"Offtopic", last time...we should use the existing threads for it indeed...
And now, average(64) + fine + aliasing does something : a very fine trace (1px). To All: Do run a self cal.
Instead of using averaging, you could decrease the memory depth to make the trace thinner.
See also Daves video
That is no difference between x10 and x1. No matter how you adjust the probe, the little overshoot won't disappear.
Although I don´t have the issue with my 5074, it sounds like a mismatch problem - we bought a couple of probes cause the originals were mostly "vanished"...
On some scopes there was no problem to adjust them.
Other scopes showed exactly your problem - we couldn´t compensate the overshoots completely, it´s a matter oft the Input capacity.
Only with their original probes everything was fine.
Either the input capacity on your rigol was different (tolerance) or the probes are defective.