Archive for April, 2012
How to prepare for the future (Performer/Composer edition)
1. Get expressive control of all your fingers at musical tempos (clean 16th notes at 120-132bpm – this can be done on keyboards, guitars, new controllers like the Pico, etc.). This includes the ability to execute scales, arpeggios and other patterns in musically useful ways. Once the playing mechanism is freely independent and trained, it can easily adapt to alternate controllers with a bit of practice.
2. Get expressive control of your wind stream (pick a wind or brass instrument and learn it)
3. learn the basics of subtractive synthesis, and a DAW like Ableton Live LE ($99). Start with Figure for the iPhone at a cost of $1.
4. learn the basics of music theory and how to read notation.
5. write music constantly. Loop-based, through composed, it is all good – the future belongs to entertainers, virtuoso performers, and composer/song-writers who build multi-track environments for others to play in, and to accompany the events of life. Writing good themes, catchy hooks, and great beats will never go out of style.
6. Separate musical quality from instrumental texture. Great themes will be performed on lots of different virtual instruments. Some music (like a lot of Bach) will work equally well with very different instrumentation. Concentrating on themes and their development rather than particular instrumental expressions will help create the broadest future uses.
7. Experiment with expressive performance of virtual instruments. How much expression can you drive? How evocative can you make the sounds you use?
8. Traditionalists: relax. The violin, symphonies, orchestras, pianos, and even electric guitars aren’t going anywhere. Television didn’t kill radio. Virtuoso performers will always be in demand. Great musicians will always be appreciated, as will great instruments. I still want a Fazioli concert grand as well as an Eigenharp Alpha! Nanomaterials and composites will revolutionize traditional instruments in wonderful ways – they will sound better, project better, respond quicker, and be customizable to the player. (They will even be cheaper as on-demand manufacturing becomes common). Of course, there will always be a high-end of traditional makers, working by hand, with traditional materials. This isn’t a better/worse world. It is a both/and world where old and new will co-exist.
I just finished reading Peter Diamandis and Steven Kotler’s provocative book, Abundance. The core thesis of the book is that several exponential technologies are changing the way we live for the better, and will result in profound change in the years ahead. The exponential technologies are:
2. Networks and Sensors
3. Artificial Intelligence
5. Digital Manufacturing and Infinite Computing
In the book, they lay out dozens of examples of how these technologies are being deployed against poverty, water and sanitation problems, energy problems, etc. They lay out a convincing case that the world as a whole can experience abundance, which is certainly an enticing thought.
It does not take someone with a PhD from MIT like Peter Diamandis to realize that if you can align yourself with exponential forces, then amazing things will happen. This is relatively basic mathematics. It is immediately clear how these things apply if you are working as a geneticist, a software developer, a robotics engineer, etc. But it takes a bit more creativity to look at these trends and discover how they apply to the other areas of life. What these technologies enable in their primary sphere is only the beginning. The 2nd and 3rd order follow-on effects will dwarf the initial use cases. We already know what happens when computing becomes near free, but what will happen when education, health care, clean water, and energy become near free?
The rest of this essay explores how these forces could affect musical expression on a planet-wide basis.
The Present State of Musical Affairs
Let us first be clear that there has never been a better time to be a musician on the planet. Already instruments are cheaper, and more available than they ever have been. Digital technologies have already made pianos available for a few hundred dollars, and put a whole recording studio in the hands of anyone with a laptop and a few hundred dollars. High school kids can self fund an album recording, and promote it themselves. Music distribution is practically free on the Internet, and there are even teachers who teach over Skype. Or at least that is the truth in the First World, but it isn’t the whole story for the developing world.
Most of the world, (and even most musicians) do not use anything digital to make music. They use handmade drums, cheap guitars, wooden flutes – whatever is at hand. People WILL make music with whatever they have. But where money is plentiful, we still tend to use acoustic instruments: violins, pianos, guitars, and other “traditional instruments”. What we call “ethnic instruments” in the West are really highly developed traditional instruments, little different from our own. Even so-called “electric” guitars are not really digital devices – they are 100% analog. Where digital technology has been employed in keyboard, woodwind, and percussion instruments, it has largely been to emulate acoustic instruments, or provide a traditional control interface, so that traditionally trained musicians can more efficiently interact with music software. Of course, there has been a huge boom in electronic synthesis, but interestingly, relatively little innovation in instruments themselves.
