In Reply to: RE: Agree, and if I might be so bold as to add... posted by jkeny on January 15, 2016 at 09:22:07:
Actually, one aspect of the incoming signal is not reclocked at all. That's the 8 kHz isochronous packet rate of the audio packets. This is generated by a clock in the root USB hub and all of the intermediate hubs are slaved to this clock. So if there is jitter on this clock it will be present on all packets, regardless of the regeneration. Actually, each hub in series will increase the jitter of the packet start due to various technical issues involving squelch circuits, start of packet detection, and details of how synchronization is done between asychronous clocks. (We discovered all of these issues in the repeaters for the 10 Mb/s Ethernet, which only worked because there had been a low limit on the number of repeaters that could be connected in tandem. Ring networks had many more nodes strung together and had much more complicated jitter problems.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- RE: Agree, and if I might be so bold as to add... - Tony Lauck 01/15/1613:57:34 01/15/16 (7)
- RE: Agree, and if I might be so bold as to add... - jkeny 16:11:55 01/15/16 (6)
- RE: Agree, and if I might be so bold as to add... - Tony Lauck 16:48:57 01/15/16 (5)
- RE: Agree, and if I might be so bold as to add... - jkeny 16:57:44 01/15/16 (4)
- RE: Agree, and if I might be so bold as to add... - Tony Lauck 17:05:58 01/15/16 (3)
- RE: Agree, and if I might be so bold as to add... - jkeny 17:19:04 01/15/16 (2)
- RE: Agree, and if I might be so bold as to add... - Tony Lauck 17:46:19 01/15/16 (1)
- RE: Agree, and if I might be so bold as to add... - jkeny 18:19:40 01/15/16 (0)