IMHO:
- Despite all the loud announcements, network coverage for LTE-M & NBIoT is still poor.
- I am based in switzerland with (i think) a avarage to good mobile infrastructure. Nevertheless, the LTE-M and NBIoT Bands are not enabled on every antenna and the coverage is poor also from an international perspective. When you check the world coverage maps there is terra incognita all over the caribbean, africa etc.
Since the SARA 410 on the mkr1500 has no fallback to 2G or 3G it's a problem if you have a mobile device or want to operate in such areas. - The mkr1500 has no voice despite LTE-M can do voice.
Ergo: If your application is fixed on a place with a good LTE-M covarage and you don't need voice: yes.
- The main advantage of LTE-M/NBIoT they say is saving energy and better signal penetration into buildings. If you have no energy saving issues nor coverage its a: No.
- The mkr1500 is not unstable, the timing and the communication with the modem (which is a complete subsystem) only over the AT console makes it very difficult to write and debug stable code (as also with the mkr1400). The MKRNB library is only a guide to this galaxy and made to demonstrate the tech and the examples.
I wish Arduino will present a MKR1600 with a FULL LTE 5G Modem with voice over I2S and a 2/3G fallback as a worthy successor of the mkr1400!