Matt Pearson's visuals for Groove Armada's Red Light show
When my kids ask me, “What did you do in the Visual Art Wars, Dad?” I can tell them, with pride, that I fought on the side of the future. I'll tell them about the environments in which we fought—the festival fields of Europe, the night clubs of Ibiza—and I'll tell them about the forces we fought against—the Anachronists, the Atemporals, the Pareidolics, and, worst of all, the FakeDJs.
It might seem an odd place to fight for the future of Visual Art, but an area had emerged with a particularly pressing need. The decades either side of the millennium had seen a certain breed of, previously anonymous, electronic musicians kicked out of their clubs and thrust onto festival stages. Here, nerds hunched over their laptops were having to compete with “real” live acts. Their shows weren't quite cutting it. No one could really see what they were doing [on stage]. They weren't even prepared to show their screens. There was music coming from somewhere, but it was difficult to tell what was making it. The de facto solution became attempting to take attention away from the men behind the curtain by creating a visual spectacle to distract the audience. Which, ultimately, made them the unwitting pioneers on a new front line of Visual Art.
Will my hypothetical kids believe me when I describe the state of live music visuals in the first decade of the 21st Century? The decade when a live act could be billed as a “full A/V show” and then play pre-recorded videos. They'll laugh nervously, not sure if I'm joking, when I tell them this was the norm, that bands played massive gigs--headlining Glastonbury, closing the Olympics, playing for the Queen--with sets that, at some points, involved showing visuals that had been pre-rendered, sometimes months or years before the moment all were gathered to share. And they won't believe that the audiences just went along with it. That they didn't care. That they didn't walk out.
We shouldn't blame the audiences though. The music lovers of the last decade of the 20th Century and the first decade of the 21st, suffered the confusion of transition. The third era of visual art was just taking off, and the technology was struggling to keep up with public expectation. So it was being widely faked. We cannot blame the gig goers for their impatience. They had seen fictional representations of the exciting future ahead, and confused it with the technological reality. Which was why they were complicit in the sham, why they were happy to experience a video of what a live electronic music gig should feel like, rather than getting to experience the real thing. There were also charlatans abound, making it purposely difficult for them to know what was real and fake. Only their guts knew the difference.
For some audiences this was fine, it was as good as they thought it got. Others had better imaginations and demanded something better, and this was the cause we fought. We fought on the side of the future. And the future would win eventually. Because The Third Era of Visual Art had already begun.
Visuals from Groove Armada's Red Light show. Photo credit: Jim Howells
The Transition To Real-Time Art
The First Era of Visual Art began not long after Homo Habilis discovered tools. Around the same time (give or take a few thousand years), our ancestors discovered rubbing twigs together could make fire, they worked out ways to draw lines in the dirt, rub chalk against rock and, later, mix pigments of plants and get them to stick to a surface. This was the birth of visual art, the era of the static image. For a long time this was the only mode of visual expression we had.
The Second Era of Visual Art, the age of the moving image, is considerably more recent. It stuttered into existence through the inventive marvels of the 18th and 19th Centuries—magic lanterns, zoetropes, flip books—technologies that enabled an artist to flicker still images at a viewer faster than their eyes could process. Today three-quarters of the world's households have a device capable of casting the light of sequential images at 25 frames per second or more. We working folk love these magical devices. They are the drug of a nation.
The Third Era of Visual Art is the age of the real time. A new set of digital technologies in the last half-century have bought new techniques of creation. Super-human techniques. We have machines that cannot only display imagery faster than a human eye can see, they can also create it at that speed.
The limit of the human eye is, at best, fifty to sixty frames per second. This is called the flicker fusion threshold, the point when our brains can no longer discern any difference between a continuous (i.e. real life) image, and a flickering one. My laptop has a processor capable of generating imagery at ten times this rate, far beyond the needs of my limited animal eyes. This excess of cycles is where the magic can take place. That power can be used to make decisions, lots of them, before an image reaches the screen.
Real-time art means a moving image no longer has to be something created in the past for consumption in the future. A visual can represent the now, not the then. It is a temporal paradigm shift. We can create visual responses that react to us faster than we can to them. This is the basis of most interactive technologies (computers, tablets, phones); input and output as a real-time exchange. We almost take it for granted. The area has been much explored on the web, in video games, product design, and in the digital art world. Although the message is still filtering through to the live electronic music circuit.
Of course, the dawning of a Third Era does not mean the previous two are over. The static image, which has grown through painting, printing, photography, Photoshop and generative art, remains as relevant as ever. There's plenty of life left in the medium. Similarly, the second era will not be subsumed by the third. You might argue the First Era passed a zenith during The Renaissance, or that the Second Era peaked in the late 1960s. But the works of Hitchcock, Godard, Bergman and Kubrick are not the "last word" artworks of moving image. Just as the aesthetic perfection of Caravaggio and Botticelli did not end painting, it simply forced it toward deconstruction, Surrealism and Post-Modernism. Which wasn't necessarily a bad thing. Some of my best friends are post-modernists.
