What does faster mean really in this contest? Should we consider ARM7 and ARM9 as microcontrollers, in which case this Microchip's product would pale into insignificance? We should consider performance per Watt, performance per Dollar among other factors such as software eco-systems, security of supply etc. It's not just about an arbitrary CoreMark number alas....
Microchip's new 32-bit connectivity MCU might well be targeted for the IoT applications where power is not a problem. The features look interesting. I would like to evaluate how the DSP engine along with 159 additional DSP-centric instructions helps in improving the performance of the MIPS microAptiv Core.
It certainly isn't the fastest 32-bit MCU. While it might outperform Cortex-M4 when not using floating point, it has caches and doesn't run from flash. So it should be compared to Cortex-R4/R5, which run at much higher frequencies (Samsung SSDs use 3 at 400MHz) and have better IPC as well.
I am glad that Microchip has actually started to do something with their 32bit line of products. FOr the longest time, they were not really all that great compared to their dsPIC line of products. I still am hoping that they might decide to come out with something that has a FPU on it. As for their claim as the fastest 32bit device, I will have to dig into that a bit more.
January 2016 Cartoon Caption ContestBob's punishment for missing his deadline was to be tied to his chair tantalizingly close to a disconnected cable, with one hand superglued to his desk and another to his chin, while the pages from his wall calendar were slowly torn away.122 comments