|
General Asylum General audio topics that don't fit into specific categories. |
|
|
|
General Asylum General audio topics that don't fit into specific categories. |
|
In Reply to: Re: Some thoughts on filters posted by John Atkinson on April 22, 2004 at 04:33:28:
> I think we were both incorrect in referring to "transients": the band
> limiting of the ADC's antialiasing filter has already done its damage
> and smeared the transient information in time.So why then add insult to injury by smearing the transient
information even *further* with another brickwall filter during
playback?> In this light, perhaps the biggest step forward in recording hardware
> has been the recent adoption of high sample rates with low-pass filters
> that are better-behaved in the time domain. When the original data are
> downsampled to the CD Standard the filter can be otimized for the
> musical demands, trading off some of the out-of-band rejection against
> reduced dispersion of transient energy. (See the Keith Howard
> Stereophile article in January for some discussion of this.)Fascinating article, and I can't wait to hear more from him. (He
hinted that he would reveal his theories in the future.)But it seems like a contradiction to me that you would accept
out-of-band energy when done in the downsampling process before the CD is
made, and yet decry it when it happens in the playback DAC.> Allowing image energy into the DAC output to approximate something like
> the original waveform might have been is inelegant.Agreed, but it may be the best possibility left to us after the
signal has been trashed by the brickwall filters during the A/D
process.The basic problem is that the sample rate chosen for CD is too low.
After looking at this issue for years, I think the only way to get
great sound from digital is to sample at least 10x the highest
frequency of interest. In this way, *no* low pass filter is needed,
either for recording or for playback.In other words, using a non-OS record and playback system sampling at
192 kHz would sound phenomenal. The problem is that there are many
knuckleheads on the pro side of things that put far too much faith in
specs. So they add brickwall filters to their high sample rate
equipment in order to achieve the best "numbers", even though it
limits the sound quality.That is, a 20 kHz brickwall filter (used for CDs) sounds worse than a
44 kHz brickwall filter (used for 96/24) sounds worse than a 88 kHz
brickwall filter (used for 192/24). But SACD (even with its plethora
of *other* problems) sounds better still. Why? Simply because there
are *no* filters used for recording, and only minimal filters used for
playback.But a non-OS, high sample rate PCM system would sound best of all. It
would have the lovely transient response of SACD, yet without the
horrible out-of-band noise problems (talk about a *real* source of
system dependent effects!), and be able to use bog-standard equipment
for recording, mixing, editing, processing, etc.But since that's not going to happen anytime soon, we have to be
realistic and live with the CD standard and try to get the most out of
it. I haven't personally tried a non-OS playback system (yet), but to
dismiss it out of hand seems unwise to me. There are as many rational
arguments in its favor as against it. The final factor of merit is how
it sounds. I would be very curious to see a full review (including
measurements) in Stereophile.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Some thoughts on filters - Charles Hansen 04/22/0409:33:43 04/22/04 (2)
- Re: Some thoughts on filters - John Atkinson 10:38:01 04/22/04 (1)
- Re: Some thoughts on filters - Peter Qvortrup 03:22:21 04/24/04 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
