I've been reading a lot about the Squeezebox2 around here and now 2 companies are modifying them here on AC...
Can someone comment as to why streaming audio would be better or worse than a hard drive in a PC being used as a transport?
Quite often people talk about getting a music server in another room... and then pulling the data off the network... and/or using a wireless router to send the music to the other room. In this case, sending audio bits across long wires, through the air, or whatever... ...
Using an SB has several advantages over a PC + soundcard, including:
1) The SB can be located far from the PC and any noise it generates.
2) The SlimServer software which serves audio files to the SB can be remote controlled. I use TelCanto running on a wireless PDA, which allows me to browse and play files and playlists from my audio collection through well-organized lists shown on the PDA. Unbelievably convenient and powerful.
3) Multiple SBs can be located throughout the home, all driven independently or simultaneously by one PC running SlimServer.
4) SBs are bit-perfect audio sources. Sound quality of the SB2 D/As is excellent. But SBs feature coax and SP/DIF digital outputs should you wish to use an external D/A or signal processor (as I do).
Audio streamed to an SB via ethernet (wired or wireless) is bit perfect, since TCP/IP includes checksum validation. Jitter is inapplicable in the network data transfer domain. When the audio stream decoded by an SB2 and streamed to it's D/A, a dedicated stable XTAL is used as a timebase which provides excellent stability.
However, an SB2 doesn't sound better than a top-notch PC sound card, such as an EMU0404. Both have superb analog performance, as validated by Rightmark testing. The real advantage of SB is convenience, as outline above. If you use the SP/DIF or coax connectors on either an SB or a PC sound card, signal quality is theoretically perfect, except for esoteric jitter arguments. Jitter significance in audio gear is massively over-hyped, by any reasonable scientific standard.
With respect to using WAVs instead of FLACs, you're doing yourself an injustice. FLACs are bit-identical to the data in the WAV files from which they are generated. But, they consume 1/2 the disk space and provide embedded Vorbis tag information. WAVs don't support tag information at all.
Why would you want to waste disk space and not be able to conveniently and automatically index your audio?