Skip navigation
> RoadTest Reviews

NXP Layerscape® LS1046A Freeway Board - Review


Product Performed to Expectations: 10
Specifications were sufficient to design with: 10
Demo Software was of good quality: 10
Product was easy to use: 9
Support materials were available: 10
The price to performance ratio was good: 10
TotalScore: 59 / 60
  • RoadTest: NXP Layerscape® LS1046A Freeway Board
  • Buy Now
  • Evaluation Type: Development Boards & Tools
  • Was everything in the box required?: Yes
  • Comparable Products/Other parts you considered: Raspberry Pi 4, Nvidia Jetson Nano
  • What were the biggest problems encountered?: Setting up Network and VNC access, Hardware Interfacing

  • Detailed Review:



    First of all, this is not an expert review on the NXP LS1046A Freeway development board, but an honest review through the eyes of an electronics hobbyist. I will explain how I felt about the board as a new user and how I overcame the problems in setting it up and running. I believe this would really help a new guy to get acquainted with the board and start his journey from there. I would avoid any technical jargon as much as possible to keep this review simple.


    I like to take this opportunity to thank rscasny for giving me the opportunity to roadtest this amazing product. I would also like to mention the technical team members of NXP team Bosco, Ran and Arjun for their support throughout my road test.




    When I applied for this roadtest, I had so many things in mind to be able to set up and running on this piece of beast. But it is never an easy path. We have to assume the worst when we start with any new board. This product is still in its developing stage and the software packages are in their maturing state. We can expect it to become more stable, user-friendly and simple in the near future.




    The package was delivered in a nice enclosed box.


    The box itself had form packaging where the freeway board and its components were safely placed.


    The package contained the following:


    1. Freeway board with enclosure

    2. Power supply cable

    3. Universal power adapter (a real lifesaver)

    4. Antennas

    5. USB cable

    6. Tiny little 1.27MM jumpers

    7. Preflashed SD card

    8. Information sheets



    You might also need the following just in case:


    1. Ethernet cable

    2. Micro SD adapter


    The specs of this board are as follows:


    • Arm® Cortex®-A72 cores @ 1.6GHz
    • 4GB DDR4 with ECC @ 2.1GT/s
    • 2x USB3.0 (host)
    • Micro SD slot
    • 2x M.2 slots for Wi-Fi, 4G/LTE, SSD
    • 4x 1GE Ethernet
    • Mikro-Click expansion for additional elements like NFC, Sensors, BLE/ZigBee, LoRa, etc
    • Expansion Headers for I2C, SPI, Clocks, Interrupts, GPIO
    • Passive cooling heat sink (no fan)
    • LSDK / OpenWRT / Ubuntu / EdgeScale support


    First Look


    The board can be booted out of the box with the help of the pre-flashed SD card. Don't forget to connect the two antennas provided. Once it boots up, it creates an SSID of the form "FRWY-LS10A-<mac_id>". Connect to it using the password provided in the information sheet. Now the default interface can be accessed using the local IP address This contains the various pre-flashed use cases for the board including Performance Benchmarking, Data Analytics, Machine learning, and some Demo videos. These are a takeaway place for advanced hardcore users who wants to go in-depth into performance and stuff. Even though I was more interested in the other hardware and software features of this board, this is a nice place to start learning about the stuff the board is actually capable of.


    GUI / System Performance


    There were Benchmarking with CoreMark and Dhrystone. By default, it starts with CoreMark. But me being not so geeky about this stuff, I won't be going much deep into it (The real fact is even if I run them, I won't be able to understand any of it ).






    This page showed, in brief, the Performance, Power consumption, temperature and Utilization of the CPU cores. However, there is an option for a detailed analysis of the system. This was something quite interesting. For real geeks, you might want to check this out as it gives a fully detailed analysis of each and everything that runs on the board in form of a complete timeline, graphs, charts, and values.


    Here are some interesting snapshots.


    Some interesting facts that I came across in this detailed analysis page are:

    1. The CPU detailed analysis page is very interactive and informative. All links to each part are shown on the right side. We can even export the analysis data according to our resolution (interval) required.

    2. On start, the service httpd takes the most CPU time 4*100%.

    3. On start, the service netdata took the highest RAM and keeps on using around 36MB throughout.

    4. On start, the service netdata takes most disk usage to 2.73mbps read speed and around 750 kbps write speed.

    5. After being idle for some tilme, 3.25 GB out of 4 GB of RAM was free.

    6. Detailed data is streamed at port 19999.


    There were some constant warnings in the detailed CPU analysis page which I guess would be normal. Also, device heated up on top (slightly) even after almost 30 mins of idle usage


    GUI / Data Analytics


    This contains a simulation of cold storage to demonstrate the Edge Computing capability of the board. All the sensor data are simulated. In the real case scenario, the data can be extracted from real sensors that are interfaced with the board. Eg: MikroClick accelerometer sensor.


