Fixing ComputeSpectrum()

Today i stumbled upon a couple of blog posts that made my year!

I love having the ability to visualize stuff, especially audio whether it be a live stream or on demand sound. A couple of features i have played with a lot since the release of ActionScript 3.0 are SoundMixer.ComputeSpectrum() and BitmapData.draw().

I’ve always had a good handle on both (yet i’m still learning) but one thing that drove me absolutely crazy was the lack of documentation specifically regarding to what ComputeSpectrum actually spits out. There have been many experiments, usually followed by bouts of depression and hours of frantic searching looking for something…anything…or anyone who might be in a position to attempt to shed some light on things.

The most unanswerable questions where:

  • Why is FFT biased towards low frequencies?
  • What frequency band does ByteArray.position = N relate to?
  • Are the 256 bands per channel equally spaced?
  • How can i correct the data to produce a realistic display?

Well here’s the first step:

Ben Stucki has started to attempt to explain his findings in respect of better understanding ComputeSpectrum Ben Stucki – The Math Behind Flash FFT Results

He also references another blog post What’s In An FFT

Lets just say i’m on this big time…


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: