30 Sep
2009
30 Sep
'09
6:17 p.m.
On 30 Sep 2009 at 20:05, Michael B. Brutman wrote:
On a related note, does anybody have a good method for
doing a
floating point multiply in software? Before I steal an existing
floating point emulation lib, I'd like to see if I can cheat and just
write enough code to do the one multiply that I need.
I need to figure out how to do this:
float elapsed = ticks * 0.85;
What range of arguments and and to what precision? You can always
simplify division by a constant to multiplication by the binary-
scaled reciprocal (with some loss of precision). Multiplication, of
course, can be decomposed to a series of adds and shifts.
--Chuck