I disagree about encoding your songs into 128kbs, this is fine on an iPod but I think that the sound quality loss is clear when listening to music through reasonable speakers. I would also recommend Sonos as when coupled with a NAS it is a superb system.
This is not the case. There are many self-professed audiophiles who claim this but there is not a single scientific, double-blinded study that substantiates this oft-repeated claim.
The problem lies in the fact that CD audio isn't much fidelity to begin with, and it isn't difficult to produce acoustic transparency with a codec that not only utilizes linear prediction compression methodologies but also uses perceptual encoding descended from techniques utilized in Dolby SR-D/AC-3 perceptual encoding schema.
There isn't a tremendous data requirement to produce the same kind of fidelity that is achieved by the severely limited 16-bit Linear PCM format of CD audio. By contrast, 24-bit Linear PCM has an ungodly degree of fidelity that is very difficult if not impossible to reproduce with AAC at various bitrates. Even FLAC doesn't hold its weight if the source material is 24-bit Linear PCM.
The difference? 16-bit Linear PCM has a paltry 65,536 possible amplitude values per sample whereas 24-bit LPCM has 16.7 MILLION possible amplitude values per sample. Even SACD is far behind 24-bit LPCM in terms of dynamic range, noise floor, and frequency response. SACD boasts superior fidelity but has not only an elevated noise floor which negatively impacts dynamic range, but also allows roughly 2.7 million possible amplitude values per quantization interval. 24-bit LPCM is roughly eight times the resolution of SACD... THERE you can argue a very perceptible difference. But between 128 Kbps AAC, using dramatic data reduction through perceptual encoding, and 16-bit (1411 Kbps, recording every bit of junk including the junk you can't perceive) Linear PCM? Forget it. AES can't tell the difference and their ears and equipment are better than yours and mine.
Trust me, CD audio is absolutely nothing fantastic... and 128 Kbps AAC, as stated by the Audio Engineering Society, is acoustically transparent relative to this format... that is, no study has been able to show that any statistically relevant number of individuals can discern the difference.
Audiophiles also claim that vinyl possesses a better response than digital formats, but they seem to ignore factors such as groove width, the limit of which severely hinders dynamic range, and groove degradation upon repeated playback, which kills waveform resolution and thus frequency response and amplitude dynamics. But few people on message boards seem to possess the courage to challenge their unsubstantiated dogma as if this is somehow an off-limits topic akin to personal religious beliefs. Sound reproduction is a science, not a religion, and has measurable factors. It is not a matter of opinion.
I will say it unequivocally and repeatedly... If you think you can tell the difference, you're imagining it... because a perceptible difference does not exist. There is a difference between data and information, and as formats such as ADPCM, home theater DTS (ATX-100) and Dolby Digital AC-3 clearly demonstrate, there are many ways to reduce the amount of data required in order to
reconstruct the same analog information to a degree of accuracy indistinguishable from the original by human ears.
If you want to claim otherwise, the only thing that will prove it is a double-blinded study up to par with scientific standards... and not online ABX testing which, I might point out, surveys which format people think "sounds better" which is itself a fuzzy and meaningless metric when we are trying to evaluate actual acoustic transparency of one format vs. another... None of these ABX tests aim to verify how consistently people can distinguish PCM from AAC, nor do they do it according to scientific standards (true random sample, double-blinded, etc.). It requires demonstrating degree of accuracy greater than by chance alone (the placebo effect) in correctly identifying not which format sounds better, but which format is which... and under controlled, double-blind conditions with identical equipment in each trial.
There's no debate here unless the evidence to the contrary is on the table. Personal anecdotal claims do not qualify as scientific evidence... for indeed if they did, astrology would be published in science textbooks right next to pastafarianism.