Digital Audio Lossless/Lossy Codec Demonstration
The topic of CDs versus MP3s is indeed an old topic that started around when MP3s first started gaining popularity in the mid-to-late 90s. The debate was probably at its hottest point around when the Napster controversy started up in the year 2000. However, anyone who remembers that controversy will know that it was about copyright infringment, not quality loss. What I intend to do with this article is give you audible examples of the quality loss inherent in the codec compression process.
What I’ve done here is I’ve taken a lossless audio file (WAV, specifically), put it into my digital audio workstation (Pro Tools 8), and I’ve meticulously lined up the waveforms of some lossy, compressed versions (MP3 and AAC) of the same audio file so that they’re exactly accurate with each other. Let me explain the last part of that a little better: I zoomed in to the waveforms down to the level of individual samples and lined them up perfectly, or at least as perfectly as the human eye, ear and the Pro Tools digital audio workstation technology can manage. When I say “samples,” I’m referring to the same samples as the Sample Rate of CDs (which is 44.1 kHz). For anyone who doesn’t yet know, the Sample Rate refers to how many times in a single second that the audio source is sampled. You can think of it like many, many snapshots taken in a row.
Anyway, when you line up audio files perfectly and play them together, what you’ll hear is an increase in volume due to what’s known as constructive interference. In the case of the inverse, destructive interference, you would hear a decrease or a total cancellation of volume due to the waveforms being precisely opposite. The topic of constructive/destructive interference is indeed an interesting one. Anyone who’s taken a Physics class in school has probably covered it at some point, but I won’t delve into it too deep in this article.
Moving on, once I lined up the audio files (one lossless, one lossy), I inverted one of them so that when the two files were played together, all that was audible was the audio data that was thrown away during the codec compression process. I’ve often explained it as being completely convinced that there’s jelly on your peanut butter & jelly sandwich, but without any jelly actually being there. Also, I should mention that I encoded these lossy files through iTunes. The MP3 is 320kbps CBR, the highest quality MP3 available, and the AAC (m4a) is “High Quality” 128kbps, the iTunes standard quality (which I consider to be cassette quality at best). For your convenience, I’ve also included links to the files in case your browser doesn’t allow flash or isn’t up-to-date. Anyway, without any further ado, enjoy the demonstration!