Vinyl vs. Digital: Which Sounds Better?
Which medium sounds better, vinyl records or digital audio? It’s an ongoing debate that has inspired arguments almost as heated as political ones!
The answer depends as much on your subjective experience as it does on objective measurements, so we’re unlikely to settle the debate once and for all right here. But we’d like to try to shed some light on the issue, and take a look at the cold, hard facts of science so you can come to your own conclusions.
Analog or Digital: What’s the Difference?
First of all, it’s important to understand the technical differences between analog and digital recordings.
Vinyl records, as you likely already know, are analog, meaning they store audio information in an analog manner — specifically, as tiny undulating grooves on the surface of the record. These grooves actually look like the sound waves they represent; in other words, they are analogous to the sound waves, hence the name analog. When you place a stylus (or “needle”) down on that groove, it wiggles with the undulations as the record moves beneath it. Those wiggles are converted to an audio signal, which is then amplified and played on speakers or headphones.
By contrast, digital-audio files store audio — well, digitally. A song or album’s original recorded audio signal enters an analog-to-digital converter (ADC), which, as you can probably guess from its name, converts the analog signal into digital information, a series of 1s and 0s. Without getting too technical, the analog audio waveform is “sampled” many times per second, recording the instantaneous level of the waveform as it rises and falls using several digital bits to encode the level value of each sample.
On CDs, the waveform is sampled 44,100 times per second, and the level of each sample is encoded with 16 bits. This leads to specs you might have seen: a sampling rate of 44.1 kHz and a bit depth of 16 bits, often written 16/44.1 or 44.1/16. So-called “high-resolution” digital audio files use higher sampling rates and bit depths; the most common are sampling rates of 48, 96, and 192 kHz with a bit depth of 24 bits.
When you put a CD in a CD player, a laser reads the digital information, which is sent to a digital-to-analog converter (DAC) that converts it back into an analog audio signal then pumps it out your speakers or headphones. Likewise, when you play a digital-audio file from a computer, smartphone, or other digital device, a DAC converts the digital information to analog, then it’s amplified and played on speakers or headphones.
But which sounds better??? We’re getting there, hang in…
Compress This!
Surely you’ve heard about “compression,” and how compressed digital files can result in a reduction in audio quality. Let’s discuss exactly what that means.
The digital-audio information on a CD or in a high-res music file is a direct representation of the original analog signal; it’s often referred to as being uncompressed. Such formats include WAV (Waveform Audio File Format) and AIFF (Audio Interchange File Format). However, in much of the digital music heard today, some of the data have been removed to reduce the file size and bandwidth required to transmit the file. Such files are referred to as being compressed.
This is accomplished by removing certain frequencies, which you supposedly can’t hear anyway because they are “masked” by other nearby frequencies. Amazingly, up to 90 percent or even more of a file’s information can be removed in this way, and it still sounds more or less—but not exactly—like the original. Because this process removes frequencies that cannot be recovered, it is called lossy compression. Examples of lossy compression systems include MP3 (MPEG-1 Audio Layer III) and AAC (Advanced Audio Coding).
It’s also possible to reduce the file size without irrevocably losing any data, which is called lossless compression, but this can only reduce the file size and bandwidth requirement by about 50 percent. Examples of lossless compression schemes include FLAC (Free Lossless Audio Codec) and ALAC (Apple Lossless Audio Codec).
To compress a digital-audio file, it passes through the first part of a codec (coder/decoder), which encodes the data with the selected form of compression. Then, to play the file, it passes through the second part of a codec (the decoder), which reconstructs an uncompressed version of the data. If the original compression was lossy, the decoder makes its best guess about what the missing data should be; if the compression was lossless, the decoder can reconstruct the original data perfectly.
Once a file has undergone lossy compression, the most important specification is its bitrate—that is, how many bits per second it uses to transmit online or from one device to another. MP3 is a very common form of lossy compression, which can have bit rates from 32 kilobits per second (kbps) up to 320 kbps. The higher the bitrate, the better the sound quality; music shouldn’t be encoded at a bitrate lower than 128 kbps, or it will sound pretty crappy. Lower bitrates are used only for speech.
By comparison, uncompressed CD audio has a bitrate of 1.4 megabits per second (Mbps), which is more than 10 times the MP3 bitrate of 128 kbps. (Notice that MP3 at 128 kbps reduces the bandwidth requirement by a factor of roughly 10, or 90 percent, compared with CD.) Uncompressed high-res music files have bit rates even higher than CDs.
What’s the Objective?
Ever heard an audio snob complain about digital audio sounding “brittle” or “sterile”? Let’s learn about the science that explains why by taking a look at objective, scientific measurements.
The two primary characteristics of any audio system are dynamic range (the difference between the softest and loudest sounds the system can accurately store and reproduce) and frequency response (the range of frequencies the system can accurately store and reproduce at roughly the same level). For reference, the maximum dynamic range that a person with normal hearing can perceive is around 120 to 130 decibels (dB), while the maximum frequency range of normal human hearing extends from 20 Hz to 20 kHz.
Keep in mind that only newborn babies can actually perceive frequencies up to 20 kHz; as we age, the upper limit of our hearing decreases, especially if we attend loud metal concerts! Young adults typically can’t hear anything above 15 or 16 kHz at best, and seniors are often limited to 12 kHz or even lower.
