1 1
tkaiser

sbc-bench

Recommended Posts

On 10/5/2018 at 6:07 AM, malvcr said:

This are numbers without the "elapsed" parameter.
 


Without AFALG

type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes  16384 bytes
aes-256-cbc      19776.55k    23565.55k    24981.25k    25360.04k    25556.85k    25471.66k

With AFALG

type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes  16384 bytes
aes-256-cbc       8008.18k    27638.99k   251372.80k  1167974.40k         infk  4513792.00k

 

I am not sure ... but it seems that afalg it is not available in the stock Armbian ( I checked this with a supported M2+).  With SBC seems important to have this available to use the machines potential.

 

ALG and Cryptp Blocks can be a bit complicated...

 

Based on your numbers - what moves more bits?

 

ARM does...

SOFTWARE

Doing aes-256-cbc for 3s on 16 size blocks: 3708058 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 64 size blocks: 1104719 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 256 size blocks: 292752 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 1024 size blocks: 74300 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 8192 size blocks: 9329 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 16384 size blocks: 4662 aes-256-cbc's in 3.00s

ALG

Doing aes-256-cbc for 3s on 16 size blocks: 129499 aes-256-cbc's in 2.95s
Doing aes-256-cbc for 3s on 64 size blocks: 115145 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 256 size blocks: 78540 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 1024 size blocks: 34189 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 8192 size blocks: 5404 aes-256-cbc's in 3.00s
Doing aes-256-cbc for 3s on 16384 size blocks: 2756 aes-256-cbc's in 3.00s

 

That being said, ALG is likely lower CPU usage overall, but running on ARM in your case is the right choice....

 

 

 

Share this post


Link to post
Share on other sites
On 10/5/2018 at 3:07 PM, malvcr said:

I am configuring another BPI-R2 machine, and I was checking the benchmarks.  For this I am using a 4.14.71 with Ubuntu.  The numbers are not better than my previous attempts

did you ever tried cryptodev for the R2. I think you had to compile openssl as well (as far as I now, there're some issues with openssls cryptodev implementation, but honestly never cared). Just out of curiosity.. :P I only gave it one shot back then cause the board had bigger issues than crypto in those days.

 

 

Share this post


Link to post
Share on other sites
Quote

did you ever tried cryptodev for the R2. I think you had to compile openssl as well (as far as I now, there're some issues with openssls cryptodev implementation, but honestly never cared). Just out of curiosity..  I only gave it one shot back then cause the board had bigger issues than crypto in those days.

 

Interesting ... I though my reply was not stored or deleted.  Anyway ...

 

I was checking with cryptodev and the BPI 4.4.70 some time ago (I have some posts in the BPI forum about R2).  It was faster, but what I didn't like about it is that cryptodev it is not accepted as an official kernel module.  This is why I was working with AF_ALG.  Also, although openssl works, it is a very big piece of code that could hide some "issues" for my security centered work (this already happened with openssl).  This is also the reason I am not continuing using BOTAN, a very complete ciphering platform, and I prefer something embedded in the Kernel or my own small and light framework.

 

In fact I am not 100% satisfied with AF_ALG.  It is a very artificial method (with a terrible documentation) and it doesn't work very well for small block sizes.  I was trying to mimic the openssl "speed" benchmark while encrypting with very big block sizes, but the "encript" openssl option  doesn't permit me to work them.  In such case, I prefer not to use openssl and to work my own tests with my code.

 

Let me see if I can have a minimum AF_ALG testing and basic ciphering tool to share with clear enough source code to play with.

Share this post


Link to post
Share on other sites

@tkaiser

Hi Thomas. I'm planning to make a video about the use/uselessness/problems of benchmarking SBC's.

Today I got a message from a subscriber about the Odroid H2, where he claimed the XU4 was a very slow SBC. His claim was backed up by "benchmarks" from ExplainingComputers. Here are those results.
afbeelding.png.5e0a3f2d9112c091762fa05ac76d8d42.png

Video ExplainingComputers :  Six SBC Benchmark: ODROID XU4, ROCKPro64 & More!

