Search by Tags

Verdin iMX8M Mini Power Consumption

 

Article updated at 06 Jul 2022
Compare with Revision

Overview

This is one of a series of posts about Toradex modules' power consumption measurements. This article provides information about the power consumption of Verdin iMX8M Mini.

The Verdin iMX8M Mini's power consumption depends on the devices connected to it. Below are the power consumption results of several test cases, demonstrating realistic system configurations at various CPU loads.

Disclaimer

Toradex tests its modules following internal protocols as a way of standardizing the results. The overall power consumption of each module depends on multiple factors, such as:

  • Environmental temperature
  • Particular Silicon Revision of the SoC
  • Versions of the Software Components installed on the module
  • Use cases

The following results were achieved by simulating an ideal scenario. Applications in different conditions may not achieve the exact same results. It is suggested that a customer measures its SoM's power consumption under its own particular use cases before designing hardware based on this data.

Testing Conditions

  • The tests were performed using 3 different modules of each model.
  • The provided data is the average of the results obtained for each module group.
  • All of the SoMs were heated up first (using a stress test) to simulate the conditions of a continuous use scenario.

0055 Verdin iMX8M Mini 2GB WB IT V1.0B

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idle 1.667 1.789
Idle (headless) 1.385 1.400
HD video playback 2.067 2.635
glmark2 stress 2.671 3.407
Maximum power consumption 3.750 4.585

0055 Verdin iMX8M Mini Quad 2GB WB IT V1.1A

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idling mode with screen 2.21 2.64
Idling mode without Ethernet (cable unplugged) (with screen) 1.55 2.07
Idling mode without Ethernet (cable unplugged) (no screen) 1.32 2.13
Idling mode with screen for 20 minutes 2.21 2.76
Idling mode without screen 1.92 2.34
Stress 1 cores 2.58 3.11
Stress 2 cores 2.90 3.54
Stress 3 cores 3.19 4.01
Stress 4 cores 3.50 4.40
All cores stressed without screen 3.22 4.17
CPU stress with iperf3 eth 3.49 4.22
CPU stress with iperf3 wifi 2.4GHz 4.87 5.76
CPU stress with iperf3 eth and wifi 2.4GHz 4.86 5.21
All cores stressed and glmark2 3.49 4.20
glmark2 2.82 3.33
CPU neon stress script 3.64 4.48
Max stress: all cores stressed, glmark2, Iperf3 Wi-Fi 2.4GHz and ethernet 5.03 5.53
Max stress: all cores stressed, glmark2, Iperf3 Wi-Fi 5GHz and ethernet 4.40 5.16
Iperf3 Wifi 2.4GHz no screen 3.89 4.35
Iperf3 Wifi 5GHz no screen 3.28 3.67
Iperf3 ethernet no screen 2.16 2.63
All cores stressed, glmark2, ethernet and Iperf3 3.72 4.48
Suspend mode 0.12 0.30
RTC clock (proper shutdown) 850nW 910nW
RTC clock (power cable removing shutdown ) 860nW 910nW

0055 Verdin iMX8M Mini Quad 2GB WB IT V1.1E

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idling mode with screen 2.16 2.24
idling without screen 1.85 1.88
idle mode with no eth phy 1.20 1.21
Stress 1 cores 2.21 2.44
Stress 2 cores 2.52 2.86
Stress 3 cores 2.82 3.36
all cores (4) stressed without screen 3.09 3.70
CPU stress with iperf3 eth 3.28 3.66
CPU stress with iperf3 wifi 2.4GHz 4.50 5.03
CPU stress with iperf3 wifi 5GHz 4.29 4.67
CPU stress with iperf3 eth and wifi 2.4 GHz 4.71 4.98
CPU stress with iperf3 eth and wifi 5 GHz 4.32 4.70
all cores stressed and glmark2 3.75 4.50
glmark2 2.66 2.91
cpu neon stress script 4.19 4.22
max stress: all cores stressed, glmark2 ,iperf3 wifi 2.4GHz and ethernet 5.13 5.71
max stress: all cores stressed, glmark2 ,iperf3 wifi 5GHz and ethernet 4.69 5.21
Iperf3 Wifi 2.4GHz no screen 3.23 3.34
Iperf3 Wifi 5GHz no screen 3.19 3.34
iperf3 eth no screen 2.11 2.16
All cores stressed, glmark2, ethernet iperf3 3.75 4.20
suspend mode idle 1.03 0.00
suspend mode deep RAM 0.26 0.00
poweroff state 0.00 0.00

