Not being a 'computer guy' myself, I must admit a lot of the above is Greek to me. But when I got my Alesis MasterLink CD-R recorder that has a built-in hard drive, I was curious to hear how the sound compared. This unit, for those who don't know it, is widely considered to be of professional mastering quality, and is of course a dedicated component designed just for audio. (Although it's capable of recording and playing back at higher than CD-resolution, I'm only discussing uncompressed 16/44 here.)
Well, using my external Theta DAC, the MasterLink does give superior sound playing from its hard drive, ripped at 4X speed, than it does playing the same CD live in real time. But, both options still can't compare to playing the same CD live on my Theta Pearl transport (not that this method sounds like there's no room for improvement to me, but it clearly delivers significantly more information to the DAC than the MasterLink can). I'm not talking about preference here -- on some disks, I actually preferred listening to the lower-fidelity rendition provided by the MasterLink, as it could occasionally be easier on the ears -- but rather things like spatial delineation and ambience, transient attack and decay, fine microdynamic and harmonic detail, stuff that represents true transparency which is simply being attenuated via the MasterLink used as a transport/server vs. the Pearl (which employs the Pioneer Stable Platter drive mechanism).
These results would seem to indicate that merely using a hard drive to store and retrieve the bits is not a panacea all by itself as far as sonics go. (FWIW, all signals fed to the DAC were jitter-reduced through a Monarchy DIP 24/96, all digital separates were fed balanced, isolated, and voltage/waveform-corrected AC, identical powercords and digital IC's were used, and one digital source was turned off while the other one was in use -- yes, the sound is better that way.)