STREAM results for Apple G4/400 system

From: Anton Rang (
Date: Wed Mar 22 2000 - 14:02:55 CST


I just got a G4 at work and noticed that there weren't any STREAM results
for it posted yet. Here's the results for the "standard" PPC604-optimized
executable (which does have some cache control instructions in it). If I
find time over the next week or so, I'll put together a hand-optimized
version using the new Altivec instructions for cache line streaming and run
tests with that, too.

I didn't rebuild this binary, so the compiler flags etc. are the same as
whenever it was last built....


-- Anton

This system uses 8 bytes per DOUBLE PRECISION word.
Array size = 400000, Offset = 0
Total memory required = 9.2 MB.
Each test is run 10 times, but only
the *best* time for each is used.
Your clock granularity/precision appears to be 10 microseconds.
Each test below will take on the order of 21532 microseconds.
   (= 2153 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
Function Rate (MB/s) RMS time Min time Max time
Copy: 383.7391 0.0174 0.0167 0.0217
Scale: 295.6940 0.0227 0.0216 0.0258
Add: 493.3450 0.0197 0.0195 0.0200
Triad: 347.4862 0.0278 0.0276 0.0281

This is an unauthorized communication.  "The statements and opinions
expressed herein are my own and do not necessarily represent those of

This archive was generated by hypermail 2b29 : Tue Apr 25 2000 - 01:49:24 CDT