hi all

 

i try unixbench (Version 5.1.3)

   compared with  orange pi PC's  System Benchmarks Index Score    202.0

 

 

 

Benchmark Run: 章 5月 03 2016 23:39:28 - 00:07:31
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        7827597.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                      915.0 MWIPS (9.8 s, 7 samples)
Execl Throughput                                459.3 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        176763.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           53800.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        431407.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              430477.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  73425.8 lps   (10.0 s, 7 samples)
Process Creation                               1360.4 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   1188.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    783.3 lpm   (60.1 s, 2 samples)
System Call Overhead                         969171.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         16700.0    7827597.5    670.7
Double-Precision Whetstone                            55.0           915.0    166.4
Execl Throughput                                                43.0           459.3    106.8
File Copy 1024 bufsize 2000 maxblocks     3960.0     176763.8    446.4
File Copy 256 bufsize 500 maxblocks         1655.0      53800.0    325.1
File Copy 4096 bufsize 8000 maxblocks     5800.0     431407.5    743.8
Pipe Throughput                                           12440.0     430477.6    346.0
Pipe-based Context Switching                      4000.0      73425.8    183.6
Process Creation                                              126.0       1360.4    108.0
Shell Scripts (1 concurrent)                              42.4       1188.6    280.3
Shell Scripts (8 concurrent)                                6.0        783.3   1305.6
System Call Overhead                                 15000.0     969171.8    646.1
                                                                   ========
System Benchmarks Index Score                                         336.2
surely odroid-c2's score  is higer  than orane pi PC ( 336.2 X 202.0) .

 

 

and

old intel PC ,

smpboot: CPU0: Intel® Celeron® CPU    540  @ 1.86GHz (fam: 06, model: 16, stepping: 01)

System Benchmarks Index Score                                         616.3

 

only result , i cannot analyze .

---

regards

i try unixbench

 

https://twitter.com/brendangregg/status/618573280375484417

 

https://github.com/GoogleCloudPlatform/PerfKitBenchmarker/issues/125

 

Please people, stop fooling yourself using such crap and comparing numbers without meaning! It's absolutely useless to run this stuff from last century.

i see unixbench was outdated .

 

i want to mesure cpu speed  not hard disk speed .

i do not know the tool .

 

Once upon a time  there was super pi and i used it .

at present whic one is easy to do  to measure cpu speed ?

 

orange pi PC  with armbian is very tuned up  and is fast and i think it may be faster than raspberry pi 3 with rasbian  .

i think odroid-c2 must be fast considering its hardware .

Why do you care about measuring cpu speed if you already noticed what's important?

 

It always depends 100% on the use case. If you want to do number crunching... then CPU performance is important if you can't use GPGPU. Trying number crunching on slow ARM SoCs is moronic. So why not just dropping the whole idea to measure CPU performance at all?

 

When you use a desktop Linux then get fast storage with superiour random I/O (modern eMMC) and have a look whether you get 2D acceleration (not CPU related) and 3D acceleration (not CPU related) and video acceleration (not CPU related).

 

And then stuff like IRQ balancing, good dvfs operating point tuning to get the best thermal behaviour and so on is more important (that's what Armbian does: focusing on the relevant tuning parameters!) 

 

Simply stop stupid CPU performance benchmarking. Most if not all of the times this is only producing worthless numbers (that look pretty nice when graphs were made of so that really dumb people can draw conclusions from the length of graphs instead of switching on their brains to understand what's really important to get good 'real world' performance). These useless kitchen-sink micro-benchmarks are for morons and/or for phoronix.com.

 

Stuff like "quad core is better than dual core IF the workload makes use of parallelism" are so easy to understand that no one needs benchmarks to 'verify' this.

i tried sysbench .

 

sysbench  --test=cpu run

 

the result is

 

dmesg | grep CPU0
 

CPU0: Intel® Celeron® CPU          540  @ 1.86GHz (fam: 06, model: 16, stepping: 01)
Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          23.2413s
    total number of events:              10000
    total time taken by event execution: 23.2258
    per-request statistics:
         min:                                  2.20ms
         avg:                                  2.32ms
         max:                                 10.99ms
         approx.  95 percentile:               2.39ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   23.2258/0.00

 

 

odoroid-c2 over internet

 Threads started!
Done.

Maximum prime number checked in CPU test: 10000


Test execution summary:
    total time:                          9.6456s
    total number of events:              10000
    total time taken by event execution: 9.6433
    per-request statistics:
         min:                                  0.96ms
         avg:                                  0.96ms
         max:                                  1.42ms
         approx.  95 percentile:               0.97ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   9.6433/0.00

 

 

odroid-c2 is much faster than Celeron®  1.86GHz

odroid-c2 is much faster than Celeron®  1.86GHz

 

Plain wrong, sysbench ist just one of the many broken 'benchmarks' that can not be used to compare between different platforms. Unless your use case is calculating prime numbers sysbench results are 100 percent worthless. If you let it run on a more recent ultra slow Intel Atom it will outperform S905. And that S905 looks that good with this one specific sysbench test is due to using ARMv8 instruction set.

 

Some details: http://forum.odroid.com/viewtopic.php?f=136&t=19158

 

And the basic problem regarding mornonic kitchen-sink benchmarks: http://www.brendangregg.com/blog/2014-05-02/compilers-love-messing-with-benchmarks.html (by choosing wrong compiler settings your ODROID-C2 might be 100 times slower. And that's why this sort of 'fire and forget' benchmarking is always crap)

i give up benchmark .

 

at odroid-c2(armbian and over internet )

date ; ./pi.bat ; date
2016å¹´  6月  7æ—¥ ç«æ›œæ—¥ 05:15:29 JST
3.14159265359
2016å¹´  6月  7æ—¥ ç«æ›œæ—¥ 05:16:06 JST
37 sec

 

at Intel® Celeron® CPU          540  @ 1.86GHz (kali linux)

$ date ; ./pi.bat ; date
2016å¹´  6月  7æ—¥ ç«æ›œæ—¥ 05:16:17 JST
3.14159265359
2016å¹´  6月  7æ—¥ ç«æ›œæ—¥ 05:16:28 JST
11 sec

 

 

here

 cat pi.bat

 

#!/usr/bin/env python
def f(x):
  return 4 / ( 1.0 + x**2 )
n = 10000000
sum = 0
step = 1.0 / n
for i in range(0, n):
  x = ( i + 0.5 ) * step
  sum += f(x)

pi = sum * step
print pi
 

 

 

i get pi information at http://netbuffalo.doorblog.jp/archives/4783572.html

 

 

i tried arch linux on odroid-c2 , but almost all aur fails .

and

in openbsd , source(ports) is common  to all architecture .  .this is suprising .