Skip navigation
> RoadTest Reviews

RoadTest Review a Raspberry Pi 3 Model B ! - Review

Scoring

Product Performed to Expectations: 9
Specifications were sufficient to design with: 10
Demo Software was of good quality: 10
Product was easy to use: 8
Support materials were available: 10
The price to performance ratio was good: 10
TotalScore: 57 / 60
  • RoadTest: RoadTest Review a Raspberry Pi 3 Model B !
  • Buy Now
  • Evaluation Type: Independent Products
  • Was everything in the box required?: Yes - There was no sdcard, and an old NOOBS card from a Pi B (not the B+) didn't work.
  • Comparable Products/Other parts you considered: The Pi rules!
  • What were the biggest problems encountered?: an old NOOBS card from a Pi B (not the B+) didn't work.

  • Detailed Review:

    Introduction

    The topic I promised  to discuss in my roadtest application is heat dissipation related to use of the Pi 3 for home automation.

    There were some rumors on the net that the processor temperature can be as high as 100 degrees celsius at 100% CPU.

     

    Current setup

    Currently I'm running an home automation system using a Pi 2 model B, which controls the lights in the house, measures temperature and also a connected camera detects intruders.

    For monitoring the system itself I have a temperature sensor in the cabinet and also monitor the CPU temperature.

    Here is a picture of the webpage which shows me the graph of the current month:

    rpizolder_-_temperatuur_logger.png

    You will see the temperature of the power supply (green), the case itself (purple) and the CPU (top one).

    You can see that it follows the ambient temperature, including day/night rhythm.

     

    What I tested

    So I was curious about the heat transmission is om the Pi 3. I saw some tests on the internet who states that the CPU under 100% load can go higher than 100 degrees celsius. (http://www.gizmodo.com.au/2016/03/the-raspberry-pi-3s-quad-core-cpu-can-hit-a-toasty-100c-under-load/ )

    For home automation I don't need a lot of computing performance, while power dissipation is much more important.

    First I did a CPU test using sys bench, on three types of the Raspberry Pi, my current home automation system (Pi 2), a Pi B+, and the Pi 3 from the roadtest.

    Here is a picture of the Pi B+ (left) and Pi 3, laying on my desk.

    Screenshot_27-05-16_22_16.png

    sys bench can be installed using: sudo apt-get install sysbench

    For monitoring the temperature I used RPiTemperature from: http://www.roylongbottom.org.uk/Raspberry_Pi_Stress_Tests.zip

     

    wget http://www.roylongbottom.org.uk/Raspberry_Pi_Stress_Tests.zip
    unzip Raspberry_Pi_Stress_Tests.zip 
    cd Raspberry_Pi_Stress_Tests/
    chmod +x RPiTemperature 
    ./RPiTemperature passes 700, seconds 2 &
    

     

    Here is the result of sys bench on all three boards. Of course I know that --num-threads=4 doesn't make much sense on the Pi 2 and Pi B+, but I don't expect it influences the results a lot.

     

    Pi B+ cpu sys bench:

    pi@zolderpi ~/stres_test $ sysbench --num-threads=4 --test=cpu --cpu-max-prime=20000 --validate run
    sysbench 0.4.12:  multi-threaded system evaluation benchmark
    
    Running the test with following options:
    Number of threads: 4
    Additional request validation enabled.
    
    
    Doing CPU performance benchmark
    
    Threads started!
    Done.
    Maximum prime number checked in CPU test: 20000
    
    
    Test execution summary:
        total time:                          1329.8390s
        total number of events:              10000
        total time taken by event execution: 5318.6407
        per-request statistics:
             min:                                351.47ms
             avg:                                531.86ms
             max:                               1132.11ms
             approx.  95 percentile:             551.17ms
    
    Threads fairness:
        events (avg/stddev):           2500.0000/0.00
        execution time (avg/stddev):   1329.6602/0.13
    

    Pi 2 B cpu sysbench:

    sysbench 0.4.12:  multi-threaded system evaluation benchmark
    Running the test with following options:
    Number of threads: 4
    Additional request validation enabled.
    
    
    Doing CPU performance benchmark
    
    Threads started!
    Done.
    
    Maximum prime number checked in CPU test: 20000
    
    
    Test execution summary:
        total time:                          293.8832s
        total number of events:              10000
        total time taken by event execution: 1175.3010
        per-request statistics:
             min:                                117.00ms
             avg:                                117.53ms
             max:                                160.74ms
             approx.  95 percentile:             118.04ms
    
    Threads fairness:
        events (avg/stddev):           2500.0000/11.02
        execution time (avg/stddev):   293.8253/0.04
    

     

    Pi 3 B cpu sysbench:

     

    pi@pi1pi3test ~/stress_tests/Raspberry_Pi_Stress_Tests $ sysbench --num-threads=4 --test=cpu --cpu-max-prime=20000 --validate run
    sysbench 0.4.12:  multi-threaded system evaluation benchmark
    
    Running the test with following options:
    Number of threads: 4
    Additional request validation enabled.
    
    
    Doing CPU performance benchmark
    
    Threads started!
    Done.
    
    Maximum prime number checked in CPU test: 20000
    
    
    Test execution summary:
        total time:                          246.9762s
        total number of events:              10000
        total time taken by event execution: 987.6050
        per-request statistics:
             min:                                 96.08ms
             avg:                                 98.76ms
             max:                                214.33ms
             approx.  95 percentile:              97.46ms
    
    Threads fairness:
        events (avg/stddev):           2500.0000/54.59
        execution time (avg/stddev):   246.9012/0.02
    

     

    During the tests I logged the temperature with a 2 seconds interval and plotted the results using plot.ly. Below you will see the results.

      
    Plotly___Make_charts_and_dashboards_online.png
    We can conclude that the Pi B+ is the slowest of the three, which of course was expected. The temperature also is a lot lower.
    I was surprised that there seems to be not much difference between the Pi 2 and Pi 3. The Pi2 is slightly slower. The temperature of the Pi 2 although is much lower than its version 3 counterpart. Note that the Pi 2 was in a cabinet with power supply, while the Pi 3 was laying open on my desk. I expect that the temperature of the Pi 3 in a cabinet will be even higher.

    I also made some thermal images when running the tests simultaneously on the Pi B+ (left) and the Pi 3 laying on my desk.

     

    IR_0169.jpgIR_0171.jpgIR_0170.jpg
    The difference in temperature can clearly be observed, although I didn't see higher temperatures than 60 degrees (C). Which is much lower than reported by:(http://www.gizmodo.com.au/2016/03/the-raspberry-pi-3s-quad-core-cpu-can-hit-a-toasty-100c-under-load/ )
    Power consumption of the Pi 3 during this test was  with 0.43 A really acceptable:

    Screenshot_27-05-16_22_20.png

     

     

    Final remarks

    My conclusion is that I continue to use the Pi 2 for my home automation, and use the Pi 3 for more demanding tasks, such as software defined radio.

    During my experiments A Comprehensive Raspberry Pi 3 Benchmark was added to Element14 by Christopher Stanton.

    I hope my findings above will contribute a little bit to that.

     

    Do not hesitate to contact me when you have some questions.


Comments

Also Enrolling
Enrollment Closes: Dec 19 
Enroll
Enrollment Closes: Dec 8 
Enroll
Enrollment Closes: Dec 8 
Enroll
Enrollment Closes: Dec 12 
Enroll
Enrollment Closes: Dec 18 
Enroll
Enrollment Closes: Nov 30 
Enroll