True true. I remember running sine wave tests through Mcintosh and Yamaha “hi-fi” amps and the stereo meters not matching. I have no experience with studio gear, and am not at liberty to say they are not accurate in such.
That's pretty sad because I've worked on playback software in the past, using portaudio (like audacity does), and there was a way that it could report the latency well enough that we could display the playback cursor over the time-domain audio signal at the right place at the right time, even when using very long playback latency. The only "downside" with long latency was the time between keyboard input and the sound changing (one feature was switching rapidly between different versions of pretty much the same thing).I have not tried them recently but the software VU meters tend to have latency that makes them out of sync with music. They can't anticipate how long it takes audio samples from the player to get to the output of the device. For this reason, while they are pretty to look at, I don't find them as satisfying as real VU meters.
Exactly, if I'm not mistaken, McIntosh has models that only display voltage (even though it says watts) and other models that actually do a power measurement. I don't mind the voltage variety because it's just for looks as mentioned before, and measuring current (to get actual watts) could be more intrusive in the signal path.As for one attached to a power amp in a home stereo...does it matter? It's just for looks.
They are (or should be, if to the ANSI standard) pretty accurate, easily to a needle's width. But, on steady tone only. Their dynamic performance is very poor on programme material as their deliberately long integration time is designed to indicate 'average' levels more consistent with perceived loudness than signal peak level.Aren’t VU meters notoriously inaccurate?
Someone on reddit did this incredibly cool KTB+Atom+VU meters custom build:
https://www.reddit.com/r/headphones/comments/cdb64p