have been sued for attempting to deceive the customers…
The equivalency hoax relies on two myths.
The lens gathers light
The lens is not a vacuum cleaner, it doesn’t suck photons and then push these photons toward the sensor.
The lens transmits the light within its field of view from the outside toward the sensor. Notice the word toward and not “onto the sensor”.
The light hits the sensor
The claim is that since the sensor is smaller in an APS-C camera than in a full frame camera, a portion of the light is lost (where did the rest of the light go? It must be spilling somewhere outside…).
In fact, the light doesn’t hit the sensor. The CMOS sensor is made of millions of electronic sensors that record the light. On top of each of these electronic sensors, there is a photosite. The photosite is a small microlens that sits on top of the electronic sensor. And so far, it’s been a 1 for 1.
The light transmitted by the lens hits the photosite/microlens and activates the electronic sensor underneath.
Each photosite is illuminated by the same amount of light whether it’s a very small sensor or an APS-C or a full-frame or even bigger sensor like the medium formats.
Sekonic is a Japanese company that makes optical and electronic measurement instruments. It’s one of the very few companies that still makes light meters. The Seikonic light meters range from reasonable prices to very expensive.
I use the Sekonic L-308S which measures incident light. It’s one of those that is reasonably priced and I bought it on sale. It gives me a reading of the light “hitting” the subject. I use this reading to set the exposure on my cameras. It’s the same exposure for both my APS-C and for an old film Rolleiflex with Ilford FP4 (120 roll film: 12 exposures) which is 4.5 times larger than the APS-C camera. My light meter doesn’t care about which camera I use, and it doesn’t care about which lens I use. It only cares about the aperture of the lens (the amount of light transmitted), the ISO (the sensitivity of the film/sensor) of the film/sensor and the shutter speed of the camera (the amount of time for the exposure).
Taking a photo on both cameras, with the same exposure, results in the same brightness of the image.
If the equivalency was true, why wouldn’t Sekonic document it or even better, place a setting for it on their light meters?
Many, many film/sensor sizes have always existed, at least since 1888 (which is the start of film being used commercially for photography). From Kodak Disc/Minox, Kodak 110, Kodak 126, 35mm, 6x4.5 (cm), 6x6 (cm), 6x7 (cm), 6x9 (cm), 4x5 (inch), 5x7 (inch), 8x10 (inch), 11x14 (inch), Polaroid Instant 16x24 (inch), and I’m sure that there other sizes as well…
Many of the cameras, such as the Speed Graphic, would have different camera backs such as: 4x5", 3¼x4¼", 2.¼x3¼". Same lens, same shutter, same exposure…
The first light meter was a selenium light meter and it was created by Charles Fritts in 1884.
The first mention of the equivalency started in 2003 on a few forums.
So what happened between 1884 and 2003 with light metering? 120 years of wrong exposures?