Here the ROCK64 seems to outperform the XU4. We all know that's not right.

I would use SBC-bench if ok for you and Blender to show the difference in results using different platforms, kernels and settings on the same SBC. I think with the M4. Lubuntu xenial armhf vs Lubuntu bionic armhf vs Lubuntu bionic arm64 and Armbian stretch vs bionic. Those differences are big.
And the Raspberry 3B+ with ram overclock and without to show importance of ram+cpu.

I would make different subjects. Of which :

 

Problems with Benchmarks and sollutions

SBC-Bench (how does it work, what does it do)
Differences between different cortexes A7, A53, A72, ... (I'll need to do a lot of homework for that, if you could elaborate on it, please do)

Importance of RAM speed with CPU speed, and other parameters

Cheating manufacturers (Amlogic with C2, Raspberry Pi with 3B+, any others I should mention?)

Conclusion...

So with this I ask your permission to use SBC-bench, and quote you out the readme.md and eplanations and insights in the results.md file. And if you would like me to mention something, please tell me. Or if you want you could record an audio/video file with your words to add in the video(just a thought)

Did you start a draft for the "Interpreting results" part yet?

I'll be busy for at least another week gathering information. When done I'll share my results, and I''ll say what I'm going to use from your texts in the video.
Sorry for the long post.
Greetings.
 

Share this post


Link to post
Share on other sites
2 hours ago, NicoD said:

Here the ROCK64 seems to outperform the XU4. We all know that's not right.

I would use SBC-bench if ok for you and Blender to show the difference in results using different platforms, kernels and settings on the same SBC. I think with the M4. Lubuntu xenial armhf vs Lubuntu bionic armhf vs Lubuntu bionic arm64 and Armbian stretch vs bionic. Those differences are big.
And the Raspberry 3B+ with ram overclock and without to show importance of ram+cpu.

I would make different subjects. Of which :

 

Problems with Benchmarks and sollutions

SBC-Bench (how does it work, what does it do)
Differences between different cortexes A7, A53, A72, ... (I'll need to do a lot of homework for that, if you could elaborate on it, please do)

Importance of RAM speed with CPU speed, and other parameters

Cheating manufacturers (Amlogic with C2, Raspberry Pi with 3B+, any others I should mention?)

 

Gah - watched the video - and a lot of problems across the board (pardon the pun).

 

Different kernels, built with different versions of GCC, userland (for example, Raspbian userland is all ARMv6 with exception of the kernel for the A7/A53 boards)....

 

(I wouldn't have included the any of the Pi's in the set of boards being evaluated because of the userland - <soapbox> nothing against Pi's in general, one must appreciate that 35M+ boards means they're doing something right, and they've spawned an entire HW/SW ecosystem around their platform, that's ok - and that ecosystem has in turn made affordable ARM boards available for hobbyists, makers, and developers - before Pi, if one wanted to do development around ARM, boards were expensive, and SW support was very limited to the vendor BSP - these days, it's a lot more open - not perfect, but much better than it was</soapbox>)

 

Rock64 vs Odroid XU4 - Quad A53 vs A7/A15 big.LITTLE - the big.LITTLE is a challenge for the scheduler, and depending on the BSP from the OEM, it's easy to get wrong, where threads can land on the lesser preferred core, this is an issue even on Android, where much work has been done outside of the mainline kernels (ARM and Qualcomm, I know they've done a lot of research there, but much of that has not been pushed back to mainline).

 

In my experience, with supported boards (for me this is Tinker and NanoPi NEO), Armbian is generally faster than the vendor's images - and that's doing Byte-Unixbench, which is discounted because it is compiler sensitive - that being said, it's still a useful tool when comparing apples to apples (e.g. tweaking settings on the same OS/Platform, but comparing Platform A to Platform B, one has to take the results with a grain of salt)

 

I haven't found a lot of evidence of cheating by any of the SBC vendors - it's really hard to do with FOSS, compared to Android, where cheating has occurred with certain OEM's and specific benchmark APK's - Android has enough hooks to enable this kind of cheating in any event.

