(Cleaning and rewriting 2026).
Synthesis was once a desire to hear new sounds, make new music, go places that hadn’t been heard before. That creative stance started to die with Tomita and is now truly dead when trying to emulate some noise from decades ago.
Synthesis has become terribly OLD FASHIONED.
This pisses me off. Fuck recreating the Blade Runner soundtrack now we’re past the year in which the film was supposed to take place.
Software can now pull a sound apart, make a wavetable, or a spectral image, change every aspect of it, build entirely new sounds from audio atoms – and people are still talking about ladder filters?
We need to clarify what we mean by ‘hardware’ and ‘software’, because quite a lot of ‘hardware’ is a container. A Moog model D is all hardware while a Blofeld is software in a box. Recent work by the Usual Suspects has transformed a whole range of hardware into software – with the same sound.
How do we define ‘better’? We are talking about musical instruments being used for music. Implicit in music is a listener and that something humane is being communicated – awe, hope, fear, something worth listening. What does the sound of a device offer for others? Slavishly copying a sound plucked from some antique recording is not music. Matching the shape of a waveform on an oscilloscope is not music. These are feats for athletic carnivals.
You may be pleased by the feel of the knobs, or the type of wood at the cheeks of the thing, but what does this do for a listener? Maybe the wood inspires you to delicate adjustments, but it’s still not a violin. The way your machine looks, how pretty the lights, the styling of it – it may as well be a Dyson vacuum cleaner or an Apple phone. I’m tempted to judge that only the resulting sound matters, but I have to concede that interface might make reaching that sound more likely.
Exemplars of ideas
I have been buying and reviewing more software because the majority of hardware is cowardly, conservative and dead boring. I’ve sold a ‘legendary’ (and very heavy) old Roland Juno for some good profit. The reason being that if you divided the sound by the amount the damn thing weighed, you’d have no change left over for coffee. Any half decent virtual analogue could make that noise – especially Roland’s own. Can the audience really hear the difference?
The time spent racking, un-racking, cabling around the back of things, assembling A frames etc. is like the days when people would run clothing through a mangle before hanging it on a clothesline. Generally when I visit other people’s hardware studios the damn things are NEVER FINISHED – a great excuse for why no actual music is being produced. In my case I’m now trying to reduce my hardware down to exemplars. Carefully limiting the orchestra (like cholesterol) so that you can actually score music.
Some hardware is interesting. For example, I keep my UltraProteus because of the weird thought process behind its operation but I’ve sold the Yamaha FS1r because as crazy as it is, the sounds it makes aren’t that great. And that’s what matters.
But if you you spend enough time with spectral synthesis and wavetables you will find a world of new sounds. Or take a recording and hack away at it with spectral editing – that’s what I did with my records Donut and Aversion. It’s synthesis, but it’s not hiding in the last century, terrified of stepping outside ‘analogue warmth’.
To strive for new sounds future is the ambition that synthesis is all about.