ARM mBed officially End of Life

[ralphrmartin]

I had an email from ARM today, with the following key points:

The Mbed platform and OS will reach end of life in July 2026, when the Mbed website will be archived and it will no longer be possible to build projects in our online tools
The device software - Mbed OS - is open source and will remain publicly available, but is no longer actively maintained by Arm
The Mbed TLS project is unaffected by this announcement and continues to be supported as part of the TrustedFirmware community project.

A community fork of Mbed OS, Mbed CE, is under active development.

[coppice]

So, what does this change represent? ARM don't think its worthwhile, or Mbed has developed enough momentum that it doesn't need them to drive it?

[ataradov]

ARM don't think its worthwhile

This. The idea behind mBed and CMSIS is to make all vendor's devices appear the same. Guess what do vendors not want? To appear the same and easily replaceable. So, the support from vendors was non-existent or very minimal.

Some lower level parts of CMSIS were adopted because it made sense. But all the high level stuff that made everyone look the same was also ignored.

[ralphrmartin]

The Arduino ecosystem also (more or less) tries to make everything the same, and that has worked out well for its community. I suspect the difference is that it is not mainly the chip makers (apart from perhaps Espressif) doing it in this case, it is the suppliers of development boards (or their users). They have much less in the way of resources than the chip manufacturers, and it helps them to leverage existing libraries.

[ataradov]

Arduino is a a board vendor and a community effort with a lot of support from Atmel. Atmel obviously had interest in Arduino success. But as Arduino gained more support from other vendors, Atmel interest reduced quite a bit.

The only party benefiting from mBed support is ARM and all the vendors will lose vendor lock-in if mBed is successful.

[ralphrmartin]

Seems like a strange time to give mBed up, when ARM may need all it can get to keep everyone (chip makers, board makers and developers) on board, rather than jumping ship to RISC-V.

[ataradov]

Major MCU vendors are not going to jump to RISC-V any time soon. The companies you see going to RISC-V were not primarily ARM customers anyway.

Maintaining mBed costs a lot of money and it does not have wide adoption. And why go against your actual customers?

[SiliconWizard]

That's an odd decision, not that I personally care, but those who have relied on it will not be happy.

[YTusername]

ARM killed Mbed 2 years before. The contributors mentioned the untold decision and that is documented here and here. In my opinion, the successor will be Zephyr OS, although there is a different methodology and structure between both. Arduino announced that Zypher will replace Mbed-core in its SDK.

[mac.6]

Major MCU vendors are not going to jump to RISC-V any time soon.

True for customer facing cores, but for non-customer facing cores, some have switched to risc-v.