 

sbc-bench, in my humble opinion, is a good benchmark for supported boards - as long as the boards being compared are all on the same version of Armbian - and this is made clear in the script comments (please review the script on github, and @tkaiser has been pushing updates, so if one has cloned the repo, it's worthwhile to do a git pull to get the latest revision.

 

To answer your question about the different versions of Cortex...

 

Small Cores - A7, A53 are the low power cores focused on efficiency

Big Cores - A15, A12(A17), A72 - big cores... 

 

Think of it like Atom (Small Core) vs Core i3/i5/i7 (Big Core) - even at the same clock, the big core is going to get more work done, but perhaps at the cost of heat, so thermal solution needs to be considered.

Share this post


Link to post
Share on other sites
On 10/20/2018 at 11:52 AM, NicoD said:

o with this I ask your permission to use SBC-bench, and quote you out the readme.md and eplanations and insights in the results.md file. And if you would like me to mention something, please tell me. Or if you want you could record an audio/video file with your words to add in the video(just a thought)

Did you start a draft for the "Interpreting results" part yet?

 

Up to @tkaiser for results on sbc-bench...

 

working on an addition - byte-unixbench and sorting out things... removing some gcc over optimizations, looking at threads...

 

https://github.com/sfx2000/byte-unixbench

 

It's a better bench than sysbench, and portable... Doing a -c 1 -1 and -c4 -i 1  keeps things short - however - letting it run thru pushes heat/throttles...

 

UnixBench is interesting from a system perspective...

 

RPI3 B Plus vs Tinker....

 

Tinker is 15 pounds of power in a 5 pound sack - RPi3 B+ is a CPU that can do better that it is with raspbian....

 

Tinkerboard - Cortex-A12/A17 - Armbian
------------------------------------------------------------------------
Benchmark Run: Sat Oct 20 2018 17:02:37 - 17:31:22
4 CPUs in system; running 1 parallel copy of tests

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    8709974.2    746.4
Double-Precision Whetstone                       55.0       1031.4    187.5
Execl Throughput                                 43.0       1095.7    254.8
File Copy 1024 bufsize 2000 maxblocks          3960.0      91960.7    232.2
File Copy 256 bufsize 500 maxblocks            1655.0      26583.4    160.6
File Copy 4096 bufsize 8000 maxblocks          5800.0     246267.0    424.6
Pipe Throughput                               12440.0     149851.8    120.5
Pipe-based Context Switching                   4000.0      25850.9     64.6
Process Creation                                126.0       2429.0    192.8
Shell Scripts (1 concurrent)                     42.4       2061.9    486.3
Shell Scripts (8 concurrent)                      6.0        432.0    720.1
System Call Overhead                          15000.0     442992.8    295.3
                                                                   ========
System Benchmarks Index Score                                         258.2

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   13538575.0   1160.1
Double-Precision Whetstone                       55.0       1982.4    360.4
Execl Throughput                                 43.0       1752.7    407.6
File Copy 1024 bufsize 2000 maxblocks          3960.0      87122.4    220.0
File Copy 256 bufsize 500 maxblocks            1655.0      22948.6    138.7
File Copy 4096 bufsize 8000 maxblocks          5800.0     281302.7    485.0
Pipe Throughput                               12440.0     321233.1    258.2
Pipe-based Context Switching                   4000.0      40012.9    100.0
Process Creation                                126.0       3820.3    303.2
Shell Scripts (1 concurrent)                     42.4       3399.0    801.7
Shell Scripts (8 concurrent)                      6.0        433.6    722.7
System Call Overhead                          15000.0     952658.0    635.1
                                                                   ========
System Benchmarks Index Score                                         373.1