In the main, this is easy to understand. The piano has been continuously developed for a few hundred years. Same with the violin family of instruments. Pipe organs are even older. Drums are older still – and the human voice the oldest of all. While human beings have an amazing array of shapes and sizes, the basic mechanics of the human body have resulted in several basic classes of instruments: things we blow into, things we bow, and things we hit or strum. Our existing instruments are VERY evolved versions of these ideas. The difference between a concert flute and a child’s recorder is in execution, not idea. Both are tuned pipes, excited by breath, with pitch change accomplished by holes that shorten the alter the apparent length of the tuned pipe. Of course, there are huge differences in materials, craftsmanship, valves vs. no valves, etc, but the basic control mechanisms are quite similar. The violin family of instruments is hugely evolved from its beginnings and modern makers are equaling any work done in the past. Pianos are considered a largely “solved” problem, though we will no doubt see carbon-fibre or composite soundboards that are better than wood at some point, with composite actions that need little regulation and are indistinguishable from their wooden counterparts in function. In the end, analog design is hard – and with hundreds of years into perfecting certain expressions, we have some pretty entrenched ideas of what a musical instrument is. Making an instrument more expressive than a cello is not likely to be a weekend project by a lone inventor. Making a portable polyphonic instrument as capable of musical expression as a 6-string guitar is a significant undertaking. Perhaps equally significantly, almost all our musical education is tied to those instruments, which may suit us less over time.
In the West, many have outsourced music making to their iPhone, even as schools have cut back on art education. It is now possible to get world-class music anytime, anywhere, without ever touching an instrument. So we have a population that LOVES music, and consumes it almost exclusively digitally, but an educational system for music that is tied to analog instruments and expressed through digital recordings. This is broken. The concepts needed to understand folk music, rock and roll, and even traditional orchestral music can be learned in a year or two by any junior high or high school student. Intellectually, the essential components of music theory are no harder than high school Geometry. The equivalent of one college semester of music theory is enough to understand most popular music. But, that is cold comfort compared to what is needed to learn to play an instrument. Learning an instrument is not a primarily “understanding” based kind of learning like we do in school. It is an experiential, neuromuscular training activity more like learning a martial art or gymnastics. Musical instruments are cheaper than they ever have been, but are still uncommon in the general population for one simple reason – they are hard to learn. Musical skill is even rarer, and is handed down in a verbal tradition from a single teacher to a student much as it was 200 years ago.
The Internet has sped this up a lot for popular music, and “slow-down” software has been a boon for budding shredders everywhere. But we haven’t seen the equivalent of a Khan Academy for music. It is very easy to type math answers into a computer. It is a lot harder to play your sax solo as homework into a computer. Part of the reason is the instruments themselves. While the output of the instruments can be digitized, the actual control inputs are analog and can’t be carried electronically. Instruments cannot be configured to the level of the beginner and then expanded as skill develops. Teachers need to inspect movement to see what is going on, but even they can’t tell what is actually happening inside a student’s body. Games and other cutting-edge learning technologies have never really been developed, I believe in part due to the interface issues. Where interface issues have been solved in games like “Guitar Hero”, the interfaces are too simplistic to be musically expressive. Given the millions of plastic “Guitar Hero” controllers sold, the person who can solve the digital instrument controller problem stands to make a fortune.
Most acoustic instruments make a LOT of sound. They are optimized for this purpose, since amplification is a very new technology in the history of music. When you need people in the back of the concert hall or tavern to hear you, then a fully resonant, alive instrument is exactly what you need. If you live like much of the world in small apartments in cities – your neighbors may not be as enamored with hours of trumpet practice every evening. Most acoustic instruments have an economic value and a physical presence that just couldn’t work on a subway or a bus. Acoustic instrument are conspicuous in use. They don’t have headphone jacks, and carrying microphones, audio interfaces, and laptops is way too cumbersome outside of professional or performance situations.
Put simply, the possibility of unleashing an entire planet to be actively musically creative is available in the next ten years. We can move from a world where most people passively consume music to one where anyone can enter into music making with a digital instrument that meets them at their level of skill. There are several ancillary facts that must be understood to place the opportunity in context:
1. Over half the planet has a cell phone. Even in Africa, there will be 125 million smart phones in use by 2015.
2. Every smartphone has enough power to run a basic multi-track recording and synthesis package. Download Propellerhead’s Figure app from the Apple Store to see this. An average laptop already has more power than most musicians can use. This will only accelerate. We should assume that it is essentially free to run a full “Omnisphere-grade-synthesizer”, and full DAW package on the average cell phone within 5 years.
2. Internet access is already, and will be ubiquitous, though it will not always be accessed by laptops (think cell phones and tablets). In the first world, it will be broad-band quality or better all the time.
