How does one use the frequency domain for playback processing? Are there any advantages/drawbacks to using this instead of time domain processing?
Disadvantages of doing processing in the frequency domain are:
Increased latency involved in the FFT/Windowing stage (8ms minimum using 256 blocksize, increasing with higher blocksize). Blocksize=256 Stopband Attenuation(dB) 1_31 32-45 46-59 60-73 74-87 88-102 Latency(samples) added 384 640 896 1152 1408 1664 Latency(ms) 8 13.3 18.7 24 29.3 34.67
Increased latency involved in the FFT/Windowing stage (8ms minimum using 256 blocksize, increasing with higher blocksize).
Higher memory requirement
Possibly more setup when managing many frequency bins
Advantages of doing processing in the frequency domain are:
Less computations required
Higher resolution filterbanks/crossovers
Linear Phase
Constant, flat group delay
Sharper filter band edges.
I will attach an example system for further help soon.
filterbankeq.awd Here is the example file for the FilterbankEQ.
6:03pm
Disadvantages of doing processing in the frequency domain are:
Advantages of doing processing in the frequency domain are:
I will attach an example system for further help soon.
7:23pm
Here is the example file for the FilterbankEQ.