There are a handful of professional YouTubers who incorrectly claim sensor size affects bokeh, specifically Tony & Chelsea Northup who, despite all of the facts proving otherwise, insist on claiming that smaller sensors have less bokeh, or increased field of view. Even worse are all of the morons, pros and amateurs, who simply regurgitate their bullshit.
First, let’s get something out of the way quickly: these two are really bad at facts, including maths. If you find their videos, especially their “History of (insert product)” series, insufferable, their inability to grasp basic concepts like addition is probably why.
Their confusion lies in the “crop factor.” Bokeh is based on aperture. F-stops are directly tied to your lens’ focal length, in millimeters. Using the 50mm focal length as an example, with a ƒ/1.4 max aperture, on a “35mm full frame sensor,” that same lens used on an APS-C sized sensor would create an image with an approximately 75mm field of view. Because of this crop factor, a lens with ~33mm focal length is required on an APS-C sized sensor to create an image that best replicates the field of view the 50mm creates on a full frame sensor.
But bokeh is a product of the aperture, whose area is roughly the focal length divided by the F-stop. Back to the example of using a ~33mm lens on APS-C to get the FOV of a 50mm in full frame, that means the aperture area is smaller per F-Stop whenever you’re adapting a lens’ focal length to match the field of view. This applies to any 2 dissimilar lenses you’re comparing.
Again, a larger aperture results in increased bokeh, so when comparing dissimilar focal lengths, the lens with the smaller focal length will always have a smaller aperture at the same F-stop.
It has nothing to do with the size of the image sensor and everything to do with the lens. It’s irrational to expect a 33mm ƒ/1.4 lens to render similar bokeh to a 50mm ƒ/1.4 lens when the former has obviously a smaller aperture along the whole F-stop scale.
Sure, you could argue that smaller sensors use smaller focal length lenses, so therefore image sensor size affects bokeh, but you’d be wrong. An equally false equivalency would be to say that physically larger and heavier cars require bigger engines thus bigger cars are faster than smaller cars. Essentially, this is what a person is trying to convince you of when saying “APS-C sensors have ‘less bokeh’,” whatever that means.
Now for the proof by simply doing the math. If Tony and Chelsea are correct, then an APS-C sensor can’t have the “same bokeh” as a full frame sensor. Well, 50mm divided by 1.4 equals 35.71mm. To get the same FOV and “the same bokeh” on an APS-C sensor, you’d need a lens with a ~33mm focal length and 35mm maximum aperture. In fact, a Chinese company called Zhong Yi, aka ZY Optics, aka Mitakon, makes the Speedmaster 35mm ƒ/0.95 for APS-C sensor based cameras. That lens has a 36.84mm max aperture, about 1mm larger than the 50mm ƒ/1.4 example. That aperture should render bokeh on an APS-C sensor that’s out of focus to a nearly identical degree as the 50mm example on a full frame sensor.
The sensor in no way affects bokeh and it’s time YouTubers like Tony and Chelsea stop spewing the same, incorrect info and for others to stop regurgitating it. It’s irrational and just plain wrong.