The art of creating PC generated music
Let's first have a look at the real world !
In the real world, there are a lot of different steps and many different
professions involved in the creation and the recording of a new song.
Just to mention some major steps (without even talking about distributors),
there must be
- A composer who creates new songs, new rythms and writes the
original scores. Nowadays, composers are often helped by arrangers
who are specialized in enhancing the original song, so that it sounds richer
- Then you need an orchestra (or a music group) to interpret the
- This orchestra has to be led by a conductor, who will add his
personal touch to the artwork: he is responsible for bringing music to life,
finding the right tempo, rythm and expression for each part of the song, and
choosing the configuration of the different instrument groups (much of the
beauty of music comes from its 3 dimentional nature), and adapting the play
to the acoustic nature of the environment.
- Of course, you need musicians to interpret each single instrument
that you can hear. Musicians know how to render the musical texture
that their conductor expects.
- Now that music came to life, we need recording engineers, who know
how to capture all the different sounds without flattening and distorting
- Recording an orchestra is a very complex task because of the many
different sound textures, and the wide distances between instruments. If you
record from far away, you will loose the 3D feeling that you expect on a
stereo CD. On the opposite, if you go too near, you will hear only part of
the instruments, and have interferences when a single instrument group is
recorded by different microphones. So you have to record different
instrument groups separately using directional microphones, and you will
need an audio engineer to recreate the original spacial effect in a
studio from the different sound sources.
Now, how about PC generated music ?
The magic of modern technology is that you can simulate all these steps on a
very ordinary PC, provided that you have a reasonably good sound card. Of
course, you can't compare the results of a home made, sysnthetizer generated
song with the technical artworks of the modern music industry ! But compared to
the 'beeps' that an average computer was able to produce 10 years ago, we have
now fabulous sound samples ready and easy to use, and the final assembly is
really breathtaking !
So what are the steps ?
- You still need to compose your music ! For that task, you will need
a music processor. That's kind of
a word processor, but for writing music. the important thing with your music
processor, is that it must be capable of rendering your song in a MIDI
- Since your score is going to be played by your music processor, you will
have to insert all the necessary information that a conductor would
normally have to care of: tempo, expression, volume, instrument disposition,
- Having a virtual orchestra means that you have a lot of virtual
instruments. By default your sound card comes with a general MIDI instrument
palette, which renders your song with a quite averageish quality. The
fabulous concept, introduced by Creative
Labs with their Sound blaster sound card (and it's descendents: Audigy
and X-Fi), are sound fonts. The concept of sound fonts is
simple: just like with your word processor, where you can change the text
fonts ('Times New Roman", "Arial", etc...), you can change
the sound samples that are used to render an instrument: check the example.
- Now we need to recreate the 3D feeling. Most sound cards are capable of
simulating a wide range of spacial effects, many are pre-defined, but you
will soon feel the need to push the configuration a bit farther, to get the
effects you really want.
- Finally, you need a digital sound recording software, to capture the final
result generated by your sound card.