3. Computing power and storage in the cloud is close to free, and supports “almost-free” business models.
4. 70% of the earth’s population will live in cities by 2050. Space and noise constraints exist now and will only grow.
5. Modeled instruments or hybrid modeled/sampled instruments will likely replace sample-only instruments. They are simply more expressive. My modeled piano has infinite half-pedaling – and is more expressive than my acoustic grand piano in that regard. Samples can be very high quality but are a chore to manipulate, and hard to use in real-time. Models can react in real-time. This is what performing musicians need – real time response and control.
What is needed is a digital instrument – one that Moore’s law and mass manufacturing makes cheaper and cheaper with several important characteristics:
Open hardware design specs. Part of the reason violins work so well is no one has a patent on them. Anyone can try to improve them. Digital technologies mature very, very quickly in an open environment. Parts are cheap and getting cheaper. Having patented hardware will just slow adoption. The goal should be to make the digital equivalent of an electric guitar – we have no shortage of manufacturers or differentiation of an essentially “finished” instrument. Key hardware concepts:
– at least 2.5 octave playable range (shiftable on the fly) – that’s enough for killer solos, more in artist grade instruments
– polyphonic capability in a portable package- it is one reason why guitars are so useful
– breath control sensitive to the full range of human respiratory power (current breath controllers are very inadequate in this regard)
– pressure-based expression as sensitive as a violin bow
– velocity-based expression as good or better than a piano or a hand drum
– pitch-based expression as good or better than bending strings on a guitar
– ability to use all fingers simultaneously
– allow for expressive movement while playing – it is part of what makes electric guitar playing so watchable
– re-mappable control surface (hide the “bad” notes, learning games)
– enough data rate for MIDI 2.0 transmission
– wireless connectivity to a cell phone (Bluetooth is ideal, but likely too b/w constrained)
– USB is acceptable in the mean-time (iPad’s have adapters)
– portability: think Eigenlabs Pico form factor
The hardware goal is a base sensor and data-rate specification that allows musician-grade expression to be in every Jansport backpack in a major city. The hardware is really just an interface to software and cloud functionality, and so will be able to respond appropriately to someone with little skill or a lot. Of course, professional musicians and entertainers will use larger, more capable, more expensive, and nicer models. It is really hard to beat the Eigenlabs controllers in this regard for a “key oriented” controller. There are many other alternate controllers, but the Eigenlabs are the best thought out in my opinion, and meet most of the criteria above. There is certainly room for an equally well thought out string and woodwind controller that maps to those types of players as a natural interface. The future demands controllers that are not based on MIDI. We need high-bandwidth controllers that respond to the full range of human control inputs (ie. breath controllers actually sensitive over the WHOLE range of human breath, not just sensitive to a small portion of it), and we need rich data to feed to instrument models things like string pressure, bow speed, angle, position relative to the bridge, etc. This is what will transform static samples into real modeled instruments.
Commodity manufacturing. Production volume lowers price. The price target for a basic consumer controller should be sub $50. At this price it can be an accessory to a cell phone. Artist versions will always be more expensive, sensitive, and profitable. Some things won’t change. Already a Eingenlabs Pico can be purchased for about $500 USD. Volume production should allow it to get under $250 and new materials will enable future devices to go well below that, particularly if the large MI companies get involved. New nano-materials will be custom built for controller bodies at very low cost, with integrated sensors that do exactly what is needed. Then we can have a continuum from a musically useful “game controller” all the way to artist grade instruments. Open hardware spec will enable cottage industry for high end craftspeople to remain involved. The large MI companies should welcome this, not fear it. Significant innovation comes from the small shops, and they pose no commercial threat since they have such low volume production.
Open software. Just like computers and phones need an operating system, Digital instruments need a host application. This application needs to accept the standard inputs from hardware. The software needs to provide:
– cellphone, tablet and laptop connectivity as a plugin to music software
– MIDI 2.0 output of rich expression data – high bandwidth, nuanced data
– Extremely low latency – should feel analog in its response
– ability to host arbitrary software instruments (AU’s, VST’s) – i want to create and play my own instruments
– Map and save control surface parameters
– loop-based music creation software – like Reason/Ableton – 8 tracks to start
– connectivity to cloud-based services
– some effects like reverb or high quality instrument models are very computationally expensive for a phone. If the digital performance information is rich and nuanced, then better quality models can be accessed in the cloud, and the final project rendered there, rather than on the device.
