[time-nuts] How are iPhones' clocks set under LTE?

[Tim Shoppa]

Unlike CDMA (where time distribution was an automatic part of the low-level
protocol) I suspect the time displayed on many modern phones is not set by
the telephony synchronous protocol but rather by IP-over-Wifi packets.

And the packets don't seem to do a very effective job keeping the clock ont
he phone correct. My employer gave me a Nokia Lumia 630 "Windows Phone" and
its clock has always been off by at least a minute.

There was a few years ago, a very nice article about the effort to repair
the clocks in clock towers in many cities. What rang most true to me was
"if you visit a town they can't even keep the clock correct, who else knows
what else is wrong there?".

Tim N3QE


On Sun, Aug 3, 2014 at 4:37 PM, Brian Garrett <garrettbrian1960 at gmail.com>
wrote:

> Hi all,
>
> First “time”r here.  This may not rank up there with your degree of
> time-nuttery, but I haven’t been able to get an answer elsewhere.  Recently
> I was discussing the issue of how the different cellular providers set
> their time, and I told him that I’d read that CDMA phones and towers have
> to have their clocks synced to GPS as part of the protocol, whereas GSM
> phones do not, and can theoretically be set to wall time, and thus phones
> on networks using CDMA would have atomic accuracy all the time since what
> they were getting was as good as GPS.
>
> Well, obviously I was pathetically behind the times.  Most everybody these
> days including Verizon, which both I and my friend have now, uses LTE , as
> you know.  I have looked all over for info as to what LTE’s time-setting
> requirements are, as implemented by Verizon, but I’ve not seen discussions
> of it anywhere.  I’ve seen amusing anecdotes over what can happen if your
> Android isn’t set to receive the network’s time, or what can happen to your
> phone’s clock if you live near a time zone boundary, but no discussion of
> how time dissemination is handled in-network.  I know my iPhone can be, and
> usually is, 2 or 3 seconds fast or slow when checked against an accurate
> reference clock, so I’m thinking they can just use wall time like GSM did.
>
> Has this been discussed on the list before?  I haven’t seen anything in
> the archives, and no-one at Verizon that we of the unwashed masses have
> access to will know the answer  Pointers, anyone?
>
>
> Thanks in advance,
> Brian
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>