Synthesis as Fashion

"The producers didn't understand the capabilities of the instrument—and they still don't. It takes imagination to think of a sound no one has ever heard before." -Walter Sear

Again, found on Synthtopia:



These five pieces are derivative. I don't think this point is arguable. This is what happens when genres get overly specific. Any variation outside of the formula ejects the piece from classification.

To the people that say "I hate trance. It all sucks.", I say, sticking to ANY formula sucks. The more narrowly-defined a genre is, the less room for differentiation exists. If I can't hear the artist behind the formula, what is the point? Money, I guess.

The winds of fashion will soon sweep all this forgettable music away. Musical genres are a form of fashion, but synthesis technology itself is subject to fashion.

This is where it started.

"Moog was himself a witness to the power of his bass sound when he was invited to bring his synthesizer to a New York studio session where Simon and Garfunkel were recording their album, Bookends (1968). Mood set up the bass sound himself for the track "Save the Life of a Child." which opens with the sound: "One sound I remember distinctly was a plucked string, like a bass sound. Then it would slide down– it was something you could not do on an acoustic bass or an electric bass… a couple of session musicians came through. One guy was playing a bass and he stops and he listens, and listens. He turned white as a sheet." - Analog Days, Trevor Pinch

So then, someone with a Moog could find a lot of studio work thereafter. But the subtractive synthesis sound becomes dated and people wanted something new. So, if you could program FM, and knew your way around a DX7, you could find some studio work. Then sampling became popular. If you had a Fairlight or an Emulator, you could find some studio work. This process repeats itself over and over to this day. In some cases, the genres are actually defined by the technology used.

When I say the promise of synthesis has largely been squandered, this is one example. Instead of exploring potential, we spend a majority of our efforts reproducing the success of others. This comes with a danger, though. Fashion inherently dates itself and the highs are transient.

If your music is dominated by technology or genre, be prepared for the inevitable crash.

Laver's Law
Anything past the Dowdy point, the creator becomes irrelevant and the work in question is simply an example, whereas innovators, people ahead of their time are remembered.

Indecent - 10 years before its time
Shameless - 5 years before its time
Daring - 1 year before its time
Smart - current
Dowdy - 1 year after its time
Hideous - 10 years after its time
Ridiculous - 20 years after its time
Amusing - 30 years after its time
Quaint - 50 years after its time
Charming - 70 years after its time
Romantic - 100 years after its time
Beautiful - 150 years after its time 

Abstraction and Interpretation

I'm sure you've heard about the loudness wars. Synthtopia recently posted a video on the mastering engineer's perspective

As soon as the music leaves the mastering studio, the experience rapidly degrades. The signal to noise ratio of an iPod/iPhone is around 80dB and they're the predominant medium today.

Let's imagine a future world where this isn't the case, where routine everyday music quality is a fully-immersive experience and the entire dynamic and frequency range can be appreciated.

We're ultimately limited by the thermal noise of a electrical component, the sound of the atoms themselves moving around at around at room temperature: -127.5dB. A 24-bit word at 6dB per bit comes out to accommodating 144dB range. In theory, a 24-bit audio file can encompass the full dynamic range the physical world has to offer. Most audio interfaces come up far shorter, but something with a 120dB range would be considered excellent.

How we master music today, compressing the total dynamic range into a few bits, is going to sound comical and flat. This brings me to a side point about the historical significance of the 20th century recording industry and usage rights. This whole business model is a minor blip in the history of music. Since our lifetimes happen to fall during this blip, we are conditioned to believe that this is The Way, but it isn't. Not in the grand historical view. Rights encumbered media will rightly give way to other, more useful models such as Creative Commons. Culture is resampling. Yes, multitrack audio can be remixed and remastered, but most will not and our ancestors will see this period of music as both rights encumbered (due to copyright creep) and low resolution.

Where I'm going with this is the issue of abstraction and interpretation.



I've said in the past, the role of a composer is to get the sound out of their head into the head(s) of the audience. The composer could perform a piece directly, but more complicated orchestrations involved ensembles of musicians. In this case, an abstraction of the sound is required in the form of a written language we know as musical notation.

This language is very low bit rate. The data compresses to a very compact form. This requires interpretation on the part of the musician, and this itself is an art. The composed piece can be performed various ways, but the heart of the original piece beats within.

In the 1940s, composers began to be able to manipulate sound directly thanks to the invention of magnetic tape. The audio became the final product. Audio quality has since improved dramatically, and many examples of early musique concrète, sound old, which may or may not be an issue, depending on circumstance.

We hear Beethoven's music today, not because it was recorded when he was alive, but because we are able to interpret his musical abstraction and render it using the latest audio technology. This reinterpretation can happen over and over and re-rendered with the latest technology.

