Andrew Sega
Also known by the moniker Necros, is an American musician best known for tracking modules in the 90s demoscene as well as for composing music for several well-known video games.
I have a love for cheesy music. I don't want to list any bands and embarrass myself ;D
We started out in the middle ages creating music which had certain desirable physical properties (for example, a major chord sounds "nice" because the frequencies are in integer ratios to each other). And then as society evolved, we created these emotional contexts for certain instruments and progressions. Major-chord arpeggios sound "happy", minor chords sound "sad", chromatic scales can sound "scary", et cetera. In the 20th century, film soundtracks reinforced this point as people associated certain kinds of music with certain visual and emotional experiences. It's a giant feedback loop, really; once you grow up in a given culture, it leaves this musical fingerprint on you which colors your experiences.
One of the most common failings [when writing a drum track] is repetition. Nobody wants to hear that same stupid 16-line bass-snare pattern throughout the WHOLE song. Didn't your mother ever teach you that variety is the spice of life?
A song is not just a collection of melodic riffs, it is an emotional statement.
A lot of my tracked music was written when I was very young, relatively speaking. I was very optimistic, I had finally discovered some sort of public musical outlet, it was generally a very happy time. So, the music of that time sort of reflects that, I think? By the time I started the Alpha Conspiracy project, I was older, a bit more sophisticated (and, well, cynical), and so the music got more complex.
If I was obsessed with making money off of my music, I wouldn't have released it for free on the internet for the last 5 years.
Certainly there are songs out there that are massively popular, and you can use some reverse analysis to see how they're put together, but it's difficult to reliably engineer a "hit". Add to this the constantly swirling winds of cultural taste, and you can see that the music industry is more akin to playing the lottery than anything else. Having some talent or taste can give you a bit of an edge, but it's still a huge roulette wheel..
A human is a complicated organic/electrical system, which is immersed in a culture. Try raising a monkey like a human, it won't work, you need the human's certain brain characteristics (including self-reflection) in order to create a truly intelligent creature. The brain processes inputs and perception in very particular ways, and I think until we understand the underlying processes better, there is no way to really simulate it in software. Computers will continue to be good at simplistic analysis, and raw processing power, but the subtlety of emotions is something intrinsic to the human organism and culture.
Stylistic evolution comes from listening to what other people have done and making a Darwinistic modification.
Ideally, a song should contain both elements of high melodic tension, and low melodic tension. No listener wants to sit through a totally high-energy 180 BPM non-stop 6-minute ride through synth mania unless they are already busy grooving madly on some dance floor in a smoky club somewhere. Also, unless your listener is on heavy sedation, he or she will not enjoy your sparse 18-minute ambient tune which consists of the same languid piano riff repeated over and over again.