 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
68.49.187.199
In Reply to: Re: Ack! dAck! 2.0 - HDCD?? posted by Todd Krieger on May 2, 2005 at 19:03:11:
It may be better with an HDCD decoder, but with redbook, can we assume that the control bits are pseudo-random noise....it's better to assume that these bits degrade the redbook signal...any superiority observed in redbook playback should be attributed to the mastering.
Follow Ups:
"It may be better with an HDCD decoder, but with redbook, can we assume that the control bits are pseudo-random noise"The HDCD trigger code could be integrated with pseudo-random noise, but I don't think it's in itself pseudo-random noise.
Hi Laznos,I can't imagine that HDCD would degrade the redbook signal particularly much if it is truly pseudo-random. It would appear to us as real random dither. In most systems, the analog noise floor is quite a lot higher than 16 bits (typically 14 bits is about right) due to power supply limitations, so being able to hear even past the dither is rare anyway in most commercial products. On some products with terrifically low noise floor (e.g., battery-powered stuff), you can hear down to 16 bits but if the dither is encoded properly I don't think there's a big problem with hearing the encoding. Again, the algorithms are proprietary so we can't know for sure.
I'm not very impressed by the idea of HDCD because I don't think it really adds all that much useful of information to the datastream. If you think about the actual amount of data necessary to generate an extra 4 bits of resolution, even for the rare conditions in the waveform where we need the dynamic range extension, one would have to have a huge amount of data compression to pack it in. In order to get pseudo-random encoding *on top* of the dither already present, one needs quite a lot of samples to distribute the data over. So, the compression would most likely be extremely lossy (e.g., the resolution is very selectively applied, and much of the resolution enhancement comes from interpolation, not real data).
Additionally, I'm not convinced that 20 bits is necessary in the vast majority of recordings anyway (let alone 24); recording conditions limit what you hear in many cases. Lots of masters don't have 20 bits of resolution so what you are hearing will be the equivalent of the tape hiss of yore, when the noise floor of your vinyl system exceeded the analog noise floor of the recording equipment :). People who like non-oversampling very much dislike the effect of the sinc kernels used in FIR filtering, more so than 4 bits of dubious utility so HDCD hasn't been a big priority for us. Seems it's not a huge priority for a lot of other high-end companies too... big shots like NAIM don't include it either.
-Chris
Ack! Industries
Very articulate and interesting post. However, the comment about Naim not offering HDCD is incorrect. See link, below.
Thanks for the correction. I had a CD5 for quite some time and it did not do HDCD, and I seem to recall that their other players of that era didn't incorporate it either. The chipset is not free in any case - if a manufacturer doesn't have pressure from customers to use it (meaning there's not a widespread perceived value) then there's good incentive not to include the feature. A few years back it appeared flat earthers didn't care much for the resolution enhancement stuff on the market.-Chris
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
