stream results for $199 computer seem rather high...

From: Roldan Pozo <roldanpozo@hotmail.com>
Date: Tue Aug 08 2006 - 20:40:52 CST

John,

I have just put together one of the cheapest 64-bit Linux boxes: a $199
Emachines T3506 (3.2 GHz Intel Celeron D, 512BK cache, 512 MB RAM running
64-bit Fedora Core 5) and it seems to be getting rather high stream numbers.
  Could this be a new Mflops/$ high score? (I used gcc v. 4.1.1, -O6
-funroll-all-loops).

Regards,
Roldan Pozo
http://math.nist.gov/pozo

------------------------------------------------------------
STREAM version $Revision: 5.6 $
-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 2000000, Offset = 0
Total memory required = 45.8 MB.
Each test is run 10 times, but only
the *best* time for each is used.
-------------------------------------------------------------
Printing one line per active thread....
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 51019 microseconds.
   (= 51019 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) Avg time Min time Max time
Copy: 996.0721 0.0330 0.0321 0.0344
Scale: 1025.1106 0.0318 0.0312 0.0332
Add: 1167.4626 0.0420 0.0411 0.0430
Triad: 1170.3061 0.0419 0.0410 0.0429
-------------------------------------------------------------
Solution Validates
-------------------------------------------------------------
Received on Wed Aug 09 08:58:11 2006

This archive was generated by hypermail 2.1.8 : Sun Sep 17 2006 - 09:41:36 CST