sortbench.lha utility/shell - Comments

Spectre660 (76.76.182.224) @ 13 Nov 2016, 03:29.03

Sun, 13 Nov 2016 15:29:03 +0100

By: Spectre660 (76.76.182.224)

Sam460ex closed all running clock apps before running . 
 
------------------------------------------------------------- 
SORTBENCH 1.1 (Gunnar von Boehn) 
Its a CPU benchmark that stresses CPU, DCache and branch prediction. 
------------------------------------------------------------- 
1 K Element :  1646.91 MB/sec 
2 K Element :  1712.88 MB/sec 
4 K Element :  1694.81 MB/sec 
8 K Element :  1679.20 MB/sec 
16 K Element :  512.01 MB/sec 
32 K Element :  426.25 MB/sec

marko (83.248.160.12) @ 13 Nov 2016, 03:23.33

Sun, 13 Nov 2016 03:23:33 +0100

By: marko (83.248.160.12)

Sam440ep-flex 800 MHz 
 
With plain Workbench (without WBStartup progs): 
 
------------------------------------------------------------- 
SORTBENCH 1.1 (Gunnar von Boehn) 
Its a CPU benchmark that stresses CPU, DCache and branch prediction. 
------------------------------------------------------------- 
1 K Element :  1199.11 MB/sec 
2 K Element :  1209.99 MB/sec 
4 K Element :  1210.13 MB/sec 
8 K Element :  1199.02 MB/sec 
16 K Element :  382.94 MB/sec 
32 K Element :  303.50 MB/sec

marko (83.248.160.12) @ 13 Nov 2016, 03:09.46

Sun, 13 Nov 2016 03:09:46 +0100

By: marko (83.248.160.12)

On Sam440ep-flex 800MHz 
 
------------------------------------------------------------- 
SORTBENCH 1.1 (Gunnar von Boehn) 
Its a CPU benchmark that stresses CPU, DCache and branch prediction. 
------------------------------------------------------------- 
1 K Element :  1066.03 MB/sec 
2 K Element :  1085.57 MB/sec 
4 K Element :  1078.07 MB/sec 
8 K Element :  1053.71 MB/sec 
16 K Element :  344.92 MB/sec 
32 K Element :  272.60 MB/sec

zzd10h (109.9.30.119) @ 12 Nov 2016, 06:21.13

Sat, 12 Nov 2016 18:21:13 +0100

By: zzd10h (109.9.30.119)

previous comment was from me, zzd10h, 
I wanted to reply to ddni and I wrongly write "ddni" as username...

ddni (109.9.30.119) @ 12 Nov 2016, 06:17.55

Sat, 12 Nov 2016 18:17:55 +0100

By: ddni (109.9.30.119)

Ah yes, you are right, it's better :) (tested twice) 
 
17/0.RAM Disk:sortbench> gcc -o sort -O2 sortbench.c 
 
17/0.RAM Disk:sortbench> sort 
------------------------------------------------------------- 
SORTBENCH 1.1 (Gunnar von Boehn) 
Its a CPU benchmark that stresses CPU, DCache and branch prediction. 
------------------------------------------------------------- 
1 K Element :  2116.08 MB/sec 
2 K Element :  2122.17 MB/sec 
4 K Element :  2123.24 MB/sec 
8 K Element :  2144.95 MB/sec 
16 K Element :  2141.67 MB/sec 
32 K Element :  1657.02 MB/sec

TSK (77.86.211.251) @ 12 Nov 2016, 03:52.04

Sat, 12 Nov 2016 15:52:04 +0100

By: TSK (77.86.211.251)

Compiling the utility with -O3 gave slower speed than with O1/O2. Compiling it with O0 (o zero) made the speed to drop to snails speed. These kind of tests really doesn't test a CPU as long as you can't disable multitasking completely and don't write it using assembly to avoid any compiler related tricks.

