I really think you should use a more representative benchmark. Heck, even just take Chrome and run some browser benchmark on it like speedometer.
replyI run a large number of benchmarks; the Hackaday article of course focuses on the headline findings, but you can see the rest here: https://github.com/geerlingguy/sbc-reviews/issues/111
reply