Mail Archives: djgpp/1999/07/27/08:26:36
Martin Str|mberg <ams AT ludd DOT luth DOT se> wrote:
> Klaas (klaas AT ns DOT sympatico DOT ca) wrote:
> : Rob Kramer wrote:
> : > One of my applications (involving a GIF-decoder) I recently tried on a
> : > 386 SX machine. The decoder is terribly slow in that case. The original
> : > pre-DJGPP obsolete 16-bit version of the application is way faster. (I'm
> : > not sure whether the 32-bit version is faster than the 16-bit version on
> : > a 386 DX)
> [Klippa, klapp, kluppit.]
> : Do you use any floating point math? The SX would have to emulate, which
> : wuold be considerably slower.
> Eeerh... Please correct me if I'm wrong, but the 386(whatever) didn't
> have a FPU at all, no matter what XX. I'm looking at a fine sample of
> 386DX-25 right now, and it _does_ _not_ have a FPU.
Correct.
> It is the 486DX that has a FPU while 486SX don't (or it even have one
> but disabled - so the result is the same as not having one).
Right.
If memory serves, the difference between the 386SX and SX is the bus
width: the SX only has a 16-bit bus ("32-bit? What for? Who needs such
computing power?"), thus reducing the cost of the system by some
factor (probably a large factor is I remember the prices at that time).
Thus running 32-bit code (or more exactly: using 32-bit data) takes a
huge performance hit when performed on a 386SX.
--
Alain Borel
<Alain DOT Borel AT icma DOT unil DOT ch>
- Raw text -