Sound quality is in the ear of the beholder
As someone who worked on the development of the reference psychoacoustic model for AAC circa 1998-2000, I must say that the complainant to the Nokia ad was somewhat naive in their assessment that that AAC cannot be the same listening quality as a CD.
Yes, the transported bit-rates are starkly different. This is in fact the objective!
My point is this: during the circa 2000 MPEG listening tests (which were conducted in the manner described in the ITU-R BS1116 standard "Methods for the subjective assessment of small impairments in audio systems including multichannel sound systems"), we found that the AAC low-complexity profile appeared transparent to the 90% of the 100+ listeners @96kb/s joint stereo. This is using the MPEG critical audio items from the SQAM library.
Therefore, anybody who claims to hear the difference between a CD and the equivalent AAC stream @128kb/s is a big fat liar.
Nokia may want to consider referring to "listening quality" in their adds in future.