Digital music plays by numbers

Recordings: Unlike analog LPs, CDs convert sound into the bits recognized by computers.

November 01, 1999|By Lou Dolinar | Lou Dolinar,NEWSDAY

You hear a lot about digital music these days. So let's start with this: What the heck is it?

When people mention "digital" music, they usually compare it to something called "analog" music. The former means that music has been transformed into the 1's and 0's that computers use to do their thing; analog corresponds more to the real world.

A good example of analog is the old-fashioned vinyl record. For our younger readers, this is one of the large black platters that your parents occasionally listen to when they've had too much to drink.

Mom and Dad's LPs work on the same principle as their grandma and grandpa's: With those early gadgets, performers sang or played into a little horn connected to a vibrating needle. The needle was placed against a plastic-like cylinder, which was rotated at high speed by a mechanical clockwork.

The needle inscribed a continuous groove into the surface of the cylinder. Its depth and width and other characteristics corresponded to the sound being made. To play the sound back, you put the needle in the groove and cranked her up again. The sound -- not very good sound -- came forth from the horn. We've added all sorts of fancy electronics to improve audio quality since then, but the idea is the same.

Compact discs, now more than a decade old, operate on a different principle, recording sound that has been fed into a device called a digital-to-analog converter (DAC). The DAC takes a "sample" of the sound milliseconds apart -- like the individual frames that make up a strip of film.

The samples are then converted into the 1's and 0's of computerese. These bits, as they're called, are then burned into the surface of a CD in a circular track. Although the medium was different, the whole process was based on the way computers record data on their disk drives. To play the sound back, a laser beam reads the 1's and 0's, which go through another DAC that converts them into the analog signal that amplifiers and speakers need.

There's a lot to like about digital CDs. Digital media -- and this pertains to my satellite TV, as well as my CD player -- tend to be far more resistant to physical damage: everything from wear to dust and scratches to stray electromagnetic interference.

On the downside, you have an imperfect "sample" of reality, and a truly golden ear can hear the difference between a digital CD and an analog record. On expensive equipment, the analog record will often sound better.

Most important is the digital-to-analog conversion. Simply put, this can be done badly or well, depending on how much money and hardware you throw at the problem. Cheap CD players and cheap PC sound cards tend to have cheap DACs.

Still, for the past decade or so, the music CD has been dominant, and the medium didn't escape the attention of the computer industry.

Looking for a cheap way to store and distribute data, Microsoft's Bill Gates and other industry leaders realized that the CD format, though not ideal, was cheap and easy to implement on a PC. Instead of designing a new storage medium and building new factories to make players and disks, the computer guys piggybacked on the recording industry's work.

Even so, until recently, CDs for PCs and music didn't have a lot to do with each other. True, you could play a CD on a computer and listen through tiny, tinny speakers. But PCs didn't have enough hard disk space to store the music from CDs -- a single album requires 600 megabytes of data. Moreover, PCs weren't fast enough to manipulate the data with any sophistication.

But you know how engineers are -- give them an overactive computer and one afternoon they'll invent the Internet. In 1988, a band of engineers founded the Motion Picture Experts Group (MPEG for short) to figure out how to shrink the size of digital files for movies.

This was important because both cable television and phone companies were planning to sell movies on demand. To make it possible, the engineers came up schemes called MPEG-1 and MPEG-2 that compressed movies by a factor of 10. And, as a sort of afterthought, they came up with a standard for compressed audio called MPEG Layer-3.

According to the Fraunhofer Institute, which owns the most advanced implementation of MPEG, "The encoder analyzes the spectral components of the audio signal by calculating a filterbank (transform) and applies a psychoacoustic model to estimate the just-noticeable noise level. In its quantization and coding stage, the encoder tries to allocate the available number of data bits in a way to meet both the bit rate and masking requirements."

In English, this means the computer throws out stuff it thinks you can't hear, then takes what's left and scrunches it mathematically. It can throw out more or less data to produce a larger or smaller file. At 10-to-1 compression, it's pretty close to the original CD. If all you want is AM-radio quality, you can shrink the file 24 times. If all you need is telephone quality, you're looking at a 96-to-1 compression ratio.

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.