Rpi 3B+ - Cortex-A53 - VCOS/ThreadX - Raspian
------------------------------------------------------------------------
Benchmark Run: Sat Oct 20 2018 17:02:32 - 17:30:38
4 CPUs in system; running 1 parallel copy of tests

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    4324740.1    370.6
Double-Precision Whetstone                       55.0        957.4    174.1
Execl Throughput                                 43.0        908.8    211.4
File Copy 1024 bufsize 2000 maxblocks          3960.0     140312.9    354.3
File Copy 256 bufsize 500 maxblocks            1655.0      40618.4    245.4
File Copy 4096 bufsize 8000 maxblocks          5800.0     353296.2    609.1
Pipe Throughput                               12440.0     280908.2    225.8
Pipe-based Context Switching                   4000.0      50734.2    126.8
Process Creation                                126.0       2212.2    175.6
Shell Scripts (1 concurrent)                     42.4       1780.5    419.9
Shell Scripts (8 concurrent)                      6.0        575.7    959.5
System Call Overhead                          15000.0     594784.0    396.5
                                                                   ========
System Benchmarks Index Score                                         302.2

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   17082008.4   1463.8
Double-Precision Whetstone                       55.0       3803.4    691.5
Execl Throughput                                 43.0       2240.8    521.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     228921.9    578.1
File Copy 256 bufsize 500 maxblocks            1655.0      62777.0    379.3
File Copy 4096 bufsize 8000 maxblocks          5800.0     578721.9    997.8
Pipe Throughput                               12440.0    1112342.2    894.2
Pipe-based Context Switching                   4000.0      98478.8    246.2
Process Creation                                126.0       4789.7    380.1
Shell Scripts (1 concurrent)                     42.4       4464.7   1053.0
Shell Scripts (8 concurrent)                      6.0        589.0    981.7
System Call Overhead                          15000.0    2289227.2   1526.2
                                                                   ========
System Benchmarks Index Score                                         705.6

 

Edited by sfx2000
clean up post formatting

Share this post


Link to post
Share on other sites

@sfx2000

I don't know how to interpret those numbers. Why are the 2 different scores?
What do you read out of those numbers? Why are they added up and made as a percentage? It are all results of different tests right? Then aren't you adding apples to pears and oranges? Just asking. I don't know.

This is what I've got until now. Still need to do a lot. Now Stretch with desktop, nightly in bionic and many Blenders.
I'm also going to use the Tinkerboard since many different images are available for it.
If you know any more reasons why benchmarks can differ, please let me know. Cheers.
I guess Thomas is on vacation. Haven't red anything of him since Thursday. (at least nothing new)
 

Reasons for difference in performances
--------------------------------------
throttling
32-bit/64-bit
Difference in cores A53/A7/A15/A72
distro (ubuntu/debian...)
distro version
kernel version
driver versions
compiler version
software version/outdated repositories
desktop Mate/Xfce/LXDE/...
display resolution/headless
background processes
cpu clockspeed
ram clockspeed/latency
ram useage /swap/zram
updates
process sheduler 
optimizations for the system/distro
crypto engine for encryption
Undervoltage 
config settings


SBC-Bench with NanoPi M4
------------------------
Armbian at 1.5Ghz/2Ghz
Lubuntu at 1.4Ghz/1.8Ghz

                       |SBC bench result  |CPU Miner          |7-zip small core |7-zip big core|7-zip multi avg. of 3 |Blender 
Armbian bionic          http://ix.io/1nLh  10.23kH/s           1335              2005           8352                   1h13m50s
Armbian bionic nightly  http://ix.io/1pDo  10.24kH/s           1329              1990           8292                   1h13m28s
Armbian stretch desktop http://ix.io/1odF  8.66kH/s            1350              1977           8400                   1h14m12s
Armbian stretch dsk nightly  //ix.io/1pM0  8.80kH/s            1359              1993           8500                   1h15m04s
Armbian stretch core no fan  //ix.io/1pKU  8.80-8.65kH/s       1353              1989           8461
Armbian stretch core withfan //ix.io/1pL9  8.76kH/s            1354              1988           8456
Armbian stretch core nightly //ix.io/1pLf  8.82kH/s            1357              1994           8494


