Speed of floating point double vs single precision

[Beamguy]

I am writing a numerical program and am pondering which precision to use for floating
point numbers. A bunch of years back there was a class of computers where it became
faster to do floating point arithmetic in double precision, since the hardware floating
point was double precision. I am not sure about PC's. Are hardware floating point
units on PC's double or single precision these days, and based on this which is faster
to use?

Thanks.

 

[Carey Frisch [MVP]]

IEEE Standard 754 floating point numbers
http://stevehollasch.com/cgindex/coding/ieeefloat.html

--
Carey Frisch
Microsoft MVP
Windows XP - Shell/User

Be Smart! Protect Your PC!
http://www.microsoft.com/athome/security/protect/default.aspx

-----------------------------------------------------------------------------

:

| I am writing a numerical program and am pondering which precision to use for floating
| point numbers. A bunch of years back there was a class of computers where it became
| faster to do floating point arithmetic in double precision, since the hardware floating
| point was double precision. I am not sure about PC's. Are hardware floating point
| units on PC's double or single precision these days, and based on this which is faster
| to use?
|
| Thanks.