Maybe someone can enlighten me...

A sensor that is specified as being a 1/3 inch sensor is obviously 1/3 of an inch in size, normally measured on the diagonal of the rectangle.

So far, so good.

I've recently started seeing sensors described as being 1/2.7.

What the hell is that? Mixing a vulgar and a decimal fraction makes no sense! (To me, at least.)

Image sensor format

Quote:Sizes are often expressed as a fraction of an inch, with a one in the numerator, and a decimal number in the denominator. For example, 1/2.5 converts to 2/5 as a simple fraction, or 0.4 as a decimal number. This "inch" system brings a result approximately 1.5 times the length of the diagonal of the sensor. This "optical format" measure goes back to the way image sizes of video cameras used until the late 1980s were expressed, referring to the outside diameter of the glass envelope of the video camera tube. David Pogue of The New York Times states that "the actual sensor size is much smaller than what the camera companies publish – about one-third smaller." For example, a camera advertising a 1/2.7" sensor does not have a sensor with a diagonal of 0.37"; instead, the diagonal is closer to 0.26".[29][30][31] Instead of "formats", these sensor sizes are often called types, as in "1/2-inch-type CCD."

Due to inch-based sensor formats being not standardized, their exact dimensions may vary, but those listed are typical.[30] The listed sensor areas span more than a factor of 1000 and are proportional to the maximum possible collection of light and image resolution(same lens speed, i.e., minimum F-number), but in practice are not directly proportional to image noise or resolution due to other limitations. See comparisons.[32][33] Film format sizes are included for comparison. The following comparison is with respect to the aspect ratio of 4:3. The application examples of phone or camera may not show the exact sensor sizes.

The fact that there's a Wikipedia entry justifying this nonsense doesn't make it any less nonsense. In fact, I'm having trouble actually comprehending what 'This "inch" system brings a result approximately 1.5 times the length of the diagonal of the sensor.' even means.

What would be really nice is if manufacturers just gave the physical X and Y of their sensors, along with the pixel count in each axis. That tells you everything you need to compare two sensors without them using disingenuous ways of saying how large the sensor is.

Or did I miss something?

That would be to easy and they wouldn't have anyway to try and "trick" you thinking one is better than the other.

I'm pretty sure my old maths teachers would have me in detention if I used a decimal with a fraction, regardless how many paragraphs of nonsense I used to try and justify it!

I am also 1/2.7th (is *th right for this?) confused about the reason for this... I like your solution better unseen... just tell me the dimensions and the pixels period!