|
Kewlness, you should take some care when analizyng tracks. For example:
1. Don't analyze the whole track. If you want to compare freqs between two tracks you should choose a similar passage: rythm+bass+chords, only bass, only rythm, only pads... etc
2. Only analyze tracks proceeding from the same source (CD, Vinyl, Mp3). Each format has its own rules: Vinyl has an unavoidable crosstalk between L&R channels, MP3 usually cuts off frequencies over 16 Khz, CD has a -0.2 dB headroom and is dithered, tapes limit the stereo range... and so on... Tracks should be digitalized at the same frequency (44.1, 48, 96...) and wordlength (16, 20, 24, 32...).
3. The analisys window (time region) should be as equal as possible (the resolution of the freq analisys depends on the number of samples).
4. Overall... frequency analisys won't give us the "secret" of the pro sound. It's only a math trick. The freq analisys supposes that the ear is a linear system. There is nothing so strange, personal, surprising and nonlinear than the human ear. There are things more important than single freqs, like timbre, dynamics, aural (stereo and even more than stereo) perception.
Sometimes it's better to turn off your uv-meters, spectrographs,... and just hear what you are making (of course IMHO).
|