This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

Why to go with Zephyr RTOS in nRF5340?

Summary: What are advantage bullet factors of Zephyr RTOS over FreeRTOS to choose this in nRF Connect SDK v1.5.0
Details:
We have Implemented our previous project using Nordic's nRF52832 SoC with FreeRTOS.
Now we are developing a new BLE Mesh based project with nRF5340 and it requires a RTOS.
Based on analysis
For nRF5340 based development we need to use nRF Connect SDK v1.5.0 which is merged with Zephyr RTOS
BLE based samples are all Zephyr RTOS based.
We would like to progress like below
1. Simple interface check without RTOS
2. BLE Mesh check without RTOS
3. Finally upgrade the BLE mesh with RTOS.
But the nRF Connect SDK v1.5.0 based development looks like not supporting without Zephyr RTOS.
Shall we get the below clarification:
1.Is any possibility to start the "BLE Mesh + nRF Connect SDK v1.5.0" based project without Zephyr RTOS?
2.What are the advantage bullet factors of Zephyr RTOS over FreeRTOS to choose this in nRF Connect SDK v1.5.0? .
Conclusion:
Please list the advantage bullet factors of Zephyr RTOS over FreeRTOS?

Awaiting for your response.

Regards,

S Mohan

Parents
  • Mohan, did you proceed any further with your considerations?

    The more we progress here in our current development, the less I see Zephyr as of any advantage at all. Yes, it is too big to fail, and eventually it will work - if your project can wait a couple of years.

    The mandatory Zephyr has caused us at least 6 months delay so far, and very basic functionalities are still not working in Zephyr. Our project is one year old and there are bugs we can't seem to solve unless we decide to learn the realms of Zephyr and change low level issues ourselves - after which we will have a hard time to use main Zephyr updates...

    We are seriously considering to 
    propose to our customer another 6 months delay on the project, and come up with our 'bare bone' solutions for nRF5340 with FreeRTOS or something similiar... Or maybe we should go as far as redesigning the HW to use a silicon that does not force Zephyr into their developers...

Reply
  • Mohan, did you proceed any further with your considerations?

    The more we progress here in our current development, the less I see Zephyr as of any advantage at all. Yes, it is too big to fail, and eventually it will work - if your project can wait a couple of years.

    The mandatory Zephyr has caused us at least 6 months delay so far, and very basic functionalities are still not working in Zephyr. Our project is one year old and there are bugs we can't seem to solve unless we decide to learn the realms of Zephyr and change low level issues ourselves - after which we will have a hard time to use main Zephyr updates...

    We are seriously considering to 
    propose to our customer another 6 months delay on the project, and come up with our 'bare bone' solutions for nRF5340 with FreeRTOS or something similiar... Or maybe we should go as far as redesigning the HW to use a silicon that does not force Zephyr into their developers...

Children
Related