rhoadley.net music research software blogs
aru seminars m&t critski focm1a cmc circuit bending mic2b sensor technology comp 3 sonic art major project
youtube vimeo facebook
Linking old and new technologies
Understanding how something works is an understanding of how a typical, but unreal instance of how that thing works. It does not allow a detailed understanding of the precise details of a real example of the thing itself, including in this its physical aspects, including its location.
Real Recording Abstract Since the 1950s there has been considerable development in the area of computers and music. One of the major areas within this has been audio recording and manipulation and nowadays, especially with the rise in the power of the desktop computer in recent years, there is a plethora of software available at all levels for this purpose. Another area that has developed substantially in recent years has been MIDI. While a less resource hungry medium, MIDI has become especially popular in commercial musics, games and across the internet. Each of these areas has advantages and disadvantages, but neither records all information concerning a musical performance. While an audio recording will record all sound data, no attempt is made to provide any means to make the individual aspects of a recording - for instance, the two parts of an instrumental duet - available for editing within the same recording. Two separate recordings must be made and then, optionally, mixed together. Within a single recording, however, the only editing options are global ones that effect all aspects of the recording. MIDI is not like this - it records, stores and plays back certain 'musical' parameters such as pitch, velocity, programme information and volume. If a manufacturer has provided it, it can even store information concerning the internal state of a sound module. All this information can subsequently be edited. However, none of this information includes details of the actual sound of the performance. Instead, it is used to trigger sounds made by a MIDI compatible sound module. While this can produce acceptable results in some environments, clearly, it is much less acceptable in others. A good example of the latter is where there are significant live and physical performance aspects, such as in the playing of a stringed instrument. In these cases, all aspects such as bow control, minute finger movements and other sounds created by the bow and fingers are lost. In some respects, no live performance can ever be correctly reproduced via MIDI, although MIDI does allow for the editing of individual parts of the performance. Our proposal is to make ground in this matter through a number of routes that are based around the specifically musical and technical matters that arise from the problem. One is that a significant amount of data pertaining to musical performance is lost because the majority of the physical actions of the performer are in the case of both forms of recording, discarded or simplified irreparably. A simplified situation will be arranged in which the recording of, for instance, a violin player will be made. But this will be a recording of the physical movements of the player - most obviously, fingers and hands, but also other bodily movements. This is in itself not dissimilar to current techniques used for creating computer animation. However, in this case this is a side product only. The substantive point will be in investigating and applying this data to other sound sources. This would mean investigating how precisely the original 'sound' or more appropriately the original 'performance' could be simulated, but creatively there are many more uses - most specifically, the use of the data to enhance 'standard' audio compositions and performances with 'live' data. The project would not require a vast number of resources:- 1) a computer with the hardware and software ability to a) record and playback video information b) process and/or synthesise audio information, ideally although not necessarily in real time. c) construct any required original software. 2) video equipment. 3) hardware components to sense movement data from a performer's body. It would be envisioned that, given sufficient time, the first results of any such research would be available after a period of one year. Richard Hoadley, Music Douglas Nunn, Audio Technology February 2000
The interface between music and technology - one of my current interests.
Also, the linking of old and new technologies.
I want to talk about these technologies and their effects. Why I feel they are, currently at least, radically different
How much experience of new technology do you have?...
How much do you want?
If we get around to the computer, things might not work - this is part of the thing.
If you think about it this is the interface between a piece of hardware, an object, and a human being.
By definition, this will be very much about ideas. It's worth pointing out that there aren't necessarily answers to these questions - certainly not conclusive ones, but there are issues that you can bear in mind while thinking about it.
Before the following question, think about the following thing - in an orchestral piece, someone knocks a music stand over. Is this music?
Question: What is a musical instrument? What features make a musical instrument? While you're at it, come up with some definitions of a piece of music - what is it about a piece that makes it one, what characteristics mean that something isn't a piece of music?
Discuss
[This is important as there are many things that, convincing as they may be, aren't included on analogue or digital audio recordings. You cannot edit an individual performer's actions - or at least not convincingly. How might you go about doing this? You can edit a MIDI track, but then the sounds aren't live.
]In order to think a little more about what a musical instrument actually is, let's think about something. Don't worry if some of the ideas seem obscure or impossible, that's not the point.
