Quote:
Originally Posted by jayfrigo  Bob's point was referring to something that has already been mp3, being turned into mp3 again. Perceptual coders of this ilk, including things like AC-3 and ATRAC, are designed to throw out data for parts of the signal that will be masked and therefore won't be missed by the casual listener in the typical consumer environment. However, if you take the resulting uncompressed PCM stream and data-compress it again, the critical decisions the original coder made become painfully obvious.
Double-encoded mp3s sound bad as the fragile perceptual coding falls apart upon being presented previously coded material. Dolby E was made in part to address this problem for the digital television broadcast chain since multiple encode/decode passes may be unavoidable during the many links in the chain from source to local affiliates. |
Thanks for that interesting info jayfrigo thumbsup, I haven't really dealt with that reality which I am sure has been documented many times over. I've also haven't paid too much attention to all the subtleties of mp3 sound delivery. I am assuming we are talking about a
lossy compression aspect ratio of 4:1 or better. Although I receive several mp3 samples to master every week
, what I return to the customer clearly demonstrates the advantages of mastering even a mp3 file. I don't accept mp3 at or less than 128Kbps. I am adamant in accepting mp3 as source material period tutt. It is only the last recourse. Having said all that, what I meant on my previous post, I was referring to (or adding to the discussion) the practice of converting from 24 bit to 16 bit prior to making a mp3 copy ----
Regards,