Lubuntu arm64           http://ix.io/1oGJ  9.24kH/s CPU Miner  1056              1551           6943                   1h28m13s
Lubuntu Bionic armhf    http://ix.io/1pJ1                      1111              1769           7705                   2h02m54s
Lubuntu Xenial armhf	http://ix.io/1oCb                      989               1507           6339                   2h20m51s 

 

Share this post


Link to post
Share on other sites
7 hours ago, NicoD said:

I don't know how to interpret those numbers. Why are the 2 different scores?
What do you read out of those numbers? Why are they added up and made as a percentage? It are all results of different tests right? Then aren't you adding apples to pears and oranges? Just asking. I don't know.

This is what I've got until now. Still need to do a lot. Now Stretch with desktop, nightly in bionic and many Blenders.
I'm also going to use the Tinkerboard since many different images are available for it.
If you know any more reasons why benchmarks can differ, please let me know. Cheers.
I guess Thomas is on vacation. Haven't red anything of him since Thursday. (at least nothing new)

 

Tinker looks bad on those numbers - it's better than they suggest... just that UnixBench, by default, it is going to run for about 21 minutes, and that's a strong load for any thermal issues - and Tinker has a fair amount of that - like mentioned - 15W of load in a 5W basket - Pi3B+ just does better there, as would any similar chip...

 

The big core A12/A17 on Tinker does show and indicate the challenges with big.LITTLE where we have mixed cores - and we've seen this with Android, and excessive throttling there on certain handsets running Android.

 

The problem with Tinker is that it's power hungry and runs hot, it's big core only - so under sustained load with a typical install - e.g. Asus provided heatsink, and a decent MicroUSB power supply that can drive the board most times - it does tend to suffer a bit, it becomes heat soaked, and hopelessly throttled - and the numbers above show this...

 

To get best performance out of Tinker - one does have to look at driving power thru the 40-pin interface with a bench quality power supply, and active cooling for the chip - properly powered and cooled, it's a good challenger for Intel's Baytrail/Braswell chips - look at chromebooks, where the RK chip does a decent job...

Edited by sfx2000
fix typo

Share this post


Link to post
Share on other sites
On 10/20/2018 at 8:52 PM, NicoD said:

@tkaiser

afbeelding.png.5e0a3f2d9112c091762fa05ac76d8d42.png

Video ExplainingComputers : &nbsp;Six SBC Benchmark: ODROID XU4, ROCKPro64 & More!

Here the ROCK64 seems to outperform the XU4. We all know that's not right.

@sfx2000
I can now explain how these scores came to be. Sysbench does 10x better on 64-bit than 32-bit. Also different versions are used.
Gimp only uses 1 core. And also different versions. This works better in 32-bit than 64-bit.
Boot times can't be compared. Rasps have Raspbian which is optimized for the Rasp. Different SBC's take longer to start the boot process. He used different OS'es(Ubuntu/Debian).

@tkaiser

There is a small problem with 7-zip scores because it's not using 100% when doing multi-core. The percentages differ with the same board and different OS'es.
M4 Armbian stretch desktop http://ix.io/1odF  Tot avr load: 518

M4 Armbian bionic desktop  http://ix.io/1nLh  Tot avr : 525
Small differences, but it makes it a bit harder to assess.
Doing for example 2*2005 + 4*1335 = 9350 (armbian bionic)         Gives a number of the potential performance since those scores are 100% of the used core. Maybe it's a bit more exact to compare. Also with Intel devices which do get 100% here.
I also don't think it's good to use those total MIPS that's an average of the compressing and decompressing. It are 2 different numbers of different tasks. So I've got a problem with mixing them. I use decomressing MIPS here so it's a result of 1 task.

