Does anyone have a K7 RH6.0 system out there who would be willing to
run a simple benchmark? It's just a simple matrix-multiply type code
(source code below).
If you could compile the code and run it and let me know the time, I'd
appreciate it. You may also find the comparison with other machines
informative; Linux on the cheap Abit BP6 with two OC Celerons looks
really good even compared to Sun machines costing more than 10 times
as much!
Note: use the egcs gcc compiler if you have it - it makes quite a
difference!
Run it as "time bench" and you should get a summary line like:
8.960u 0.050s 0:09.36 96.2% 0+0k 0+0io 91pf+0w
Thanks!
Jim
[EMAIL PROTECTED]
/*
%s cc -O -native -fast -o bench bench.c
% gcc -O4 -o bench bench.c
*/
#include <time.h>
#include <stdio.h>
#define Dim 512
/**
benchmark results:
On Linux, gcc -O4
On Solaris: cc -O -native -fast
brokk, dual Celeron 400 (556/92)MHz egcs-2.91.66
Linux rask.gso.saic.com 2.2.5-15smp #1 SMP Mon Apr 19 22:43:28 EDT 1999 i686 unknown
4.77 (Unstable)
------------
brokk, dual Celeron 400 (540/90)MHz egcs-2.91.66
Linux rask.gso.saic.com 2.2.5-15smp #1 SMP Mon Apr 19 22:43:28 EDT 1999 i686 unknown
TBD (Stable)
------------
tyr, Pentium II/400 with Linux 6.0
Linux tyr.gso.saic.com 2.2.5-15 #1 Mon Apr 19 23:00:46 EDT 1999 i686 unknown
8.960u 0.050s 0:09.36 96.2% 0+0k 0+0io 91pf+0w
------------
bor, UltraSparc 60/450
SunOS bor 5.7 Generic_106541-06 sun4u sparc SUNW,Ultra-60
10.0u 0.0s 0:10 94% 0+0k 0+0io 0pf+0w
------------
rask, dual Pentium II/300 on Linux 6.0
Linux rask.gso.saic.com 2.2.5-15smp #1 SMP Mon Apr 19 22:43:28 EDT 1999 i686 unk
nown
10.680u 0.040s 0:10.73 99.9% 0+0k 0+0io 96pf+0w
------------
vali, UltraSparc 60/333
SunOS vali 5.6 Generic_105181-11 sun4u sparc SUNW,Ultra-60
12.0u 0.0s 0:13 92% 0+0k 0+0io 0pf+0w
------------
bragi SPARC ?? running Linux 6.0
Linux bragi.gso.saic.com 2.2.5-15 #1 Mon Apr 19 22:22:52 EDT 1999 sparc64 unknow
16.260u 0.060s 0:16.31 100.0% 0+0k 0+0io 73pf+0w
------------
ull, Sun Enterprise 3500:
SunOS ull 5.6 Generic_105181-03 sun4u sparc SUNW,Ultra-Enterprise
18.72u 0.06s 0:19.14 98.1%
------------
edda, Ultra 5
SunOS edda 5.6 Generic_105181-05 sun4u sparc SUNW,Ultra-5_10
21.0u 0.0s 0:22 94% 0+0k 0+0io 0pf+0w
------------
rind, UltraSparc 2
SunOS rind 5.5.1 Generic_103640-08 sun4u sparc SUNW,Ultra-2
23.0u 0.0s 0:23 96% 0+0k 0+0io 0pf+0w
------------
hymir, Sun Enterprise 3500:
SunOS hymir 5.7 Generic_106541-04 sun4u sparc SUNW,Ultra-Enterprise
24.0u 0.0s 0:34 70% 0+0k 0+0io 0pf+0w
------------
suttung, UltraSparc 1
SunOS suttung 5.7 Generic_106541-04 sun4u sparc SUNW,Ultra-1
25.0u 0.0s 0:25 98% 0+0k 0+0io 0pf+0w
------------
sept.jw.org, AMD K6/200 running Linux 6.0
Linux sept.jw.org 2.2.5-15 #1 Mon Apr 19 22:21:09 EDT 1999 i586 unknown
36.400u 0.130s 0:36.53 100.0% 0+0k 0+0io 90pf+0w
------------
amd K6-2 350 (373/83) Linux5.2 egcs-2.90.29
Linux rh1 2.0.36 #1 Tue Oct 13 22:17:11 EDT 1998 i586 unknown
39.44 sec
------------
P-133 (133/66) Linux6.0 egcs-2.91.66
Linux webcam 2.2.5-15 #1 Mon Apr 19 22:21:09 EDT 1999 i586 unknown
47.57 sec
------------
kon, Sparc 20
SunOS kon 5.5.1 Generic_103640-08 sun4m sparc SUNW,SPARCstation-20
49.0u 0.0s 1:30 53% 0+0k 0+0io 0pf+0w
------------
mothir, Sparc 5
SunOS mothir 5.7 Generic_106541-06 sun4m sparc SUNW,SPARCstation-5
54.0u 0.0s 0:55 98% 0+0k 0+0io 0pf+0w
------------
oct.jw.org, Pentium 133 running Linux 6.0
Linux oct.jw.org 2.2.5-15 #1 Mon Apr 19 22:21:09 EDT 1999 i586 unknown
61.980u 2.160s 1:04.32 99.7% 0+0k 0+0io 90pf+0
------------
P-233 (233/66) [laptop] Linux6.0 egcs-2.91.66
Linux ringhorn 2.2.5-15 #1 Mon Apr 19 22:21:09 EDT 1999 i586 unknown
67.550000
------------
*/
main()
{
/* const int Dim = 1024;*/
double A[Dim+1][Dim+1], B[Dim+1][Dim+1], C[Dim+1][Dim+1];
int i, j, k;
double s;
clock_t start, stop, diff;
printf("\nRemember to update the comments section in bench.c with your
results\n\n");
for ( i= 0; i < Dim; i++ )
for ( j= 0; j < Dim; j++ ) { A[i][j]= i+j; B[i][j]= i+j-1; }
start= clock();
for ( i= 0; i < Dim; i++ )
for ( j= 0; j < Dim; j++ ) {
for ( s= 0, k= 0; k < Dim; k++ )
s+= A[i][k]*B[k][j];
C[i][j]= s;
}
stop= clock();
diff= stop - start;
printf ("Start: %0d ... Stop %0d\n Time used: %f\n",
start, stop, 1e-6*diff);
#ifdef NO
cout << "Start: " << start << " ... Stop: " << stop
<< " Time used: " << diff*1E-6 << endl;
#endif
return 0;
}
--
To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
as the Subject.