Software mimics real life: document, mouse, desktop
Most of you will have come across software that tries to mimic real life. In fact, most of today's software does this - a document looks like a document, the pointer of a mouse mimics the movement of your hand, the operating system mimics a desktop. A sheet of manuscript paper mimics a sheet of manuscript paper. This is, in a sense, obvious.
make it appear to be real, make it behave realistically
If you want to do something on a computer, the easiest way is to make is appear as if it is that thing, and try to make it behave in a way that makes it appear as if it's real. It usually works, too - and when it doesn't we get frustrated and often respond in ways that have no relevance to the computer - but it's so convincingly real, we can't help it. We bash the mouse, we curse the screen, we tap of the keyboard's keys harder.
Think of a synthesiser. Why are most synthesisers provided with piano-like keyboards?
Stuff about analogues, pianos mechanically connected, digitals no real connection, or often the connection can be edited.
Imagine...
...a real instrument. Say, a flute. Imagine you have a very complex computer programmed to be a flute. You don't need the details, but if you want them, assume that it looks, feels and sounds exactly like a flute. If you want, you can have headgear, gloves, all sorts of things that you might think are required. It doesn't matter. All you have to assume is that the thing you think you're holding and are about to play is a flute. It is, to all intents and purposes a real flute.
So what's the point?...
...the virtual flute is, in fact, a combination of complex hardware and software. In fact, it needn't be software, and if it is hardware, it is, effectively a flute. If it contains any software, then it is not. Why? The whole point of something containing soft, or even firmware (an intermediary between soft and hardware), is that the code can be modified. Who by? Depending on proprietry controls, copyrights and the like, anyone who knows how to do it and has access to the code.
It's highly likely...
...that the flute has some way of changing certain things to make it more useful, so, maybe, you can extend it's range a little, make it sound like a piccolo if necessary, so the performer doesn't need to carry around extra boxes. Why not, using the same logic, extend it to an alto flute or a bass flute?
In fact...
...a great many flute players are in fact woodwind players. If the software/hardware exists, why not extend beyond the flute family and include a 'clarinet' button. It could be that, like a synth, the 'flute' simply sounds more like a clarinet, but we can do better than that - we can make it appear to be clarinet.
So how far can you go?
What might you want your 'flute' to do? It already has a vast range, can sound like another instrument - why not make it sound like any instrument? Or any combination of instruments? Why not, in fact, make a much better synthesiser than the ones we're used to? (Because it's very difficult and expensive.)
This is actually a computer
In fact, we do have a sort instrument like this, although it's not terribly good yet, and guess what, it's called a computer. So we're back where we started. And what have we proved?
Conventional musical instruments can't be altered!
That, probably, what makes a conventional musical instrument what it is is that it can't be altered, much. Any such instrument, in comparison to a computer program, is extremely limited in the repertoire of things that it can do, but far from being a disadvantage, it's one of the things that makes playing an instrument really well possible.
If you consider a really good piano player... ...how do they hit so many correct notes so accurately? If you can't play the piano, you realise just how difficult - apparently impossible it is. But if you can, you think it's easy, because to you, it is, it's become a part of you.
Limited repertoire means intense study and therefore much skill
This is why until now, while the computer is often used as if it were a musical instrument, there are very few examples of good computer music. This is not to say that there won't be in future, but in my opinion musicians will be hampered by the computer's very flexibility.
Of course, existing, traditional musical instruments are technological too, they're just fairly straightforward in comparison to a computer. Most began very simply and gradually developed into what they are today through the use of mechnical and material changes - so the flute began as a fairly straightforward tube with a few holes and developed into the complex machine we use today. Over the 20th century, when most traditional instruments have mainly stopped developing, composers have further modified the use or the structure or instruments in order to obtain new sounds - so the violin's strings may be plucked hard, brass instruments may be muted with a variety of add-ons, some woodwind instruments can be manipulated to play more than one note, and so on.
As an example of this, we are going to investigate one of the more profound and interesting modifications to one of the most traditional of instruments - the piano.
Listen to these two short piano pieces:
In case you couldn't tell, these are pieces for piano. They were composed over fifty years ago, by the American composer John Cage. He's probably most famous, unfortunately, for composing the silent piece, 4:33.
The prepared piano is a good example of an extended instrument - the basics of the piano are still there, but the addition of preparations it's sound can become radically different. In this way, one is using an old technology - the piano, with an even older one - the bits and pieces simply experimented with - in a sense, making your own instrument, to create something new and vibrant. Of course one is making one's own instrument using a highly sophisticated piece of machinery - the piano.