You can combine two doubles to double precision, to 107 bits. https://en.wikipedia.org/wiki/Quadruple-precision_floating-point_format#Double-double_arithmetic

You can combine two doubles to double precision, to 107 bits. https://en.wikipedia.org/wiki/Quadruple-precision_floating-point_format#Double-double_arithmetic https://en.wikipedia.org/wiki/Quadruple-precision_floating-p...