The three eras aren't distinct time periods; they comfortably overlap. But once an age has dawned our perspective has to shift, and we run the danger of looking a bit silly choosing the wrong medium for a job.
Gantz Graf to OpenFrameworks
It's early days for the Third Era. Real-time capabilities are accelerating at a dizzying speed, but there are still hard limits on what can be achieved aesthetically. There isn't the processing power to create the HD polish we've become used to with high-end 3D graphics. Real-time will always look a little retro next to an overnight render. But we've had acceptable capabilities for a decade now, and it's only going to improve. If there's one future prediction I can make with impunity it's that computers aren't going to get any slower.
Audio visualisation is a great example of where we're at with real-time media, one with particular relevance to live music. It is a brand new art form. And already this remarkable tech is in the hands of bedroom coders. We have affordable, portable processing power capable of listening to an audio waveform, breaking it down, and then drawing a visual extrapolated from that data to a screen fast enough that our eyes don't perceive the delay. If you cannot appreciate how fucking incredible this is conceptually you've probably spent too much of your life in the 21st Century.
To my eyes, the benchmark for audio visualisation was set with Autechre's Gantz Graf video. If I had to point to a single work that opened my eyes to the possibilities of the Third Era, it was this insane data mosh. In 2002 it looked like the future. It still does today.
Gantz Graf was way beyond the kind of interactives I was working on at the time (I was still producing work on CD-Rom back then), but I could kinda see how it could be done. Mentally reverse engineering it wasn't difficult.
It was a few years before I had the processing power to try my own experiments. OpenFrameworks opened the door for me here; a tool fast enough to read and process audio in real-time and still have enough cycles left to draw something to the screen. I started porting my code sketches over to oF and putting music through them. I was getting 3-400 fps out of my initial audio-viz experiments.
I was still a long way from achieving the complexity and richness of Gantz Graf in real-time though. We all are. But my progress was in the right direction. I soon realised I could break down the process and do the audio analysis offline. It didn't need to work live if all I wanted to make was a video (Second Era principles). This allowed me to generate images slower than real-time and post-sync it after, freeing myself from the limits of what real-time allowed.
I worked, a decade after Gantz Graf, assuming these techniques had been pioneered years ahead of my own tinkering. But how wrong I was. It was only after reverse engineering the system that had been the model for my experiments that I discovered how hypothetical that system was. It had never actually existed. Gantz Graf was a fiction.
Speculative Futures and Accidental Fictions
Alex Rutterford, who created the video, said in 2002:
Everyone says ‘how long did it take you?’ How did you do it, they always want to ask me technical questions. I’d really love to be able to say to them, ‘I just wrote a computer algorithm, and the computer did it all. I wrote a program and it all just intelligently works it out,’ but it doesn’t exist, it’s fools gold thinking that someone can sit there writing a piece of software that can make intelligent decisions about pace and animation, the closest I have seen is perhaps iTunes.
Rutterford was an animator, not a coder. The mad bastard had done it by hand. It was traditional timeline animation, which I, with my Panglossian future daydreams, had assumed was the hard way of doing. It was so crazily inefficient, so unimaginably tedious, that I had dismissed it as a possibility. But Rutterford's Occam's Razor arrived at the opposite conclusion. He never considered there was, or ever would be, a way it could be done algorithmically. He may still believe this for all I know. But he was/is, I'm glad to say, very wrong.
At this point, iTunes was pretty new, but WinAmp and SoundJam had been around for years. They didn't look particularly good, but they proved the concept at least. To not see the future these technologies predicted was a failure of the imagination. Within five years of Gantz Graf, Robert Hodgin had demonstrated multiple iterations of his Magnetosphere, a Processing system that made all it's own aesthetic decisions and looked incredible. Paul Prudence's work with VVVV was similarly impressive. UVA's work with Massive Attack too. And many, many others. Another five years and the examples would be too numerous to list.
Of course, Rutterford did not set out to deceive anyone. He never claimed his piece was an algorithmic audio visualisation. We, or I at least, chose not to listen to him and instead invented a better version in our heads. Gantz Graf was an accidental fiction, and a very sexy one at that. Only we were to blame for believing it and then going on to make it real.
