@ eatenpi: as the mail states, you are following jonobr1, so you will receive an email of every post he makes.

On topic, for starters the minimum boundary is 0. So that's one down, one to go.

I believe there is no method to get the max amplitude (per band) and it's definitely not a fixed value. I tried some practical experiments and it seems the totalMax is somehow - directly or indirectly - related to the buffer size. One could elaborate this experiment by seperating it for each single band to see if there are differences. So in short, I think the solution / workaround is just to DIY, for example by just keeping track of the maximum, if needed per band.

Code Example

1. import ddf.minim.analysis.*;
2. import ddf.minim.*;
3.  
4. Minim minim;
5. AudioPlayer sample;
6. FFT fft;
7. float totalMax;
8.  
9. void setup() {
10.   size(200, 200);
11.   minim = new Minim(this);
12.   sample = minim.loadFile("jingle.mp3", 2048); // buffer size affects totalMax
13.   sample.loop();
14.   fft = new FFT(sample.bufferSize(), sample.sampleRate());
15. }
16.  
17. void draw() {
18.   background(0);
19.   fft.forward(sample.mix);
20.   float[] mma = getMinMaxAvg(fft);
21.   if (mma[1] > totalMax) { totalMax = mma[1]; }
22.   println("min: " + nf(mma[0], 1, 3) + " | max: " + nf(mma[1], 2, 3) + " | avg: " + nf(mma[2], 1, 3) + " totalMax: " + nf(totalMax, 2, 3));
23. }
24.  
25. float[] getMinMaxAvg(FFT fft) {
26.   float minimum = 99999, maximum = 0, avg = 0;
27.   for(int i=0; i<fft.specSize(); i++) {
28.     float value = fft.getBand(i);
29.     if (value<minimum) { minimum = value; }
30.     if (value>maximum) { maximum = value; }
31.     avg += value;
32.   }
33.   avg /= fft.specSize();
34.   float[] mma = { minimum, maximum, avg };
35.   return mma;
36. }
37.  
38. void stop() {
39.   sample.close();
40.   minim.stop();
41.   super.stop();
42. }