[time-nuts] French Time offset
Lux, James P
james.p.lux at jpl.nasa.gov
Wed Mar 18 14:21:03 EDT 2009
> -----Original Message-----
> From: time-nuts-bounces at febo.com
> [mailto:time-nuts-bounces at febo.com] On Behalf Of Michael Sokolov
> Sent: Wednesday, March 18, 2009 10:56 AM
> To: time-nuts at febo.com
> Subject: Re: [time-nuts] French Time offset
> "Lux, James P" <james.p.lux at jpl.nasa.gov> wrote:
> > Things get done for funny reasons: After all, the physical size of
> > France is why ATM "cells" (packets) are 53 bytes (48 byte payload)
> > instead of either 32 byte or 64 byte payloads.
> Huh? Can you please elaborate? I thought the 48 octets of
> payload were a compromise between the camp that wanted 32 and
> the camp that wanted 64.
> who has implemented the whole ATM/AAL5 stack in software
> (looking at cell headers and all) for SDSL.
53 bytes was the longest you could do and not require echo suppression for a path across France. The US is so huge, they had already bit the bullet on needing echo suppression, so they wanted longer packets, to reduce the header overhead (as a percentage). There were also some "timing jitter" requirements, and how much buffering was considered reasonable for "real time" response in voice applications.
More information about the time-nuts