Good Fictions and Bad Fictions
Fictions drive innovation. We imagine futures then make them realities. The only downside is the potential disappointment of confusing the fiction and the reality. I can't be the only coder who's had a client wanting a Minority Report style interface, but not quite understanding that this required a Hollywood budget just to fake it, let alone do it for real. These real-time interactives don't come off-the-shelf quite yet. It can be realised, but only with a little patience and a decent pot of money. Two things clients rarely have.
Dreaming the futures we want and then aiming for them is a good approach. It is much better than just letting the technology carry us along. The difference between these two methods could be the difference between two of our most commonly imagined futures; the machine-controlled singularity, where human needs become redundant, or the human-lead utopia, where we're all dancing our socks off on the moon.
But there are more pernicious fictions being perpetrated by the Second Era VJs of live electronic music. Not innocent or accidental fictions, but deliberate slight of hand to distract audiences from dull looking performers.
The overarching misdirection in “live A/V” is that we're meant to believe there is some kind of connection between the A and the V. The fiction comes in two main varieties, the flavours favoured by our adversaries; the Pareidolics and the FakeDJs:
1. The Pareidolics hope that if a bit of video kinda fits the music, even if it isn't linked by anything apart from happening to be playing together, the viewers brain will make their own connection. Like playing Wizard Of Oz and Dark Side Of The Moon at the same time. It's a good technique because it works, to an extent—a suitably abstract visual moving in a suitably abstract way can look like it's in time to the music. If you squint a bit. And you've had enough drugs.
2. The FakeDJs are identifiable by video whose sync is too perfect (like Gantz Graf). It can look and sound amazing, until you stop to ask yourself how it's done. If the audio and visual are that well synced, does it mean the audio track is actually part of the video itself? And if the audio is on the video, are those guys on stage in front of it just pretending to make the audio? Ah, yes, it would appear they are. Smile and wave boys. Smile and wave.
The second, to my mind, is the greatest crime of live A/V. It shows utter contempt for your audience, assuming they won't know, or care, if you're just miming to a backing track. But you'd probably be surprised how many big names, names you've likely paid money to see, use this technique live. I know because I have friends who've built these visual mime-along tracks. I've seen their work headlining Glastonbury. With well-known acts pretending to play in front of it.
The first fiction is better, and is time honoured. It's been serving us well for as long as we've had moving image. It looked great when Warhol did it for the Velvet Underground fifty years ago. We like it. It's familiar. It's safe. Comfortable. It's the way your granddad VJs. But, here on the cusp of the Third Era, maybe it's starting to feel a little anachronistic? Dontchathink?
Especially when there are better, admittedly more expensive, real time options. If you're a starving indie band, sure, go with The Pareidolics—it's easy, it's cheap, and just about anything will work. But if you're backed by record company money you need to up your game. And that doesn't mean opting for the second fiction instead. Unless you really hate your audience that much.
If you need that visual boost for a live show there are only three options really. You can look fake, you can look cheap, or you can do it for real.
There Is No Spoon
But it is difficult for me to preach while I'm perpetrating my own fiction here too. There is, of course, no “Third Era of Visual Art”. Not yet anyway. We can't label an era at its outset, before we have the temporal perspective. It's pure hyperbollocks. But I'm hoping my proposal of an era defined by the possibilities of real-time art is a believable scenario. We have the technology so the Third Era is not a vision of the future. It's a vision of a more evenly distributed present.
Gantz Graf was that believable future too. The good fiction. Gantz Graf was to generative artists what Arthur C. Clarke's stories were to communication engineers of the 60s and 70s. What Asimov's tales were to AI in the 70s and 80s. What Neuromancer was to the 'Net in the 90s and 00s. It was a feasible fiction, easily mistaken for reality. Belief in the fiction made it a reality. Much like if you believe in God hard enough, sooner or later you'll hear him talking to you.
My main worry is that to many audience members it doesn't matter if their visuals, or indeed their audio, are live or not. The post-modernist music lover just accepts that everything is artificial, pre-recorded, pre-rendered and pre-mixed. The drug of their formative years was ecstasy, fake love in handy pill form. They can chemically induce a communal moment; they don't need it shamanistically summoned by the resonant sweat of a great performance. They're happy with the fake version.
I'm confident that expectations will change with time though. As do drug fashions. And anyone caught FakeDJing will become the subject of ridicule, as they should be. This may require us to redefine the meaning of “live A/V show”. At the moment it just means a video playing over music. But if we're expecting live A, lets insist on live V too. It's not hard to do (talk to me, or one of my colleagues) and it can make for memorable shows.
Demand it. Welcome it. Let it in. The Third Era of Visual Art is gonna be great.
Matt Pearson (aka @zenbullets) is the author of "Generative Art: A Practical Guide" (Manning 2011). This article is extracted from his new book "Novelty Waves", an exploration of the emergent art forms of our new century. It's available—DRM-free, set-your-own-price—from here.