    GUI / Machine Learning & Video Analytics

    It demonstrates the AI/ML capability of the system using two demos.


    1. An industrial safety and security demo using TensorFlow

    2. A people counting demo using YOLO.


    Unfortunately, I doot possess a USB webcam. Therefore I could not carry out these demos.


    GUI / Video Vault


    This has a couple of videos about the board and the demos. The notable thing is that the videos are streamed via WiFi using the M.2 Wifi card present on the board.




    Now since the geeky stuff are done, let's get down to real business.


    Network Setup


    There are two ways through which we can access the board over the network.


    1. Using the onboard Wifi hotspot

    2. Using any of the ethernet ports available of the board


    SSH Login


    SSH access is available through Wifi, Ethernet or USB Serial port.

    You can ssh to FRWY board with username/password being “user/user”.


    If you would like to log in as root, please follow the steps below.


    1. Access the FRWY board console using Wifi or Ethernet

    2. user@localhost:~# vim /etc/ssh/sshd_config

    3. Uncomment the line “PermitRootLogin yes” and save

    4. user@localhost:~# service sshd restart


    Now you will be able to ssh into the FRWY board the username/password being root/root.


    Internet Access


    The demo software currently running on your box is the early release for demo use, with no router system is supported on FRWY.

    NXP has added the support in the later software release which will be available for download directly from With v3.1.7, the NXP board comes with router support.



    VNC Access


    VNC access is one of the most things required for this board for hobbyists like me. Till now I am not able to make it work. I will update this section once I am successful.


    Hardware Interfacing


    The board comes with Mikro-Click expansion for additional elements like NFC, Sensors, BLE/ZigBee, LoRa, etc and expansion Headers for I2C, SPI, Clocks, Interrupts, GPIO. In order to attach various peripherals, we need to expose the board outside to access the interfaces.

    However, the board cannot be completely taken out as one of the screws that attach the board to the base of the enclosure is not easily accessible using the normal screwdriver.

    I tried to interface the MicroClick Accel click board via the Micro BUS interface. For that, we need to screw out the MicroBus cover panel fixed on top of the Micro SD card slot.

    Unfortunately, I was not able to work it out yet. I will update this section too once I am successful.


    However, the sensor data can be simulated for the Data Analytics demo which uses the Accel sensor via MicroBUS connection.


    To start demo:

    sensor.html -> sensor_demo.php -> -> It will configure the adxl345 module sensor.html -> sensor_demo.php -> -> read -> It will read the data from sensor 1 by default and show on gauge and graph.


    To get all sensor details list:

    sensor/sensor.html -> sensor_demo.php -> -> read -> It will update for sensor 1

    sensor/sensor.html -> sensor_demo.php -> generate_data -> data1.txt -> It will update for 2 to 50 sensor in the list


    On selecting particular sensor data on gauge and graph:

    sensor/sensor.html -> sensor_demo.php -> generate_data -> data1.txt -> It will give the data for particular sensor


    To stop demo:

    sensor.html -> sensor_demo.php -> it will stop the showing data on gauge and graph




    The main features which I liked about this board are:


    1. Ability to run a full-fledged Linux environment.

    2. Availability of various hardware interfacing options.

    3. Availability of Network access.

    4. Ability to perform real-time A/V processing, ML and AI.


    The places where the product could really improve are:


    1. Polished bootloader with out of the box VNC and network access.

    2. Ease of interfacing.

    3. Ease of access to GPIO pins.

    4. Onboard Flash boot.

    5. Active cooling.


    Next Steps:


    1. Get the VNC working.

    2. Get the hardware interfaces running.


    This road test will be updated with any future details. Overall the product is very good. I really hope this will continue to improve in future versions.


    I wanted to compare the performance of this board with Raspberry Pi 4 and Jetson Nano. As I do not own any of these as of yet, the comparison may be done once I am able to procure them in future.




    Product Page:


    Quick Start Guide:


Also Enrolling

Enrollment Closes: Oct 5 
Enrollment Closes: Oct 30 
Enrollment Closes: Oct 29 
Enrollment Closes: Oct 29 
Enrollment Closes: Oct 26 
Enrollment Closes: Oct 5 
Enrollment Closes: Oct 9