What is it about records?

Digital forms of music have taken over from analog, but there is still a devoted following of records.

This article is about the philosophy of recording, and what people expect from it. Ever since Edison recorded sound, the focus of recording technology was about improving the quality and faithfulness of the sound. That is different from the emotional experience of the sounds, which is why people primarily bought it, no matter what medium it was provided on. The same evolution occurred with CGI for games and movies.

That does not mean that there were not hiccups with each new medium. Magnetic tape became cheaper, so an 8-track version came to be used in cars, where the motional made records impractical. With the development of the Compact Cassette by Sony and Phillips, its small format and relative ruggedness was readily take up by youth on the move, and so took over from records for them, though records might still be used at home. Companies like Nakamichi, with their innovative technology, elevated the medium to true hi-fi, with many recording their records to cassette, so keeping their records pristine.

While digital was already in use for recordings much earlier, the early CD playing equipment used some compromises to keep the cost down. Also, not every recording engineer was familiar with the requirements and limitations of digital recording. This led to some sub-par recordings and playback equipment, marring the experience for some listeners. Just to be clear, records were a sub-par medium, but people were used to its sounds, which were perceived as less annoying than digital clipping.

Nonetheless, CDs pretty quickly took over from records, but the latter still had a small following. Somehow, despite the improvements in digital recording and playback, and the significant noise and wear from playing them, records persisted. While records are analog, the recording process for most became digital, just because it prevented degradation. To keep the process as analog as possible, some recorded direct to master discs. Somehow, digital was seen by some as an anathema to accurate rendering of analog.

The raw musical data for a CD was substantial, and even when compressed formats like MP3 became available, it was not until after computer storage became large and cheap enough that widespread use of such compressed files became popular. With widening use of the internet, downloading these file became popular, if not legal. Once bandwidths became larger and cheaper, the door was opened to streaming. From there, better fidelity codecs and extremely cheap streaming plans transformed listening into the backdrop to everyone's lives.

With cheaper hardware and affordable recording software, almost anyone can record better than CD quality sound, yet records are still around, with its adherent almost appearing to be like the Amish, trying to preserve a way of life most had left behind. Some have missed having the experience of having a physical medium, with its sleeve notes and tactileness, while others want to have analog purity. While the former is a matter of experience, the latter, as a concept at least, has a significant problem. It could be called the elephant in the room, but most don't know about it.

Our senses are only analog at the surface. Everything is converted into nerve impulse that are transformed in our brains into a replica of the experience. Everybody is probably familiar with TVs relying upon our eyes only really receiving three colours, and that the perception of a range of colour is due to the receptors proportioning how much each different wavelength of light stimulates them. It is slightly more complex in that what we perceive is our brain's combining a high-res grey-level receptor grid with a lower-res colour receptor grid.

Similarly, our ears contain about 2000 hairs, each of which is tuned to different frequencies. Like the eyes, each receptor proportions how much they are stimulated according to the wavelength. The receptors generate a series of nerve pulses, increasing in frequency with louder sounds. This is frequency-coded loudness, as opposed to the pulse-code modulation (PCM) used in audio recording. It also adapts so that the longer a sound persists, the less the frequency, which effectively lowers the information level transmitted, yet somehow our brains handles all this transparently to us.

Thus, what we hear is converted into digital at our ears, and the best we can do with our recordings is get good enough to fool our ears and brains into processing our recorded sounds as if they are real. The only analog is between whomever is making the sound and the analog-to-digital converters in the recording equipment, and from the digital-to-analog converters in the playback equipment to our ears. The same occurs for those recording and mixing the audio. What we are doing with our recording processes is sort of an audio version of a Turing test. How successfully fooled are we by it?

Some may claim to not be fooled, and so would prefer imperfectly played analog than digital, but making again that a widespread and affordable reality is now impractical. The digital nature of our hearing does offer a practical limit on how good our current digital processes have to be, since they only have to be enough to fool our ears and brains, but no better.

But from how Apple tried to persuade us that so-called retina resolution was all we needed, though anyone could see that 300dpi laser printouts, which were retina, produced worse rendering of fine details than 600dpi, we cannot be over-confident in our technology, hoping that we can fool everyone into thinking that it is good enough. Many hoped MP3 quality would be good enough, and for many who had spent their days at work hearing slightly mistuned AM radio, it was a substantial improvement. But what we previously tolerated becomes inadequate after we hear what better codecs produce.

However PCM is still based upon absolute levels, meaning that better fidelity requires greater digital space and bandwidths. Perhaps there is a better way to process and store our digital versions that takes into account the frequency pulse adaption going on in our ears, with the view to significantly reducing those two resource-intensive parameters. Early compression like MP3 relied upon dispensing with audible components that seemed to be masked by other parts enough that most would not really notice them missing.

Later versions used better algorithms, or lossless compression from computing, though the latter are still substantially larger. The trick is to properly qualify what is going on in our ears to do the adjusting, and even if we cannot fully replicate it, we will build codecs that better mimic it. The bar is always set by lossless, but until that is far more efficient, we will continue searching for how to better fake it for our ears!

There will probably always be those who prefer analog, but records are still noisy, distorted and fragile, which runs counter to any hi-fi argument for keeping them, so is it just nostalgia for a time when listening to music was an experience from buying to playing it, and now lost in the ubiquitousness of streaming? Modern popular music seems to have become a parade of collaborations and similar backing rhythms that makes it seem like it does not really matter who is singing or playing. It is a gentrification of music that results in a sameness of musical architecture.

It is a change from the 20th century, especially the 1960s, when there was a plethora of individual sounds, particularly from England, all fed by different styles, providing a rich tapestry of sounds to build musical experiences out of. Yet we have come to a musical monoculture, not unlike the music of long-standing cultures that has become an artifact that readily identifies them. Of course, there are those that still experiment, but they have become hard to find in the plethora of output resulting from the affordability of recording.

Playing records, per se, is not going to bring that variety of experience back.