Vinyl records have a typical dynamic range of around 70 dB, depending on the equipment used to record the audio and cut the record. CDs have a typical dynamic range of 90 to 93 dB, though 16-bit digital audio has a theoretical dynamic range of 96 dB. Uncompressed high-res audio files with 24-bit resolution have a theoretical dynamic range of 144 dB, though modern ADCs and DACs max out at around 120 dB or so.
In terms of frequency response, vinyl records can store and reproduce frequencies as low as 20 Hz and as high as 20 kHz. Depending on the cartridge, some vinyl records have achieved frequencies up to 50 kHz, but you need very special playback equipment to reproduce such high frequencies, and no one can hear that high anyway. On the low end, vinyl can’t go much below 20 Hz because of the physical limitations of the cartridge and tonearm as well as a type of noise called rumble, which is endemic to virtually all turntables.
CDs can store and reproduce frequencies well below 20 Hz—down to 0 Hz, in fact, but what’s the point of that? At the upper end, digital audio is limited by the sampling rate. Any digital-audio system can accurately represent frequencies up to half the sampling rate; for example, with a sampling rate of 44.1 kHz, CDs can represent frequencies up to 22.05 kHz. That’s slightly above the theoretical maximum of 20 kHz for human hearing, so it’s more than enough.
Digital audio has one frequency problem that vinyl doesn’t have to worry about. If a CD tries to store frequencies above 22.05 kHz, an audible artifact called aliasing rears its ugly head, adding all sorts of nasty-sounding junk to the sound. So, when analog audio is converted to digital, all frequencies above half the sampling rate must be blocked from being digitized, and the output from a digital-audio player must block any frequencies above that value from being converted to analog. In the early days of CDs, this led many to rightly complain about how “sterile” and “brittle” they sounded compared with vinyl. These days, anti-aliasing technology has improved greatly, so this is not really a problem anymore.
High-res audio files must also beware of aliasing, but they have a higher sampling rate, so they can store higher frequencies. For example, at a sampling rate of 96 kHz, audio frequencies up to 48 kHz are no problem. Of course, that’s way above what humans can hear directly, but there are those who believe that such high frequencies can be perceived somehow, perhaps through the skin, and so contribute to the overall experience.
“Noise” is another objective measurement used to compare vinyl with digital. Aside from rumble, the electronics in turntables have some level of self-noise, and of course, vinyl records often have scratches and dust that cause pops and other added sounds, especially if you’re not super-diligent about keeping your records clean. Digital audio does not suffer from rumble, pops, and other noises endemic to vinyl, but a poorly designed digital-to-analog filter can cause something called quantization noise.
Digital audio that uses lossy compression is subject to a variety of audible problems. For example, at 128 kbps, the lowest acceptable bitrate for music, frequencies above 16 kHz are typically removed. Also, audio events can be “smeared” in time; in particular, short transient events such as a snare-drum hit might be heard before or after it’s supposed to occur. This time smearing can also cause the overall sound to have a softer, smoother character with less clarity and transparency than the original audio. In addition, MP3 is notorious for weakening low frequencies, causing bass lines to lose their punch. Other encoders, such as AAC, do better with the bass. Finally, as lossy compression removes frequencies from the signal, it adds its own noise as well.
Subjective Impressions
Subjective experience, of course, can’t be measured, as it’s different for every person. But there are some widely accepted subjective truths when it comes to analog vs. digital that are grounded in a bit of science.
Those who prefer vinyl often tout its “warmth” compared with digital audio, which is something that actually can be measured: the frequency spectrum of the input signal compared with that of the output signal. Most analog electronics, especially tubes, and vinyl records, alter the spectrum of the sound, slightly increasing certain harmonics. The result is a warmer character that many listeners find very appealing.
On the other hand, uncompressed digital audio is more accurate; it does not alter the spectrum of the original signal. This leads vinyl advocates to say that digital sounds “cold.” Harry Pearson, founding editor of The Absolute Sound, famously said, “LPs are decisively more musical. CDs drain the soul from music. The emotional involvement disappears.”
Few would argue that lossily compressed digital audio sounds better than vinyl; in fact, most would say that it sounds significantly worse. Also, brain-scan studies have shown that MP3 compression robs music of its emotional impact more than uncompressed digital audio, to say nothing of vinyl.
But how does compressed digital audio with minimal loss, or digital audio that’s uncompressed entirely, compare to vinyl?
The big draw of lossy compression is convenience—the ability to easily stream music online and quickly download and store thousands of files on a mobile device.
A large part of the appeal of vinyl comes with its physicality – holding it, looking at the cover artwork, reading the liner notes – and it’s collectability as a souvenir.
On the other hand, many music enthusiasts really appreciate the pristine accuracy, wider dynamic range, and lack of noise that uncompressed digital audio offers. Plus, unlike vinyl records, digital audio files require no care whatsoever, and they don’t wear out after repeated listens.
Vinyl has certainly been gaining steam lately. In September, CNN Business reported that sales of vinyl records have surpassed CD sales in the United States for the first time since the 1980s. Vinyl sales have grown steadily since 2005.
At the same time, we don’t have to tell you about the explosion in streaming over the past few years, with hundreds of millions of listeners choosing to consume their music solely over the internet.
So: where do you land on the debate, especially now that you know the science behind it?
Which do you prefer: vinyl or digital?
*
This post is sponsored by SVS High Performance Speakers, Subwoofers and Accessories, delivering thrilling and immersive home audio experiences and unmatched performance for the price.