ddni (86.151.32.102) @ 12 Nov 2016, 03:32.27

Sat, 12 Nov 2016 15:32:27 +0100

By: ddni (86.151.32.102)

What do you get if you build it yourself?

zzd10h (109.9.30.119) @ 12 Nov 2016, 02:34.45

Sat, 12 Nov 2016 14:34:45 +0100

By: zzd10h (109.9.30.119)

jabirulo (2.139.107.119) @ 12 Nov 2016, 12:04.46

Sat, 12 Nov 2016 12:04:46 +0100

By: jabirulo (2.139.107.119)

SAM460ex: 
SORTBENCH 1.1 (Gunnar von Boehn) 
Its a CPU benchmark that stresses CPU, DCache and branch prediction. 
------------------------------------------------------------- 
1 K Element :  1660.98 MB/sec 
2 K Element :  1662.14 MB/sec 
4 K Element :  1661.23 MB/sec 
8 K Element :  1629.98 MB/sec 
16 K Element :  474.56 MB/sec 
32 K Element :  390.30 MB/sec

rwo (95.154.21.8) @ 12 Nov 2016, 08:29.35

Sat, 12 Nov 2016 08:29:35 +0100

By: rwo (95.154.21.8)

My evil Sam 666mhz gives me 
 
1 K Element : 1006.66 MB/sec 
2 K Element : 1008.71 MB/sec 
4 K Element : 1009.42 MB/sec 
8 K Element : 1007.09 MB/sec 
16 K Element :  346.88 MB/sec 
32 K Element :  267.28 MB/sec 
 
Run from shell with no startup sequence

rwo (95.154.21.8) @ 12 Nov 2016, 08:16.13

Sat, 12 Nov 2016 08:16:13 +0100

By: rwo (95.154.21.8)

Darn... look like I included the wrong binary in the first archive. 
 
This new version is compiled with the right optimation -O2.. it should give you a nice increase.

daveyww (118.92.8.78) @ 11 Nov 2016, 11:45.53

Fri, 11 Nov 2016 23:45:53 +0100

By: daveyww (118.92.8.78)

Hmm, the SAM460 is pretty disappointing by comparison: 
 
1 K Element :  1226.22 MB/sec 
2 K Element :  1239.28 MB/sec 
4 K Element :  1233.07 MB/sec 
8 K Element :  1187.15 MB/sec 
16 K Element :  445.17 MB/sec 
32 K Element :  379.94 MB/sec

Lio (213.176.206.197) @ 11 Nov 2016, 10:18.57

Fri, 11 Nov 2016 22:18:57 +0100

By: Lio (213.176.206.197)

on my A1XEG4 I get : 
1 K Element :  1916.82 MB/sec 
2 K Element :  1916.53 MB/sec 
4 K Element :  1917.88 MB/sec 
8 K Element :  1917.02 MB/sec 
16 K Element :  1846.13 MB/sec 
32 K Element :  1831.82 MB/sec

ddni (86.151.32.102) @ 11 Nov 2016, 01:48.37

Fri, 11 Nov 2016 13:48:37 +0100

By: ddni (86.151.32.102)

PASEMI PA6T-1682M @1800mhz 
Single core 
AMIGAONE X1000 
 
------------------------------------------------------------- 
SORTBENCH 1.1 (Gunnar von Boehn) 
Its a CPU benchmark that stresses CPU, DCache and branch prediction. 
------------------------------------------------------------- 
1 K Element :  2084.88 MB/sec 
2 K Element :  2096.05 MB/sec 
4 K Element :  2105.75 MB/sec 
8 K Element :  2134.63 MB/sec 
16 K Element :  2112.34 MB/sec 
32 K Element :  1712.92 MB/sec

ddni (86.151.32.102) @ 11 Nov 2016, 08:59.40

Fri, 11 Nov 2016 08:59:40 +0100

By: ddni (86.151.32.102)

Thanks. 
I wonder what scores the new Vampire 600 or 500 get?