The dichotomy of written music and recorded music is expressed in our copyright laws. We have a SR form for sound recording copyright and PA for the written work. In the US, radio stations pay performance rights organizations which distribute money to publishers and songwriters for use, but not the artists who recorded the work. Satellite (and internet) radio stations play by a different set of rules where songwriters/publishers and artists are both paid.

I sometimes like to think of a hypothetical composer who wishes to control all aspects of a sound, unencumbered by the various limitations imposed by traditional instruments. Please don't misunderstand, I'm all for musicians, interpretation, ensembles, etc… but this process represents a relatively huge unexplored area. Sometimes trying to express a viewpoint is made more difficult by an urge to try to adopt an extremist standpoint and defend it from all edge cases. That is absurd. I just want to take you to the edge and have you think about the implications.

If you were such a composer, why wouldn't you want to work with a synthesizer? Why wouldn't you want to free yourself of every limitation of the past? I'm not saying a synthesizer answers all problems, nor am I saying a synthesizer is capable of fulfilling the "any sound imaginable" promise that was thrown around at the time of its introduction. However, I do it feel it is a powerful tool and its promise has largely been squandered. 

I think the proper role of the synthesizer is to do things that can only be done on a synthesizer. This is one reason why I like working with Risset rhythms right now. It is something the brain can understand, but would be difficult or impossible for musicians to pull of convincingly. Therefore, it is something that has to be rendered synthetically.

Successfully leveraging the synthesizer as tool for the expression of music requires a perfect storm of abilities. You have to be a composer, know how to structure a composition, form, harmony and melody. You have to be an orchestrator; an art as difficult as composition, but appreciated to a lesser extent, and complicated by the amorphous and exceptionally wide gamut of possibility offered by the synthesizer. You have to be a musician; understand phrasing, dynamics and delivery. You have to be a synthesist, understand acoustics, recording technology, electronics, principles of synthesis. You also have to have a personality that can put all this together and have a drive to get something done. Oh, and you have to have access to the tools to make this all happen. 

I'm not saying I'm successful at any of those things, but I have the desire to try. 

I think people, in general, are fairly myopic, thinking only about what happens within their life. It is interesting to watch an organization like Harvard, with its thirst for real estate, and the related planning that transcends a human lifetime, a very real version of monopoly.

Here is the thing. If you're working in the field of rendered audio, there isn't an abstraction of your music. It will never be any better sounding than it already is. There is no way to (posthumously) re-render your work to be appreciated at a higher resolution.*

…unless you're working exclusively in an environment like csound. Here we have another example of the duality of music: a score of a performance and an orchestra definition that plays it. Assuming your piece doesn't use any recorded audio and is entirely synthetic, it has unlimited resolution.

Yeah, I think about this because we leave traces of our existence like never before. Before the internet and going digital, what we left behind was limited to our genes, written works like diaries and photos and various physical belongings. Now nothing really goes away and everything is indexed and searchable. I'm not saying what I'm doing will be of interest to anyone of the future, but if I were born in 2110, I'd probably at some point look up the history of my great-grandparents. (Hi, there.)

The process of making music varies from individual to individual. Some people can work artistically in csound, and many are building better tools to make it more and more approachable. I've been flailing away at synthesizers for 25 years now and have a good handle on what sort of process works for me musically, and csound never sang to me. Likewise, my process of using volta and a modular won't work for someone else.

Much of the effort we put into our music has a lot to do with the art of rendering the audio. Not all synthetic works can be disassembled to musical notation. I'm not assigning a value judgement either way. Some of my pieces could and some could not, but the ability to abstract and reinterpret the music could be considered a form of future proofing.

related posts
You are here
Transcribing the head
Reality check

* Could recorded audio be re-rendered at higher resolutions? Yeah, arguably you could see a process like this be developed, but it would ultimately be the same as running a stereo mix through a surround encoder for multichannel audio (surround repurposing). It isn't the same thing as real surround and is universally reviled. Another example is up-sampling DVD to HD video.

Creative Commons image by Leonrw

Radians Earphone Hack

My first pair of earphones were a set of Etymotic ER-4s. At the time (mid 90s), ear phones were a curiosity, and Etymotic saw an opportunity to extend their hearing aid expertise to a wider market: people that can hear.

I found the Etymotics, although highly regarded, to be rather clinical and unflattering. Lots of high end detail, in fact, to this day, I still occasionally use the ER-4s to check for problems, but absolutely no low end.

I reported this to Etymotic, and they instructed me to try all the ear pieces to find the ones most compatible with my ear. While the various ear pieces changed the frequency response, none of them produced any low end whatsoever. I reported this to Etymotic and they suggested I get custom ear molds taken by an audiologist, and then I'd get low end.

So, I found a very friendly local audiologist who made house calls because virtually all of his clients are elderly and immobile. He seemed delighted at the prospect of shoving goop into the ears of someone who was not an octogenarian.

