Visualization.spectrum doesn't make use of pFreqData

[mattblovell]

It doesn't seem like this visualization.spectrum addon makes use of frequency information for its display.

The CVisualizationSpectrum::AudioData() function makes no use of pFreqData or iFreqDataLength. Instead, a somewhat odd sampling of the amplitude information carried by pAudioData is used instead:

visualization.spectrum/src/opengl_spectrum.cpp

Lines 455 to 482 in f3dceca

	int xscale[] = {0, 1, 2, 3, 5, 7, 10, 14, 20, 28, 40, 54, 74, 101, 137, 187, 255};
	for(y = 15; y > 0; y--)
	{
	for(i = 0; i < 16; i++)
	{
	m_heights[y][i] = m_heights[y - 1][i];
	}
	}
	for(i = 0; i < NUM_BANDS; i++)
	{
	for(c = xscale[i], y = 0; c < xscale[i + 1]; c++)
	{
	if (c<iAudioDataLength)
	{
	if((int)(pAudioData[c] * (INT16_MAX)) > y)
	y = (int)(pAudioData[c] * (INT16_MAX));
	}
	else
	continue;
	}
	y >>= 7;
	if(y > 0)
	val = (logf(y) * m_scale);
	else
	val = 0;
	m_heights[0][i] = val;

I found mention of this in a kodi.tv forum back in a 2014 thread:

https://forum.kodi.tv/showthread.php?tid=204991&pid=1846424#pid1846424

the spectrum viz doesn't use fft data... there are no vizes in mainline using the fft data. it's still borked, but that is NOT a way to test it.

Wow. No wonder it looks nothing like a spectrum analyser display...
I really think it should be using the fft data (well assuming the fft data is fixed).

So it just uses a random subset of audio samples. That would explain why it looks like random noise.

and

well, it's amplitude bins for each data set. so not entirely random but surely not a proper spectrum.

[mattblovell]

The AudioData() member in this class can be contrasted with that in visualization.starburst:

https://github.com/xbmc/visualization.starburst/blob/c4d57064d023d4038b48027a1983d3c9bac2c228/src/StarBurst.cpp#L185

The StarBurst visualization also asserts wantFreq in its override of GetInfo(), in order to get pFreqData populated.

https://github.com/xbmc/visualization.starburst/blob/e2f781fbb00ab2cd422cc856d485d2e0480cc357/src/StarBurst.h#L63

[AlwinEsch]

With #49 becomes the work about changed and to use FFT, after them is in becomes the issue here closed.