Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 19 Mar 1998 18:39:33 -0800 (PST)
From:      Alex <garbanzo@hooked.net>
To:        "Daniel O'Connor" <doconnor@gsoft.com.au>
Cc:        current@FreeBSD.ORG
Subject:   Re: More wank^H^H^H^Hspeed tests
Message-ID:  <Pine.BSF.3.96.980319183458.323A-100000@zippy.dyn.ml.org>
In-Reply-To: <199803200100.LAA24874@cain.gsoft.com.au>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 20 Mar 1998, Daniel O'Connor wrote:

> This is my 6x86/P166+ with an <Intel 82437VX PCI cache memory controller> rev
> 0x02 on pci0.0.0.
> 
> stream_d was compiled with gcc -o stream_d stream_d.c second_cpu.c
> -malign-double -O3 -lm
[...]
> Your clock granularity/precision appears to be 7812 microseconds.
> Each test below will take on the order of 164062 microseconds.
>    (= 21 clock ticks)
> Increase the size of the arrays if this shows that
> you are not getting at least 20 clock ticks per test.
[...]
> Function      Rate (MB/s)   RMS time     Min time     Max time
> Copy:          93.0909       0.1774       0.1719       0.1797
> Scale:         89.0435       0.1852       0.1797       0.1875
> Add:           96.0000       0.2595       0.2500       0.2656
> Triad:         93.0909       0.2618       0.2578       0.2734
> 
> IMHO this isn't too bad for a M/B which was one of the cheapest I could find, 
> I think it cost me around AU$180 or so, and I bought it a while ago.

-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 1000000, Offset = 0
Total memory required = 22.9 MB.
Each test is run 10 times, but only
the *best* time for each is used.
-------------------------------------------------------------
Your clock granularity/precision appears to be 7812 microseconds.
Each test below will take on the order of 117187 microseconds.
   (= 15 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
[...]
-------------------------------------------------------------
Function      Rate (MB/s)   RMS time     Min time     Max time
Copy:          93.0909       0.1838       0.1719       0.1953
Scale:        113.7778       0.1469       0.1406       0.1484
Add:          122.8800       0.2024       0.1953       0.2031
Triad:        118.1538       0.2031       0.2031       0.2031

This was with my p180 (60mhz*3) after some tweaking with the memory timing
settings using 60ns EDO ram.  No cache controller shows up upon bootup,
and I used an oldish pgcc w/ -O9 (cc -O3 were about 20MB/s slower).  I'm
not even sure who makes my motherboard, tis very generic with even
cheaper RAM used in the second bank.

- alex

A person who has both feet planted firmly in the air can be safely called
a liberal.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.980319183458.323A-100000>