I am still drowning in numbers as I work my way towards the subject of MP3’s and why they are a mixed bag when it comes to listening to and selling rock music. I feel like the Count from Sesame Street on an algebra binge! These numbers are all about compact discs, which we must examine fully before we can look closely at MP3’s, for it is the CD against which all things digital audio are measured. Last time I only got as far as sample rates. (No, I do not get paid by the word.) I intend to wrap up this section with a brief but technical rant about the compact disc’s resolution, also known as bit depth.

First, let’s clarify something: CD specs include bit depth, MP3’s specify bit rate (as in kilobits per second or kbps). Bit rate refers to the amount of data that can be processed (decoded) per second.* If the sample rate is time (and it is, remember 44.1 kHz sample rate = 44,100 audio snapshots per second), then bit depth is the quality of that time, as in how much detail is captured in each of those little snapshots of audio. Expressed more precisely, bit depth equates to voltage, and voltage equates to amplitude. While amplitude refers to how loud a signal is or can be, eight bit audio at full scale is just as loud as 16 bit audio at full scale. The difference is not in how loud but rather how many different levels of loudness (discrete voltages) can be achieved between no sound and full volume. Since bit depth can vary based on the compression of the MP3 file, and bit rate is more telling with regards to compression, bit depth isn’t generally specified for an MP3. (This will make sense when I abandon numbers altogether to explain how MP3 compression squeezes the life out of songs.)

When CD’s were making their way to market, 16 bits was considered high resolution since most digital media files and their processing rarely exceeded eight bits.** Sixteen bits isn’t considered hi-res today, even though it provides a whopping 65,536 levels of detail (amplitude) inside each and every sample.*** Do we really need that much detail? I mean its bloody rock ‘n’ roll, not rocket science. The answer is, abso-fucking-lutely! That’s because bit depth provides the Dynamic Range and Signal to Noise Ratio (SNR) of your CD audio. Dynamic Range is the difference between the quietest sound and the loudest sound (before clipping). SNR is the ratio between an amplified signal and unwanted noise from the system carrying the signal. Both Dynamic Range and SNR are represented by a single number in decibels. Since each bit provides 6dB of Dynamic Range, CD audio boasts 96dB (16×6) of Dynamic Range and possible SNR. I say possible because preamplification and signal processing like EQ can lower the SNR since they add noise to the signal. By comparison, the human ear has a dynamic range of 120 dB.

In the early days of CD’s the format was renowned for not living up to the promise of pristine audio fidelity and a sound superior to vinyl. This was partly because early CD players used 16 bit, or sometimes even 12 bit DAC’s (digital to analog converters), which lacked the proper filters required to smooth out errors that occurred during the process of turning math into rock ‘n’ roll. Sixteen bit audio into a 16 bit DAC was a “just fit”; there was no headroom, so to speak, when the math left the DAC as music. There was less room for error, but like everything else man-made, errors were bound to happen. Now pretty much all digital audio hardware uses anti-aliasing filters and 20 bit, or 24 bit DAC’s—and this has gone a long way in making 16 bit audio warmer, if not quite warm.

Today nearly all music recorded digitally is done at 24 bit resolution (that’s 16,777,216 possible levels of amplitude!). This has opened up the signal quality in ways that can be felt and heard by most everyone. The conundrum here is that the final product is still 16 bit data for standard CD’s, and sometimes is less for those dirty little downloads that have literally turned the music biz on its ear. This increased resolution means that recording and production has gotten better, and that improvement can be heard on the 16 bit CD.

You may have noticed that some of the first CD’s of albums originally released on vinyl have been re-released as “Remastered” recordings. That remastering included (or sometimes solely consisted of) a redigitizing of the album at 20 or 24 bits, or at 16 bits through 20 or 24 bit hardware. And even though those remastered CD’s and the original are both at 44.1 kHz and 16 bits, it’s clear that the remasters sound better. Sometimes a lot better. Score one for the record companies. What may on the surface look like a marketing gimmick to get you to buy more CD’s actually has real value. And as much as I like blaming record companies for many of the problems we’re dealing with in rock’s digital age, you can’t blame them for the early CD releases sounding bad.**** We were all promised that the digital audio revolution was ready; simply digitize, burn and release. Of course, there are still plenty of other things we can blame record companies for.

*CD’s have a bit depth too; it’s 1411.2 kbps for a standard audio CD. Most MP3’s are compressed at 128 to 196 kbps. Mathematically and sonically speaking, that is a huge reduction.

**By the time CD’s killed off vinyl and cassette tapes as the preferred album format, Nintendo was the video game of choice. The brothers Mario and all their friends and enemies had a resolution of 8 bits.

***65,536 = 2(16) 16 because it’s 16 bits, two because binary numbers, the type of numbers computers use, have only two different values, 0 and 1.

****Okay, technically we can blame the record companies—many early CD’s of albums previously released on vinyl were made from copies of the original master tape. So that was one big generation down from the original. This is why many of the remaster/re-issues state “From the Original Master Recording.”

Like this article? DIGG it and sign up for our RSS feed!