0057 Verdin iMX8M Mini DualLite 1GB V1.1A

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idling mode with screen 1.84 2.17
Idling mode without Ethernet (cable unplugged) (with screen) 1.51 1.65
Idling mode without Ethernet (cable unplugged) (no screen) 1.20 1.60
Idling mode with screen for 20 minutes 2.15 2.45
Idling without screen 1.84 2.17
Stress 1 cores 2.63 2.85
Stress 2 cores 3.01 3.47
All cores stressed without screen 2.67 3.08
CPU stress with iperf3 eth 3.09 3.45
All cores stressed and glmark2 3.46 4.03
Glmark2 2.78 3.12
CPU neon stress script 3.66 3.74
Iperf3 eth no screen 2.10 2.47
Max stress: all cores stressed, glmark2 ,iperf3 ethernet 3.46 3.82
Suspend mode 0.10
RTC clock (proper shutdown) 830nW
RTC clock (power cable removing shutdown ) 840nW

0059 Verdin iMX8M Mini Quad 2GB IT V1.1A

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idling mode with screen 2.17 2.32
Idling mode without Ethernet (cable unplugged) (with screen) 1.52 1.65
Idling mode without Ethernet (cable unplugged) (no screen) 1.21 1.71
Idling mode with screen for 20 minutes 2.17 2.35
Idling without screen 1.85 2.02
Stress 1 cores 2.54 2.78
Stress 2 cores 2.86 3.28
Stress 3 cores 3.16 3.68
Stress 4 cores 3.46 4.08
All cores stressed without screen 3.15 3.82
CPU stress with iperf3 eth 3.25 3.53
All cores stressed and glmark2 3.32 3.92
Glmark2 2.78 3.07
CPU neon stress script * 3.64 4.63
Iperf3 ethernet no screen 2.13 2.32
Max stress: All cores stressed, glmark2, ethernet iperf3 3.51 4.25
Suspend mode 0.10 0.15
RTC clock (proper shutdown) 873nW 881nW
RTC clock (power cable removing shutdown ) 868nW 877nW

0059 Verdin iMX8M Mini Quad 2GB IT V1.1C

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idling mode with screen 2.13 2.16
idling without screen 1.79 1.82
idle mode with no eth phy 1.16 0.00
Stress 1 cores 2.17 2.38
Stress 2 cores 2.48 2.86
Stress 3 cores 2.78 3.29
all cores (4) stressed without screen 3.05 3.63
CPU stress with iperf3 eth 3.18 3.58
all cores stressed and glmark2 3.69 4.40
glmark2 2.62 2.87
cpu neon stress script 4.15 4.17
iperf3 eth no screen 2.07 2.09
All cores stressed, glmark2, ethernet iperf3 3.70 4.24
suspend mode idle 0.98 0.00
suspend mode deep RAM 0.23 0.00
poweroff state 0.00 0.00
rtc (proper shutdown) 845uW 0.00
rtc (power unplug) 845uW 0.00
Iperf3 Wifi 5GHz no screen 3.19 3.34
iperf3 eth no screen 2.11 2.16
All cores stressed, glmark2, ethernet iperf3 3.75 4.20
suspend mode idle 1.03 0.00
suspend mode deep RAM 0.26 0.00
poweroff state 0.00 0.00

