In Reply to: one could imagine a digital XO posted by Mart on July 1, 2001 at 05:46:43:
I skimmed through Tripath's application notes and there's no hint as to sample rate or sample size. The documentation is rather vague; they consider the details proprietary.The cancellation of jitter when a common clock is used for both A/D and D/A is possible only if the system is not pipelined (unlikely). All DSPs use pipelining, meaning that there may be hundreds of clocks of delay between the input and output. Using a common clock won't help since the input and output are skewed in time.
A digital XO can be a very good thing since zero phase shift is possible. This is realized by applying the same pipeline delay to both high-pass and low-pass sections. I think the problems people encounter with "subwoofer lag" are mainly caused by phase shift in the crossover. At a crossover frequency of 50Hz, one can expect several milliseconds of delay, enough to upset the rhythm of the music. Digital XOs can fix this.
Ed
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- sample rates, jitter, and digital XO - EdG 07/1/0110:12:03 07/1/01 (10)
- jitter, TriPath sw. frequency, energy storage & DSP filters - Arbelos 18:59:03 07/1/01 (8)
- DSP filters - EdG 20:57:50 07/1/01 (7)
- not clear how you decide that the filters are at fault, care to expound? <nt> - Mart 00:25:33 07/3/01 (1)
- its a guess - EdG 09:40:44 07/3/01 (0)
- Re: DSP filters - Arbelos 04:30:38 07/2/01 (4)
- beauty & the beast - Mart 00:20:59 07/3/01 (2)
- Minidisc does something similar - EdG 09:09:12 07/3/01 (1)
- you would think it'd employ fuzzy logic - Mart 09:55:57 07/3/01 (0)
- steep filters and CD - EdG 10:54:35 07/2/01 (0)
- true but ironic - Mart 10:35:51 07/1/01 (0)