 
|
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 |
In Reply to: I wasn't implying that there was a "sample & hold" with DSD . . posted by martinsays on March 11, 2002 at 11:29:53:
....what is DSD's advantage over hi-rez PCM then?1) The gentle filter DSD uses? Can't PCM, especially 192/24, also use a gentle filter?
2) Can DSD detect more of a musical sine wave's small hills and peaks, resulting in better detail from the playback loudspeaker?
Follow Ups:
The problems with a PCM converter are in the precision required to implement it. A 24-bit converter that represents volts must meaure tenths of microvolts. This must be precisely calibrated and the slightest drift over time will throw it off. In addition, different bits represent different voltage steps, so error from sample to sample can vary greatly- another ugly implementation issue.Sigma delta modulation gets away from these problems, but it leads to a high noise floor, so noise shaping needs to be used.
In short, both encoding schemes are capable of excellent theoretical performance. This is why theoretical arguments are of little real importance. The final audio quality will be based on the quality of the implementation, from recording to final playback.
Point 1:-> > The problems with a PCM converter . . . . .
Response:
PCM has generally required electrical skill to do the theory justice. That’s why, in the early days, they had to start with 14-bit DACs with oversampling. Then came the full 16, and then 18-bit CD DACs for CD players. But when designed and built properly, then end sonic results of multibit DACs were well worth it.
In the early ‘90s, we had 1-bit (bitstream) DACs in CD players — specifically for cheapness of manufacture (although it was marketed as some sort of technological breakthrough). I’ve had CD players with both systems. On my 1-bit Technics MASH player, it had no fine detail at all during transients — exactly as described in the rant piece on IAR — so I had to replace this player very quickly. For this I used a one-bit Marantz CD player, but I bypassed its own DAC and used instead its multibit coaxial SP-DIF output to an Audio Alchemy multibit DAC. Needless to say, it cured the problem completely.
Ironically, we now have a supposed audiophile product (SACD) using a 1-bit system derived from a concept which was conceived specifically for cheap CD players.
Point 2:-
> > In addition, different bits represent different voltage steps, so error from sample to sample can vary greatly . . . .
Response:
This has not been the problem the that some people make out. Laser error correction is very good these days. If it wasn’t, then all our CD-ROMs would not work in our computers. Just imagine if only one or two bits were corrupt on a CD-ROM: if so, then your software loaded from it would crash your PCs — if it was not bit-for-bit perfect!)
Regards,
.
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 © 2025, Audio Asylum, all rights reserved ]
