Overlap behavior on staff-ranged dynamics

So,
There seems to be something funny (or some alleged Officially Works As Intended™; see below) behavior going on with the dynamics on notes shared between two staves. For instance, first guess what you'd expect to hear from this passage,

and then compare to what actually you actually hear. If it's different from my experience, then there's already a problem (for the second part, that is; feel free to think whatever you like for the first part).

For me, I expect to 4 (four) C4's played with equal velocity. Instead, that reality explodes like the chemical compound it suggests and I hear: a soft note, two loud notes, then no note.

The only explanation I can think of is something like this: the first note (and only the first one on the dynamic because it's forced to look at that spot closely) has the velocities split between the treble and bass clef, so when the bass clef is muted, it losses that velocity (the soft note). Then, for every note after that, playback isn't forced to analyze each part and uses the look-back mechanism to determine note velocity. It merges the treble and bass notes, finds that the bass clef has been muted, and does so for the entire note (ignoring the fact that treble clef isn't silenced). Any note attack on an already sustained note is unaffected.

At least, that's how I explain it to myself so I can sleep at night.

OS: Windows 10 (10.0), Arch.: x86_64, MuseScore version (64-bit): 3.4.2.9788, revision: 148e43f