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 and smoother.
Then you need an orchestra (or a music group) to interpret the scores.
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 them.
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 format.
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, etc...
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.