I just got mine Rock 5B yesterday and I’m playing with some benchmarks at this moment. And what I’ve noticed is that it was advertised as 2.4 GHz for A76 cores, but scaling_max_freq is hard capped to 2.25 at this moment.
I wonder if that was changed for some reason or that’s a bug?
Welcome to the world of PVTM.
Radxa should’ve been as smart as Khadas and adjust their marketing efforts to speak of “A76 @ 2.25 GHz” then surprising some customers with higher clockspeeds.
BTW: By tweaking device-tree settings you can get higher (up to 2700 MHz) but IMO that’s somewhat stupid since performance gains are rather laughable and consumption increase massive at the same time. Details you may find in the ‘Debug Party’ thread here in this forum.
Hi,I’m working on DT of 3588s to make it faster.Can you please provide the output of this:
dmesg | grep cpu.cpu
I have to add this to avoid forum reject “”“aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa”
This change Radxa did recently is just cosmetics. Clueless users will be happy that cpufreq scaling now lists the 2400 MHz OPP while an MCU inside the SoC still rejects higher clockspeeds. All that changes is the difference between reported and real clockspeeds 
Before:
Cpufreq OPP: 2256 Measured: 2250 (2250.953/2250.855/2250.806)
After:
Cpufreq OPP: 2400 Measured: 2250 (2250.512/2250.316/2249.826) (-6.2%)
To get higher clockspeeds higher supply voltage is needed…