[time-nuts] Msg to N.Z. time nuts

[M. Warner Losh]

In message: <C6389103.7E0F%James.P.Lux at jpl.nasa.gov>
            "Lux, James P" <james.p.lux at jpl.nasa.gov> writes:
: 
: 
: 
: On 5/19/09 1:15 PM, "Russell Rezaian" <rrezaian at motorola.com> wrote:
: 
: > At 12:58 PM -0700 2009/05/19, Hal Murray wrote:
: >> USB has a bad reputation, but I think it's way way overblown.  Yes, it's
: >> polled, but that polling is done in hardware and the time scale is 1 ms.  If
: >> you are satisfied with an accuracy of a few 10s of ms, USB works fine.  The
: >> problem is the GPS unit.
: > 
: > And I can confirm that for my very un-scientific experiments I do get
: > a pretty consistent time sync around 1 MS plus or minus mark from a
: > USB serial port adapter in best case conditions.  That's with no
: > special work on a pretty stock consumer grade computer with an off
: > the shelf USB to serial port.
: > 
: 
: I would expect that the basic "frame" timing on USB (without actually
: digging out my USB documents, because I'm lazy) is on the order of 125
: microseconds (e.g. 8kHz, to support a 8ksps audio stream, just like IEEE1394
: does), but that might just be a latency jitter bound.

Well, Yes and No.

Yes, you can get frames that fast in isochronous mode, but most host
adapters have buffering and queue things up to be dealt with on ~1ms
boundaries.  And serial port modem control pin status change messages
aren't isochronous transfers.

Warner