The reported frequency value for the PWM test is wrong. It is supposed to be 480Hz. If you look at the image of the oscilloscope readings, you will notice the average frequency was used instead of the current frequency.
Even then, if there is one thing I have learned using oscilloscopes, never blindly trust the math readings of an oscilloscope. Confirm it by looking at the actual waveform shown by the oscilloscope.
In the scope images provided, you can see the peaks repeat every 2 subdivisions and on the top left of the scope, 1ms per subdivision is reported. This means each peak is roughly 2ms apart. f = 1/0.002s = 500Hz, but since 2ms is just a rough estimate, actual frequency lines up with the "current frequency" value reported by the scope.
No one will probably read this because the review is months old now but I have seen this review's reported PWM frequency reported wherever else I have looked. I really want Notebookcheck to report correct values instead of rushing things as Notebookcheck is one of the only places that provides the values with oscilloscope screenshots, which I do appreciate.