Home Digital Drive

Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it.

[ Follow Ups ] [ Post Followup ] Thread: [ Display  All  Email ] [ Digital Drive ]
[ Alert Moderator ]

RE: PS Audio DirectStream DAC - an extremely interesting video.

Imagine for a moment the "perfect" brick wall filter, one that completely zeroes all content above the cutoff frequency and leaves the amplitude and phase response below cutoff completely unchanged. You can implement this perfect filter by FFTing the whole time series, setting all frequency samples above cutoff to zero, and iFFTing back to the time domain. The result will have pre-ringing. I think it is a natural side effect of band limiting.

You can implement a filter with no pre-ringing and only post-ringing in either the analog or digital domain, but as Ken points out it will have phase distortion.

Here is one more hypothetical to think about, which will probably seem counter-intuitive:

Suppose you start with an analog signal that is band limited so it has no content below fc, where fc < fs/2. Then you put it through an ADC-DAC loop and the reconstruction filter is a linear phase brick wall FIR type with flat response to fc. The output will just be the original signal plus dither noise. The ADC-DAC loop will not have added any ringing. Any ringing present in the output will be due to the band limiting of the original signal prior to conversion. This may seem counter-intuitive but is mathematically true.

Note that the digital test data that produces pre-ringing when put through a DAC is not properly band limited and thus violates the sampling theorem. We're used to seeing DAC measurements in Stereophile and other places which use a perfect digital impulse or digital square wave as the test signal. These are not band limited and are not representative of digital data that came from an ADC. They make it seem like the reconstruction filter adds ringing artifacts to the signal but if the signal is properly band limited in the first place, it does not.

Of course, if you start with a music signal with frequency content > fs/2, then you have to filter the data to band limit it, and the filter will introduce ringing. It could be pre- and or post-ringing depending on the filter type. So I think that putting all the attention on the reconstruction filter is misplaced. The attention should be on the filter at the front end of the process, not the back end.

Also, I agree 100% with Ken that the Achilles' Heel of CD is the requirement for a brick wall filter at the top of the audio band due to the low sample rate. That is something that cannot be fixed on the replay side.

This post is made possible by the generous support of people like you and our sponsors:
  The Cable Cooker  

Follow Ups Full Thread
Follow Ups
FAQ

Post a Message!

 Forgot Password?
Moniker (Username):
Password (Optional):
  Remember my Moniker & Password  (What's this?)    Eat Me
E-Mail (Optional):
Subject:
Message:   (Posts are subject to Content Rules)
Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:  Automagically notify you when someone responds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]