7-zip is the best tool I've found until now. It doesn't care about 32/64-bit. CPU miner seems to be ok for 64-bit. Blender also does well, but you can't compare 32-bit with 64-bit.
A lot more results are to come.
@tkaiser

Would a short compiler test be useful for SBC-bench?
I can't use sbc-bench on the Tinker Board in Elar Xubuntu bionic and Armbian Bionic. Too high system load. Even when booting into core.
Armbian Stretch on-board Wifi stops working about one minute after boot. Can't connect to router. Wifi dongles give a lot of system load, so SBC-bench doesn't start. It wants to start with all wifi off. But then it can't download the software and upload the results. Maybe give an option to start after a while even when the system load isn't 0.1.

These are not scores to compare SBC's. But to compare benchmark tools.

Blender : BMW render @ 1080p
Gimp : BMW render result 1080p Filters -> Artistic -> Van Gogh -> ok
Sysbench : sysbench --test=cpu --cpu-max-prime=20000 -num-threads= "number of threads" run
7-zip : Numbers are average of 3 of decompressing only

All tests are done with a fan when necessary so no throttling occurs.

NanoPi M4              |SBC bench result  |CPU Miner          |7-zip s/c |7-zip b/c     |7-zip multi avg. of 3 |Blender     |GIMP    |GTKPerf |Sysbench
Armbian bionic          http://ix.io/1nLh  10.23kH/s           1335       2005           8352                   1h13m50s     0m29s5   5.06s    26763
Armbian bionic nightly  http://ix.io/1pDo  10.24kH/s           1329       1990           8292                   1h13m28s     0m29s    5.12s    26733
Armbian stretch desktop http://ix.io/1odF  8.66kH/s            1350       1977           8400                   1h14m12s     0m31s    5.24s    3.1s
Armbian stretch dsk nightly  //ix.io/1pM0  8.80kH/s            1359       1993           8500                   1h15m04s     0m31s    5.32s    3.3s
Armbian stretch core no fan  //ix.io/1pKU  8.80-8.65kH/s       1353       1989           8461
Armbian stretch core         //ix.io/1pL9  8.76kH/s            1354       1988           8456
Armbian stretch core nightly //ix.io/1pLf  8.82kH/s            1357       1994           8494

Lubuntu arm64           http://ix.io/1oGJ  9.24kH/s CPU Miner  1056       1551           6943                   1h28m13s             
Lubuntu Bionic armhf    http://ix.io/1pJ1                      1111       1769           7705                   2h02m54s     0m57s    6.97s    1666
Lubuntu Xenial armhf	http://ix.io/1oCb                      989        1507           6339                   2h20m51s     0m59s    49.77s   49.7s

Asus Tinker board      |SBC bench result  |CPU Miner          |7-zip big core           |7-zip multi avg. of 3 |Blender     |GIMP    |GTKPerf |Sysbench
Tinker OS 9.5 Stretch   http://ix.io/1pRN                      1983                      7536                   2h55m00s     1m19s    189.82s  63.7s

Odroid C2              |SBC bench result  |CPU Miner          |7-zip big core           |7-zip multi avg. of 3 |Blender     |GIMP    |GTKPerf |Sysbench
Armbian Stretch Core    http://ix.io/1pZu  4.65kH/s            1390                      5342
Armbian Stretch Core Nightly //ix.io/1pZJ  4.66kH/s            1391                      5340
Armbian Stretch Desktop NGHT //ix.io/1p02  4.59kH/s            1394                      5356                                1m23s    12s      6.0s

Software versions
-----------------
                                      GIMP        Blender      GTKPerf     SysBench                     SBC-bench
M4 : Lubuntu Xenial armhf                         2.79b        0.40        0.4.12                       0.6.1
     Lubuntu Bionic armhf :           2.8.22      2.79b        0.40        1.0.11 LuaJIT 2.1.0-beta3    0.6.1
     Armbian Stretch desktop :        2.8.18      2.79b        0.40        0.4.12                       0.6.1 
     Armbian Bionic :                 2.8.22      2.79b        0.40        1.0.11 LuaJIT 2.1.0-beta3    0.6.1
Tinker : TinkerOS 9.5 Stretch :       2.8.18      2.79b        0.40        0.4.12                       0.6.1
Odroid C2 : Armbian Stretch 9.5 :     2.8.18                   0.40        0.4.12


