Why is long double only 8 bytes?

Question

when I do size of long double i get 16...

Answer 1

The size of a long double type is platform and/or implementation dependent, but it's guaranteed the size to be at least as the size of a double type. In this case, it turned out that it's a quadruple precision floating point (FP 128bit) which is awesome for some use cases.

[https://en.m.wikipedia.org/wiki/Quadruple-precision_floating-point_format]

And, the maximum possible value which can be stored in a variable of that type is 1.18973e+4932.