In Reply to: Re: you are right . . . posted by Martin says on March 11, 2002 at 10:57:39:
This is apparently the assumption in the IAR article. But it is not correct. There is no sample-and-hold that waits for a number of bits to do an average. The recent density is recomputed in a sense at every new bit. The bandwidth depends on specific implementaion. You are right that the bandwidth is nowhere close to half the sampling time, as it is in a PCM system. That is why the sampling time is so high (64*44.1 kHz). The effective bandwidth of SACD is probably about 50 or 60 kHz. I doubt it is as high as the 96 kHz from 24/192 PCM. But then again, does it matter to humans?
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- not quite correct - tunenut 11:11:04 03/11/02 (0)
- I wasn't implying that there was a "sample & hold" with DSD . . - Martin says 11:29:53 03/11/02 (0)
- So.................. - waVeman 17:16:40 03/11/02 (3)
- the advantages are implementation issues - tunenut 18:40:41 03/11/02 (2)
- Re: the advantages are implementation issues. . . - Martin says 07:36:52 03/12/02 (1)
- . . . And for those who didn't know, "multibit" DACs use PCM {n.t.} - Martin says 07:44:18 03/12/02 (0)
- agree with you, mostly - tunenut 13:15:14 03/11/02 (0)
- Think of it this way... - Ted Smith 12:35:56 03/11/02 (0)