CPU Clocks
----------
NanoPi M4 : Armbian Bionic/Stretch : 2x2Ghz + 4X1.5Ghz
            Lubuntu armhf/ARM64    : 2x1.8Ghz + 4X1.4Ghz
Tinker :    TinkerOS               : 4x1.8Ghz
Odroid C2 : Armbian Stretch        : 4x1.5Ghz



 

 

Share this post


Link to post
Share on other sites
8 hours ago, NicoD said:

I can't use sbc-bench on the Tinker Board in Elar Xubuntu bionic and Armbian Bionic. Too high system load. Even when booting into core.
Armbian Stretch on-board Wifi stops working about one minute after boot. Can't connect to router. Wifi dongles give a lot of system load, so SBC-bench doesn't start. It wants to start with all wifi off. But then it can't download the software and upload the results. Maybe give an option to start after a while even when the system load isn't 0.1.

 

With Tinker - check your power - if going over the MicroUSB, Tinker is very sensitive to poor quality cables and less than sufficient power supply current.

 

Search around the forums, and there's good tips on getting Tinker stable with power -- @TonyMac32 has done a lot of work there documenting the challenges of this particular board.

Share this post


Link to post
Share on other sites
10 hours ago, sfx2000 said:

 

With Tinker - check your power - if going over the MicroUSB, Tinker is very sensitive to poor quality cables and less than sufficient power supply current.

 

Search around the forums, and there's good tips on getting Tinker stable with power -- @TonyMac32 has done a lot of work there documenting the challenges of this particular board.

My power is stable. It's just the system load that stays too high. It did work in TinkerOS Stretch. But Xubuntu Bionic and Armbian Bionic keeps having a high system load. This because of the wifi. It doesn't work without internet, and I don't have a cable here. I'll redo it when I've finished the others. Cheers

Share this post


Link to post
Share on other sites

I am trying sbc-bench v0.6.2 on the dev branch of Orange Pi Plus 2E and see these errors:

Checking cpufreq OPP... Done.

./sbc-bench.sh: line 783: [: too many arguments
./sbc-bench.sh: line 786: 277
217 - 277
217 : syntax error in expression (error token is "217 - 277
217 ")
ATTENTION: Throttling might have occured. Check the log for details.

Here are test results: http://ix.io/1skx

Share this post


Link to post
Share on other sites
On 11/29/2018 at 6:08 AM, jeanrhum said:

An additional bench with a snapdragon 820 soc (Inforce 6640): http://ix.io/1uOW

I used default small heatsink without any fan, so it throtles on some benchmarks which heavy cpu loads like cpuminer.

 

Nice... pretty much lines up with what I've seen on other QC Kyro cores

 

expensive board, but Qualcomm could make things a bit easier and capture some mindshare if they were to bring prices down into something more affordable.

Share this post


Link to post
Share on other sites

Stock Armbian 5.65 on FriendlyARM NeoPI Nano's... clean power and good cooling.

 

NanoPi NEO v1.31 - http://ix.io/1vG7

NanoPi NEO2 v1.1 - http://ix.io/1vG2

 

Clean install on both boards and an apt-update/upgrade to get them current (Neo gets a newer kernel), no other tweaks - since both are very similar, it's a good comparison between Cortex-A7 (H3) vs. Cortex-A53 (H5)

 

Anyways - UnixBench is still interesting as it flexs the SoC across different domains...

 

NanoPi NEO - Cortex-A7

Single Core...