After a few weeks, I received my custom ear molds. I also had some Etymotic ear plugs made, with removable filters for various levels of attenuation. However, the ER-4 ear molds did not magically improve the low end of my ER-4s. The ear plugs were a different story altogether, allowing me to finally enjoy live concerts. Every other ear plug I tried produced a drastically uneven frequency response. The Etymotic ear plugs sounded great, just softer. I even started wearing them at trade shows which lowered my overall anxiety and allowed me to understand conversation better in the noisy environment.

Over the years, I continued to chase the earphone dragon, with usually disappointing results. My most recent set are a pair of Westone 3s, which I feel should be 'mind blowing' but are simply 'acceptable' I use the Westones for recording, live monitoring, traveling, occasional mix checking and any situation where I need to elevate myself out of the ambient noise floor, or need to dig down deep in the dynamic range.

Isolation ear phones are an ancestor of future augmented reality, as they are designed to minimize the real world in favor of a virtual aural reality. This is one reason why I like the idea of composing for listeners of the future, but that is a separate topic of discussion.

Ear phones aren't always comfortable. If they're to work properly, they need to seal the ear canal. I find them fatiguing after an hour or so. Etymotic was right about one thing, various ear pieces can change the frequency response. In the case of the Westone 3s, dramatically so. My favorites by far are the olive foamies. The problem with olive foamies is they're like any kind of ear plug - made of a material that you can't wash, so they have a limited lifespan. Silicone can be washed.

Last week, matrixsynth alerted me to the Radians two part silicone ear goop hack. You push this stuff in your ear, then jam your earphones in. My first reaction was the same as Chris Randalls, (why not do it right, and go to an audiologist?) Well, they're about 1/10 the cost of a set of ear molds (not including the appointment itself), and you can have them in 10 minutes instead of several weeks. They're about the same price as a set of olive foamies and these are washable, so, why not? Plus, they say you can get two sets of plugs out of one unit.





The Radians come in five colors. I chose the tan which could best be described as "generic white American' color. There are others including 'festive dildo', 'japanese schoolgirl' and 'please kick me in the nuts'

I followed the instructions based on the video and I have one thing to add: don't take your time. Once this stuff is mixed, you have to get it in your ear and jam the earphone in quickly. My first batch was a failure as I was able to get one earphone in, but not the other.

The second batch I acted much more quickly and was able to produce a working set of ear molds. You can hear this stuff cure in your ear. Pop. Zaaaaap.

The end result is something that sounds as good as olive foamies, is washable and more comfortable. I'll cautiously give this process a thumbs up.

Reality check

Attention synthesizer nerds. Some (not all) of the well-regarded synthesizer pioneers were not very good musicians. They were the first to express a pedestrian musical idea using a new technology that made it sound fresh at the time. There was a huge barrier of entry in those days, as the associated synthesizer and recording technology cost tens of thousands of dollars. Today, that same technology is available to anyone with a computer. Technology can lift a musical cliche, but only for a limited time. Technology ages ungracefully, then what is left to admire? It is like picking a beautiful apple, only to find it is hollow after biting into it.

A lot of software technology created today is based on automating a laborious process that was made popular by a pioneer. Things that were groundbreaking 10 years ago are common and easy today.

Ask yourself, if [synth pioneer x] were an unknown, and put out an album today, would it get any further than a similar effort from one of the many well-intentioned amateurs on the internet? Be honest.

If you're the first to unearth a new technique, you might get lucky, but your music must be based on more than the application of technology if it is going to stand the test of time. Emulating the output of a technological pioneer whose underlying musical ideas were marginal to begin will get you nowhere.

Do something new. Cultivate greatness.



Image by syntopia Some rights reserved

Learning a new skill

There are two phases to learning a new skill involving technology.

The first phase is driven from the pure joy of using the tool. This is aimless exploration, finding out what is possible, exploring technique.

The second phase is applying this knowledge to the expression of an idea.

Sometimes I see people that want to jump directly to step two. They have an idea they want to express, but don't know how to go about realizing it. They get the tools people tell them they need, but get frustrated quickly, because everything is complicated and getting in the way of what they want to achieve.

Sometimes I see people stuck in step one. They love their tools, but never make the transition to putting the tool in the service of an idea.

You have to have both phases, in the right order.


image by jef safi Some rights reserved

Where it rose, or whither it rushes

:beep boop beep:

100% modular.

One comment I often get from people is "I admire your dedication." Perhaps this is damning with faint praise in some cases, but in others, they're serious. Dedication? That implies choice. I wish I could quit music. Life would be a lot simpler.