– storage of projects, patches, etc
– publishing songs/social media interactions. This needs to be open so that social media vendors like SoundCloud, Facebook, etc. can write conduits to post material to their sites.
Cloud-based Features. By 2023, $1000 laptops will have the computational equivalent power of an entire human brain. Cell phones will have far eclipsed what even a high end Mac Pro can do today. But even a Mac Pro stuffed to the gills, is still stronger for being connected to the cloud. “Saving” a file locally makes little sense if it can be backed up, automatically distributed, and shared in the cloud. Collaboration is almost more important than individual creation. What I make needs to be shared, edited, mastered, produced, integrated, synced to video and a thousand other things. if we have plug-in delay compensation today, we can have network delay compensation in the future. If I am performing on someone’s song with my digital controller, they can stream the bits to me, and my playing nuances can be recorded in their software as well as mine for future editing. I can “monitor” a sound off my cellphone, while the cloud renders a much higher quality model through a digitization of a killer acoustic space, and downloads it to my phone, for playback within a few seconds after I finish playing. Recording studios will be set up for workflow, primarily, but some will have traditional rooms to record acoustic instruments. Many projects will never use a studio at all. Digital instruments interfaced through a cell phone will be able to access the best instrument models on the planet. Just like you can rent varying qualities of instruments today, you will be able to “choose” the software model you wish to use. Better models are bound to be more expensive. The cloud will make computational expense a non-issue for instrument design and effect quality. Hardware DSP will be completely unnecessary, except perhaps as used in the cloud itself or for large studios. Sound designers that make highly expressive instruments for people to play on their controllers will be very valuable. Performers will play custom virtual instruments – even commissioning them for a concert or a recording session. The cloud functionality needs to be fully open, distributed and heterogeneous, and all this will develop organically.
Music Education 2.0. Instead of learning to read and play music through out-of-copyright folk songs, or even quality classical composers, we can create a whole online academy. Just as Khan Academy has revolutionized math instruction, the same opportunity exists for music education. Levels can be created, lessons given, technique evaluated based on analyzing the streams of data coming off the instrument. If someone is pressing too hard, or not hard enough, or is consistently inaccurate in some way, this will be easily diagnosed. As nanotechnology biological sensors become common, it will be possible to wear “sleeves” that monitor muscle movement and feed that data back for analysis and trending. Even analog instrumentalists will benefit from this kind of training. It will do for musicians what treadmills, VO2 max testing, and power meters have done for sports. Once the neuromuscular activities can be quantified, then deep analysis will help students to progress at the optimal rate. Exercises and skills can be tailored to each student. Teachers can be local or remote. Control surfaces can be mapped to their level of development and progressively “unlocked”, wrong notes can be “mapped out” of beginner’s concerts, increasing confidence. Keys can light up to “show the way” through a new piece. Advanced performers can join virtual ensembles or jam with increasingly more difficult rhythms and harmonies. Smart software should automatically adjust if student performance is inadequate. Perhaps best of all, music performance can be clearly separated from music theory in a way that has never before been practical. Since controllers will be able to be both simple AND expressive, it will open up a whole new world of composition and performance opportunities. Imagine a keyboard instrument that allowed a beginner to fire off a loop that plays what a more advanced student might play live. This will allow everyone to participate in music making, and even allow students and virtuosos to share a stage in a way that is just not possible with analog instruments.
Musicians will perform in virtual performance spaces – all of which will sound as good or better than physical spaces. Some will perform as themselves. Others will have avatars that are as well-known as any international star today, but will allow their creators to move with anonymity in the real world – a welcome relief for many creatives. The fame-hungry will, of course, pose as themselves. Holographic techniques will allow musicians to appear live in multiple places at once. Real, physical performance will command a big premium. Virtual attendance to events across the globe will be commonplace. Fans will be able to attend every concert on a global tour if they want. The kind of immersive technology that underpins 3D gaming will be available to make interactive concert experiences. Different concert goers may have different experiences of the same set, or even get to play along with the band for extra charge! If you know the parts, why not play them yourself? Creative producers, musicians, video, FX, and artists will collaborate to make incredible multi-sensory shows. When physical barriers are removed, much more collaboration will be possible between highly-skilled musicians, and it will be possible to have jam sessions with people on the other side of the globe, barring sleep cycle issues. Bands will be able to sell “controller maps” or “performance walk-thru’s” to their fans as an extra revenue stream. Sounds from the recording will be purchased, or even multi-track master access for remixing or inserting one’s own solos will be available. There is a whole world of interactive content that the best musicians will create allowing others to interact with and co-create with them. The most popular musicians will be either great performers, great collaborators, or great enablers of masses of people to interact with music, or all of the above. Musicians that can make highly immersive sonic environments that adapt to different skill levels and deliver a quality experience will be as popular as any video game. If controllers are cheap and plentiful, and music education is natural, and easy, then music can again be an interactive experience, not just a concert or a recording. When everyone can learn to make some simple gestures and “play music”, then the world will be a richer place. Put all those people on the Internet, and some amazing new music will result!
