Clear Filters
Clear Filters

Can Raspberry Pi support package work with the Rock Pi 4C+ (Rockchip RK3399-T)

11 views (last 30 days)
MATLAB/Simulink Raspberry Pi support package works well with Raspberry Pi 4B and Arduino/ESP32 boards. Unfortunately, the Raspeberry Pi 4B and Arduino/ESP32 boards have very limited performance. Furthermore, the model 4B is currently suffering supply chain issues. On the other hand, the Rock Pi boards are available, reasonably priced and considerably faster that the model 4B. It would be great if the Raspberry Pi support package would support other higher perfomance processors such as the Rockchip.

Accepted Answer

Sylvain on 22 May 2023
The cuurent support packages Matlab and simulink for Raspberry Pi is not compatible with other single board computers than raspberry pi. My understanding is that it is due to the broadcom processor used on Rpi, vs Rockchip processors on other single board computers.
We may be quite a lot of people impacted by the shortage of Pi caused by the availabiliy of broadcom processors at the moment. Having a support package (or an update of the current package) would be great to keep afloat with the current shortage.
Best would be to submit a request for an additional support package for the RockChip familiy.

More Answers (1)

Prathyusha Appalla
Prathyusha Appalla on 14 Mar 2023
I understand that you want to work with other high-performance boards, specifically the Rock Pi 4C+, using Raspberry Pi support package in MATLAB. The MATLAB/Simulink hardware support package of Raspberry Pi supports the following list of hardware models as specified in the documentation link below.
Unfortunately, from the documentation, I find that currently this support package is not compatible with Rock Pi boards.
Walter Roberson
Walter Roberson on 17 Mar 2023
Could the sample rate limitations that I'm observing be due to the Raspberry Pi SPI Read/Write blocks implementation rather than the uC processing speed?
How are you transfering data to the host? If you are using USB then you have fundamental rate limitations due to the USB control cycle being relatively slow -- you do not get substantial improvements until you switch to "bulk rate" transfers rather than packeted transfers. "bulk rate" transfers require that the data be fairly buffered -- increasing the latency.
Daryn on 17 Mar 2023
Good point when operating in External (Monitor & Tune) mode and passing data to/from the Host via USB or Ethernet it will result in bottlenecks. I've been operating in the Normal (Build, Deploy & Start) mode, where a *.elf program is generated and downloaded to the Raspberry Pi then run locally. The *.elf file could also be launched in a Raspberry Pi Linux terminal without any connection to the Host.
I also tested the Analog I/O board with Arduino and ESP32 boards (no OS). It wasn't a good comparison since the master clocks for Arduino/ESP32 boards are way slower than the Raspi. Maybe if the Raspberry Pi support package could operate without a OS or maybe an RTOS instead?
Does anyone know if the staff at MATLAB/Simulink or a 3rd party developed the Raspberry Pi/Arduino/Beagle Bone support packages?

Sign in to comment.


Find more on Raspberry Pi Hardware in Help Center and File Exchange




Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!