[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
Re: Mixed radix arithmetic
Conversion between binary and decimal floating-point formats is the same
problem mathematically as converting to and from decimal strings for i/o.
Correct rounding is a well-understood problem in software.
Converting NaNs is something else. Since the standard can't say anything
about what NaNs mean, it is up to the implementation to decide how to
convert them. 754 doesn't even say how to convert NaNs between floating-point
formats, though I hope 754R will at least require that single->double->single
won't change a quiet NaN.