Micro Magic RISC-V Core Claims to Beat Apple M1 and Arm Cortex-A9

(submitted by 01100011)

· · hntooter · 5 · 8 · 5

@hntooter Deeply skeptical. Too much appeal to authority in the article, not enough benchmarks, and zero actual measurements.

I know there were RISC-V prototypes that approached 80% the speed of a then-average Xeon core in the past (the BOOM architecture); but those were not yet market ready.

As someone who is squarely on "team RISC-V", I have to call bullshit on their claims. 5GHz is IBM mainframe territory.

@hntooter Really torn about RISC-V.
I'll be interested to see if/how it pans out.

re LB: specifically, 1 core running at 4.25GHz at 0.8V, consuming 200mW, gives 11K Coremark - or 55KCm/W

also, there's something really weird going on with the comparisons. the Cortex A9 is quoted as being over twice as fast as the Apple M1, which is... not very likely, all things considered! so probably best ignore them, and focus on the fact that RISC-V appears to enable really fast chips with really tiny power requirements

Cortex A9 and Apple M1 are pretty far apart. IIRC, the Cortex A9 was ONE CORE (not a whole CPU) from like 2011 running at 1-2GHz, back when ARM was still extremely low performance.

Sign in to participate in the conversation

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!