0060 Verdin iMX8M Mini DualLite 1GB WB IT V1.1A

Results

Test case Mean Power Consumption [W] Max. Power Consumption [W]
Idling mode with screen 2.20 2.58
Idling mode without Ethernet (cable unplugged) (with screen) 2.27 2.96
Idling mode without Ethernet (cable unplugged) (no screen) 1.23 1.75
Idling mode with screen for 20 minutes 2.40 2.97
Idling without screen 1.88 2.33
Stress 1 cores 2.56 3.06
Stress 2 cores 2.87 3.38
All cores stressed without screen 2.56 3.18
CPU stress with iperf3 eth 2.98 3.43
CPU stress with iperf3 wifi 2.4GHz 4.74 5.11
CPU stress with iperf3 eth and wifi 2.4GHz 4.85 5.15
All cores stressed and glmark2 3.37 3.94
Glmark2 2.81 3.16
CPU neon stress script 3.36 3.75
Max stress: all cores stressed, glmark2 ,iperf3 wifi 2.4GHz and ethernet 5.18 5.68
Iperf3 Wifi 2.4GHz no screen 3.84 4.12
Iperf3 Wifi 5GHz no screen 3.26 3.54
Iperf3 eth no screen 2.17 2.56
All cores stressed, glmark2, ethernet iperf3 3.37 3.77
Suspend mode 0.07 0.10
RTC clock (proper shutdown) 820nW
RTC clock (power cable removing shutdown ) 840nW

0060 Verdin iMX8M Mini DualLite 1GB WB IT V1.1C

Results

Test case Mean Power Consumption [W] Max Power Consumption [W]
Idling mode with screen with phy 2,08 6,13
idling without screen with phy 1,51 4,89
Idling mode without screen without phy 0,89 4,54
Stress 1 cores 1,88 5,38
Stress 2 cores 2,19 5,72
CPU stress with iperf3 eth 2,34 5,80
CPU stress with iperf3 wifi 2.4GHz 3,79 7,06
CPU stress with iperf3 wifi 5 GHz 3,65 7,17
CPU stress with iperf3 eth and wifi 2.4GHz 3,92 7,04
CPU stress with iperf3 eth and wifi 5GHz 3,71 7,23
all cores stressed and glmark2 2,85 6,00
glmark2 2,34 5,56
cpu neon stress script 2,55 6,00
max stress: all cores stressed, glmark2 ,iperf3 wifi 2.4GHz and ethernet 4,55 7,50
max stress: all cores stressed, glmark2 ,iperf3 wifi 5GHz and ethernet 4,33 7,46
Iperf3 Wifi 2.4GHz no screen 3,13 6,87
Iperf3 Wifi 5GHz no screen 3,06 6,87
iperf3 eth no screen 1,78 5,61
All cores stressed, glmark2, ethernet iperf3 2,99 6,09
suspend mode idle 0,93
suspend mode deep RAM 0,12
RTC clock (proper shutdown) 824uW
RTC clock (power cable removing shutdown ) 821uW

Power Measurement of Verdin SoMs

The Verdin Development Board and the Dahlia Carrier Board have a current, voltage and power measurement IC. It is the Texas Instruments INA219.

You need to enable it in the device tree. Check out the articles Device Tree Customization and Device Tree Overlays (Linux) for instructions on how to do it. Below is the device tree code snippet with the required changes:

&hwmon {
        status = "okay";
};

Then the device will be available on /sys/class/hwmon/hwmonX/. X can be enumerated as 0 or 1, because there is also a temperature sensor attached on the same I2C bus.

If we assume it will be on hwmon0, read the power measurement in microwatts (µW):

# cat /sys/class/hwmon/hwmon0/power1_input

Therefore you must divide the result by 1000000 to get the value in watts (W).

Additional Resources

Product Page and Datasheet: Verdin iMX8M Mini

Verdin Family: Verdin SoM Family

Specifications and Design Guides: Carrier Board Design Guides

Thermal Management in Toradex Modules: Thermal Management

How to Suspend Your Module Running Linux: Suspend/Resume (Linux)