Hi All,
I ran some ad-hoc experiments last night with my new BDP-1 and different audio file formats.
With Redbook (44.1/16) Apple Lossless (ALAC), mpd seems to hover around 50% CPU usage during playback.
With MP3 (320 CBR/LAME), mpd hovers around 30% CPU usage.
In both cases, the CPU usage was pretty constant, although it jumps a little bit at the start of a new track.
With AIFF (big endian), even with 96/24 files, mpd stabilizes at 3% - 4%, also with a small jump when starting a new track. Wow, an order of magnitude lower CPU utilization? I was surprised to see those results, running top during playback.
So my question is this: is there a sonic advantage in minimizing CPU utilization? That is, does high CPU utilization emit enough wide-band noise to impact delivery of PCM?
I suspect this is something that has been debated, and I'm not trying to start anything, but I am sincerely curious what the thinking is on this topic.
Thanks, Ken