This year, for Easter, I composed a short dance suite. The initial desire was for something, “Tribal and Triumphant”, in the words of a dancer friend. Easter Triumph is intended to give a glimpse into a part of the Easter Story that is rarely considered. On that first Easter morning, Jesus met Mary, and then ascended to the Father. This composition explores that heavenly meeting.
1. Introduction. The piece opens with the angels milling about in heaven, waiting. There is a randomness to their song as they await his coming. The piano enters and the angels become aware of His approach. They hastily organize themselves and a welcome chorus breaks out over a Cuban beat. They are overjoyed to see the Son again, coming victorious and leading a host of captives.
2. Throne Room. Three massive gong hits sound as Jesus arrives at the throne room doors. A quick scurry and the angels are in place. Jesus is announced by a brass fanfare, and approaches the Throne to a magisterial brass march. As He stops before the Throne, the Father welcomes Him warmly with a theme on solo trombone.
3. Coronation. Jesus is crowned King of Kings and Lord of Lords and then ascends the steps to take his seat at the right hand of the Father to a fully orchestrated arrangement of “All Glory Laud and Honor”. Angels attend the celebration with stately grandeur.
4. Celebration Dance. As soon as Jesus is seated, Heaven erupts with joy – thousands of years of anticipation culminate in a massive rave in Heaven. The throne room pulses with the energy as history’s highest moment is celebrated.
I had an interesting experience today. My iPhone was playing on “random” mode, so it brought up a bunch of things that I don’t listen to that often. One song, by a currently popular nationally-known band was irking me. This is a song that is produced to the highest commercial standard, and is on an album that has sold hundreds of thousands of copies. it has been a “top-10” radio play in its genre. I hit the “skip” button and a few songs later I had the same experience, this time with another well known female vocalist. The song was at least 10 years old, but in its day held a similar position in the marketplace. Both artists would pack a venue still, and both have real talent and ability, and both have sold WAY more music than most ever will, so why was I bothered, and does it even matter?
I recently finished writing a short 6 minute dance suite for Easter. It is not at all a traditional piece in a classical sense, though there are sections that are traditionally orchestrated. Each of the three sections had its own learning experiences embedded, but the hardest part musically was the original themes that I wrote for a short brass fanfare, and for a processional march. They aren’t long by symphonic standards, and are not even developed in my piece – simply stated as part of the flow of the piece. I wrestled with relatively small changes in note value, and experimenting with the feel of different intervals, etc to get the feel correct for what I intended. Ultimately, I was satisfied, but concluded the writing session aware of how much more work would be need to develop these and additional themes in the context of a larger work.
Coming back to the two pieces that caught my attention, my first observation was that they were fatiguing to listen to. While this might be from too much compression, or other factors, I decided to analyze it more closely. In both cases, the words were about majesty, power, and such as reflected in the worthiness of the Deity. The sonic spectrum was dense with big drums, guitars, and other bits to fill in pretty much every sonic space. But the melodic themes and vocal delivery were not majestic, powerful, or even dramatic. As is the current fashion, the female singer was breathy, airy, light, and with all kinds of small vocal flourishes at the end of every line. Intimate, expressive, and soft, were the adjectives that came to mind, which wasn’t working at all with the words (or even the raging guitars and drums). The male lead singer of the band sang with some “grit” on his vocal, but had short phrases that didn’t evoke any largeness at all. In both cases, volume and sonic density was substituting for thematic content. What was bothering me was that it sounded big, but really wasn’t. The sound was big, but the musical ideas weren’t. Loudness and density can be used to emphasize a point, but what if the point is small, but the emphasis is over the top?
The experience helped me to remember that the problem of thematic appropriateness is not tied to any particular genre, musical style or subject matter. Even having a good “hook” is not the same as having thematically consistent and excellent ideas. My experience is that quality ideas are harder than quality production skills. My own recent experience in composing coupled with observing what happens when the musical themes are not well connected to the song encourages me to focus on themes and their development as the main task. It is the musical equivalent of, “Keeping the main thing, the main thing”.