For those curious about why I didn't respond to initial emails about this, there are two reasons:
a) I get a lot of email
but more to the point
b) I get a lot of email about the uCurrent and how to use it etc, and in almost every case it's people having no idea how to use it, their setup is wrong, etc etc. Genuine hardware faults are incredibly rare.
and
c) It takes a lot of time to investigate and actually respond to emails like this, so unless it's a one-liner response I often leave it unread to deal with later. I know this sounds shitty, but I have a finite number of hours in my life. And I'm not going to respond with "I'm working on it" if I'm not actually working on it.
So when I see an email about an offset problem like this, for very genuine reasons I didn't think it could possibly be the hardware, it was most likely yet another user setup problem that will take ages to solve. Then when I actually tested a current batch one, it tested fine, so what was I to do right away?
It's only when I saw the 2nd report on the forum that I really jumped on it and knew something was up and investigated straight away.
This of course turned out to be a very rare and obscure fault that was not immediately obvious. Obvious with hindsight of course.