Nikko75
Listener and Music Lover
According to those at hydrogenaudio, audio playback is perfect as far as our ears are concerned. While I have yet to receive any reply to a request for proof of such "studies", the ongoing claim is that some elusive ABX tests results show subjects could not identify between DACs (re; CD players and soundcards) dating from the mid 80s to present. Of course, they're the same group that also believe tests that suggest 256kbs MP3s are sonically as good as redbook and DSD.
Sadly, a lot of skeptics and big wigs think that way; that audio playback was perfected long ago and that further progression isn't needed. I take such claims with a pinch of salt.
Has anyone here came across Gibbs phenomenon? It's a distortion on square and fast rising waves below the bandwidth limit. It's prevalent in all on-Gaussian DAC filters. When a stimulus contains waveforms beyond the Nyquist limit (as is also the case with a perfect square waveform at any frequency) then bandwidth limiting is going to give rise to ringing unless the bandlimiting filter is Gaussian shaped. If you can use Excel, here's an experiment for you. Write out the Fourier series for a 2kHz square wave up to some high order term, then truncate above that point (i.e., perfect brickwall filter, the ideal for preventing aliasing while maintaining an essentially perfect flat response over the desired bandwidth). Now enter the equations into Excel and plot a couple cycles. It's quite bad and is present into the mid X00Hz range.
The way to avoid distortion on signals with a high rate of change is to use a milder filter with a smooth roll off. These are not as flat from 10kHz upwards, but seem to get more appreciation among people who actually listen than brickwall types do.
Last edited: