Ok, spent a bit of time figuring out how to take a frequency response ... not sure I've got it 100% but I've measured it well enough.

So, Which M3 is not like the others?
[Linked Image from i.postimg.cc]

All these speakers are M3v4's. The green line is my 2016 M3. The others are 2020 M3s.
If you guess the M3 in red was the one in my original write-up (#438551) then you guessed right. Leave it to Murphy that I grab the only dud M3 in the bunch (but then again I might not have found it otherwise).

Now before you start reading too much into my measured reponses there are a few things you need to know:
It was taken with pink noise, in a fairly lively room and using my iPads internal mic ... so it's far from a perfect measurement (i.e. don't compare it the axioms anechoic response). What it does do, is enable us to compare similar speakers relative to each other ... and we can clearly see that red's mids, upper mids, and the lower highs are down quite a bit. This is where vocals and a lot of instruments reside. In any case depending on where you place your listening viewpoint you could look at it as recessed mids & highs or bloated bass.

I'm thinking (guessing) that the drop off at the high end may be related the pink noise energy dropping below the limits of what the system can evaluate (starting around 7khz). That said we can still see that the 2016 M3 has a slight high-end bump. I'll have to test the other 2016 M3 to say for sure but when I was comparing these the 2016 M3 always sounded slightly better to me ... I'm wondering if this was why.