Last night, the urge struck me to turn off the reverbs. There was a reverb on a send and a reverb as an insert on a track. Killed both. Eureka. Aside from the summing of tracks, no DSP was used. No reverbs, delays, choruses, peak limiters or compressors. Not even an EQ. If anything, this enables you to appreciate and discern the synthesis a lot better. Reverb can hide a lot of defects. You can bet I'm going to revisit this issue in the final pass though all of 'A Funneled Stone' tracks.

where it rose, or whither it rushes by stretta

STS-124 Launch

'past of plank and nail' married with video of a solid rocket booster on STS-124.

Super SATA Cables on Sale Soon

Malcolm Steward, a technology 'journalist' has 'tested' some new SATA cables that he reports "dramatically improved audio quality."

Wait... WHAT?

He describes the product using tired terms found in detestable 'audiophile' magazines like 'detail', 'soundstage', and ‘air’, in fact, if I had a audiophile cliche bingo card, I would have definitely won with this gem:

My only guess is that the Super SATAs reject interference significantly better than the standard cables and in so doing lower the noise floor revealing greater low-level musical detail and presentational improvements in the soundstage and the ‘air’ around instruments.

The entire thing was so hilarious, I was pretty convinced it was satire. Except, if it was satire, why are comments disabled? See, that would be the fun part.

If you're a real journalist, Malcolm, do this test:

Transfer an audio file using normal SATA cables.
Transfer an audio file using these HI DEF audio cables.
Invert the phase of one.
Sum.

What is the difference, if anything? Anyway, he doesn't mention the manufacturer or provide a purchase URL, so we can't laugh about the absurd price these silly cables will obviously command.



(thanks to kernelslacker for the tip)

read crandall's more eloquent take

swapping projects

I'm the father of two kids: a four year old and a seven year old. As such, the time I have available to work on music is usually when they're asleep. I think most people would agree that using a modular isn't exactly the speediest route to success. Now combine these factors and imagine getting anything done at all.

However, for the last four weeks, they've been in Malaysia with my wife visiting family. I've been at home, trying to keep the plants from dying and apologizing more or less continuously to the cat for the lack of people in the house. This greatly expanded the amount of time I was able to work on music and I devoured it. Two weekends were consumed by short trips to visit friends and family, but the other two weekends (and all weekday evenings), I worked on music.

I got a lot done. Downstairs, at the piano, I worked on 'could have the skies', my third piano ambient project. Upstairs, at the modular, I worked on 'a funneled stone'. This meant I could start a take upstairs, then run downstairs and work on the piano while the modular was recording. It was, like, quadruple productive.

The physical location of the piano and modular dictated the separation of the two spaces, but I could keep the current project up on each computer, and simply sit down to work. I noticed a strange thing, edit decisions were incredibly easy. If I swapped between projects, it became quite clear what to remove. "Oh, that isn't working. Delete." This also accelerated the production process, and the enforced swapping technique was so useful, I've noted it for future use. At the intersection of technique and technology, I'm aware of where my threshold is and I know if I had to close one project and open another, that would preclude me from adopting the technique on a single computer. At least on a regular basis. This also assumes you routinely work on multiple pieces of music at various stages of development at the same time.

If the working process of 'could have the skies' and 'a funneled stone' became entwined during this time period, I began to consider various ways to literally intertwine them as well. I've always liked the idea of spreading puzzles across multiple releases.

While I haven't finished either project, I can see the end of each. I intend to release a full length preview of a new funneled stone piece at the end of this week.

Enter Calico

clearnotice
electronica
entercalico

Today clear notice releases its debut offering - a compilation of amazing electronica expertly curated by Kieron James and mastered by pure tone mastering. It's like discovering a new album by your favorite electronica artist, and every track is a knock out. If you read my blog, you need to buy this album.

iTunes
bandcamp
amazon

weekend

If you are kind,
people may accuse you of selfish, ulterior motives;
be kind anyway.

If you are successful,
you will win some false friends and some true enemies;
succeed anyway.

If you are honest and frank,
people may cheat you;
be honest and frank anyway.

What you spend years building,
someone could destroy overnight;
build anyway.

- Mother Teresa

Junko Mizuno

You should be aware of kawaii noir artist Junko Mizuno. That is all.



wikipedia
fan tumblr

Hacking at the root

Lawrence Lessig nails it again. This one recorded July 29th 2010, in Boston.

Modulemodule

My friend Dave Jones clued me into these interesting designs from eardrill for the Buchla system. This is a really fabulous idea for the format, which could use a way of inserting smaller functions without consuming an entire module panel. Creative! Also, check out the periodicity.

Sample Libraries Now on Soundcloud

I've partnered with SoundCloud to deliver the complete creative commons 120bpm modular loop library and the Total Harmonic Distortion library. The neat thing about this is you don't have to download gigabytes of samples if you only use one or two. You can preview the samples directly on soundcloud, then download what you need.