Re:DEC64: Decimal Floating Point
by sharpneli on Sunday March 09, 2014 @12:29PM
(#7368828)
Except that actually does apply to X86.
As an example with AVX instructions you can process 8 floats at the same time, compared to 4 doubles. So if float is enough for you you can expect double performance in either memory transfer bound or ideally vectorizable algorithms.
And in mobile computer graphics 16bit values are common.
As an example with AVX instructions you can process 8 floats at the same time, compared to 4 doubles. So if float is enough for you you can expect double performance in either memory transfer bound or ideally vectorizable algorithms.
And in mobile computer graphics 16bit values are common.