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    3719241.1    318.7
Double-Precision Whetstone                       55.0        643.0    116.9
Execl Throughput                                 43.0        696.9    162.1
File Copy 1024 bufsize 2000 maxblocks          3960.0      87513.0    221.0
File Copy 256 bufsize 500 maxblocks            1655.0      28297.5    171.0
File Copy 4096 bufsize 8000 maxblocks          5800.0     191648.4    330.4
Pipe Throughput                               12440.0     208879.5    167.9
Pipe-based Context Switching                   4000.0      47797.1    119.5
Process Creation                                126.0       1786.4    141.8
Shell Scripts (1 concurrent)                     42.4       1433.1    338.0
Shell Scripts (8 concurrent)                      6.0        387.6    646.1
System Call Overhead                          15000.0     478281.7    318.9
                                                                   ========
System Benchmarks Index Score                                         222.8

4 cores here on A7...

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   14873484.5   1274.5
Double-Precision Whetstone                       55.0       2572.7    467.8
Execl Throughput                                 43.0       1554.9    361.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     147529.5    372.5
File Copy 256 bufsize 500 maxblocks            1655.0      46842.5    283.0
File Copy 4096 bufsize 8000 maxblocks          5800.0     364298.5    628.1
Pipe Throughput                               12440.0     832938.2    669.6
Pipe-based Context Switching                   4000.0      90871.1    227.2
Process Creation                                126.0       3765.5    298.9
Shell Scripts (1 concurrent)                     42.4       3105.9    732.5
Shell Scripts (8 concurrent)                      6.0        396.2    660.3
System Call Overhead                          15000.0    1835773.3   1223.8
                                                                   ========
System Benchmarks Index Score                                         519.3

Now we do A53 - similar clocks - so we see arch improvements generally...

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    5508085.9    472.0
Double-Precision Whetstone                       55.0       1036.1    188.4
Execl Throughput                                 43.0       1155.1    268.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     151730.9    383.2
File Copy 256 bufsize 500 maxblocks            1655.0      48463.7    292.8
File Copy 4096 bufsize 8000 maxblocks          5800.0     335435.4    578.3
Pipe Throughput                               12440.0     385417.6    309.8
Pipe-based Context Switching                   4000.0      74227.1    185.6
Process Creation                                126.0       3476.3    275.9
Shell Scripts (1 concurrent)                     42.4       2217.9    523.1
Shell Scripts (8 concurrent)                      6.0        621.3   1035.4
System Call Overhead                          15000.0     600185.6    400.1
                                                                   ========
System Benchmarks Index Score                                         363.5

4 cores all busy now - same clocks at the A7 NanoPI NEO above...

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   22036695.3   1888.3
Double-Precision Whetstone                       55.0       4145.6    753.8
Execl Throughput                                 43.0       2657.1    617.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     290036.5    732.4
File Copy 256 bufsize 500 maxblocks            1655.0      83344.5    503.6
File Copy 4096 bufsize 8000 maxblocks          5800.0     710744.4   1225.4
Pipe Throughput                               12440.0    1532780.9   1232.1
Pipe-based Context Switching                   4000.0     125344.5    313.4
Process Creation                                126.0       6891.0    546.9
Shell Scripts (1 concurrent)                     42.4       4948.5   1167.1
Shell Scripts (8 concurrent)                      6.0        626.0   1043.3
System Call Overhead                          15000.0    2324664.3   1549.8
                                                                   ========
System Benchmarks Index Score                                         857.7

So not as sexy as the OpenSSL performance, but the uplift between H3-Armv7-A and H5-Armv8 is always a good thing when everything else is close to equal...

 

40 percent improvement across the board - that's A53 vs A7 in general work...

 

@tkaiser - sysbench is silly, I agree - but UnixBench is a fair flex if one is careful on compiler options not to over-optimize things there...

 

Some boards get hopelessly heat soaked on this benchmark - TinkerBoard is a good example with the Asus supplied HS and good power - but with Armbian min-clocking to 600MHz, it cannot clock back enough to recover....

Edited by sfx2000
add unixbench scores

Share this post


Link to post
Share on other sites
1 1