how much data are you copying? is it being
copied once or multiple times?
if it is just being copied once, you're probably seeing
a lot of memory-cache misses. in real situations, the
buffers tend to be small and reused to minimize latency,
which has the right cache behavior.
[am i correct in assuming the double-double/float-float copies
were normalized to the same number of bytes?]
- dave
On Tue, 4 Apr 2000, Benno Senoner wrote:
> Hi folks,after the float vs double debate, I decided to benchmark at least
> conversion between the two datatypes.
>
> The code is attached to the end of the mail
> ( compile with: gcc -O2 -o floatspeed floatspeed.c -lm )
>
> basically I am testing: (loops of the instructions below unrolled
> in sequences of 4 to minimize the overhead of the for loop)
