I wrote about this issue some months ago, but got very busy so couldn’t get around to updating the post until now, long story short I can’t find the original post, so here is an update on my finding regarding THD+N measurements on the RME.
I regularly test my equipment to ensure it is up to spec, on one occasion when I measured the UFX it exhibited a high amount of THD+N, someone at RME suggested this could only be either the test equipment was malfunctioning, or the UFX was faulty. Neither turned out to be the case, here is what I found.
When I measured the UFX I had omitted to realise that I still had the UFX output connected in parallel with an amplifier input, I do this to facilitate acoustical measurements. The result of this was a high amount of THD+N, some experiments were needed in order to understand what was going on.
I disconnected the amp to find that the THD+N figure was now correct, then I plugged in the amp again, but this time I powered it up, this resulted in a lower THD+N figure, but it was still not within spec.
So I got a big pile of gear on the bench, some very old, up to 25 years old.
I did the same experiment on all of these devices to discover they all had the same problem, to different degrees.
This leads to the question, how many manufacturers actually apply this test in the R&D process, it would seem not many.
To put this into perspective, I don’t think many people will get the problem, most studios work with just one thing connected to any given output. However, in the live audio world where I do most of my work it is quite common to have many inputs being driven from the same signal path.
I’m writing to most of the manufacturers whose devices I tested to see if any have an interest in this issue.