The first time I used a computer for music was in the early 1980s. This was long before most computers had any sort of built-in sound generator, besides a single-tone feeping speaker. I used the trick invented by geeks several decades earlier: set an AM radio next to the computer, write a looping program that sleeps a certain time between iterations, and listen to the somewhat-musical radio interference emitted by the CPU and picked up by the radio.
I didn’t get very far with the AM radio trick, it being too difficult to program, and not all that interesting, sonically.
Later in the 1980s, I used my Mac Plus to generate drum machine tracks for some songs recorded by my friend Paul & I. This was wonderful, in a lo-fi sort of way.
Over the years, as digital recording became The Next Big Thing, I would occasionally experiment with new audio converters, plugins, editors, and the like. However, I nearly always found the tools unintuitive and complex, requiring too many mouse clicks for too little return (in comparison to the beautiful simplicity of pressing a “record” button). Simulated dials and buttons just made for a messy and fussy interface, rather than somehow bringing back the intuition of a traditional studio.
And the more I exploited the plugins and other software gizmos, the less I felt like I was in a studio, and the more I felt like I was in the computer. To be clear: there’s nothing wrong with being in the computer; as a programmer, I completely enjoy the feeling of “playing” the computer. However, it seems, to me, incompatible with creating the kind of music I like to create.
(I will admit that MIDI, while a crappy protocol itself, is a wonderful idea, and very powerful in a studio. Being able to remotely control processing, including mixdown, has real potential. It’s a cool way to use the benefits of digital even with an analog signal path.)
I also felt the recorded sound quality was far less than I had been used to in the old analog world. Granted, most of these experiments were on cheap hardware. Better preamps help, as do better analog-to-digital converters, and that’s exactly what’s being advertised these days. But that’s part of the current problem with digital: cheap hardware is what most people buy, and it’s easy possible to hear the lack of depth in many contemporary recordings. 4-track analog cassette tape sounded better than some of these digital tracks.
Finally, the largest problem I’ve had with digital recording is the latency. This is a huge issue, and one that is obvious if you read the marketing materials of any of the companies producing audio interfaces. It may be fine for a MIDI-based keyboardist, looper, or primarily in-the-box sound creator, but as someone who primarily overdubs real/live instruments, the delays I encounter while recording are completely maddening. Again, this could be somewhat improved by better, more expensive hardware — eg, Firewire or PCI interfaces instead of USB. However, it seems impossible on regular Mac/PC architectures to reduce the latency to a level that is totally inaudible, especially if any sort of digital processing (compression, reverb, etc.) is being performed. I have always found it interesting that most folks selling digital audio tools are lost without an answer when I really dig into the latency problems.
The usual “fix” is to turn off any processing while recording, reducing the latency to a minimum. However, this goes against my own recording methods of using the studio as an instrument. By delaying any sort of creative texturing until mixdown, I feel like an electric guitarist forced to play without an amp, being told that I can always add that Marshall stack later.