There’s been plenty of chatter in recent weeks and months about whether this weekend’s upcoming GPS Week Number Roller (WNRO) will impact various GPS devices. In short, on April 6th, 2019 some older devices could effectively reset their knowledge of date/time, causing things like timestamps, sunrise, sunset, tides, and anything else reliant upon date to be off.
The reasoning comes from how GPS devices first transmitted date and time using a counter that rolled forward a maximum of 1024 weeks, or 19.7 years. Garmin’s site on the issue explains it in the most concise manner I’ve seen to date:
“The GPS system is world renowned for its ability to provide accurate and reliable positioning and timing information worldwide. The GPS satellites transmit to users the date and time accurate to nanoseconds. However, back in 1980, when the GPS system first began to keep track of time, the date and time was represented by a counter that could only count forward to a maximum of 1024 weeks, or about 19.7 years. After 1024 weeks had elapsed, this counter “rolled over” to zero, and GPS time started counting forward again. This first rollover occurred in August of 1999. The second rollover will occur on April 6, 2019.”
With this date approaching, there’s been plenty of reasons to fear the world is ending – however it’s mostly due to tabloid media playing it up. The norm I suppose. Certainly, there will be some devices that are impacted. Hopefully not ones you’re using for anything important. But I’m not here to talk about those.
Instead, I’m here to talk about fitness and sport devices.
I’ve done a super quick check with the major sport/fitness vendors to validate if any issues are expected. In the case of some companies, like Garmin, they have specific support pages listing some brief tidbits. However, they used terms like ‘vast majority’, which in my experience leaves just enough of a gap to fall through. Sorta like if the recipe says ‘The vast majority of the time the cookies come out fine’, you’ll immediately ask: And the other times?
So I circled back to Garmin on this, and they confirmed that all fitness/outdoor devices are compliant/compatible. It sounds like the ‘vast majority’ part is in reference to potential non-fitness/sport products they make (remember they make automotive/marine/aviation/etc) devices.
Then, I decided to hit up a few other companies. Here’s the quick and easy list:
Fitbit: All sport/fitness/outdoor devices confirmed compatible (Surge/Ionic only units that use GPS). No issues expected.
Garmin: All sport/fitness/outdoor devices confirmed compatible. No issues expected. Company page info here.
Polar: All sport/fitness/outdoor devices confirmed compatible. No issues expected.
Suunto: All sport/fitness/outdoor devices confirmed compatible. No issues expected.
TomTom: All sport/fitness/outdoor devices confirmed compatible. No issues expected. Page info here.
Next, is the one player that I can’t find any info online about. Though I wouldn’t expect any issues – mostly due to how recent their hardware is in the GPS realm.
Apple: I’ve searched high and low, and found nobody talking about Apple devices or concerns.
For the most part, if you have any other newer device – it’s pretty unlikely to be impacted. Devices most likely to be impacted are much older devices, and in the cases of GPS watches in particular, these don’t tend to be around as much (life-span wise) as larger things like car navigation systems.
Companies like Garmin note that even in rare cases where a non-fitness unit wasn’t compatible, it wouldn’t affect positioning (where you are). Just the dates. And, if you have a drone (from DJI), you’re good to go as well.
With that, you can now go and run or ride on Saturday without any worries about being unable to upload your file to Strava afterwards. All shall be well.
Thanks for reading!
It’s like Y2K all over again!
Party like it’s 1999…err…2019!
Y2K – don’t forget – 29 february 2000 was often a true bug 😉
Number of years ago with my original Fenix, occasionally my track would show as jumping across the world for a few minutes then back. Digging in I found the the date would shift to April 6th 2019 and increment from there for a few minutes. I realized that the time was coming through as 0 so being interpreted as the rollover date. Skipping the fact that the chip/watch/software should have handled the huge jump in date/time rather than assume it was correct, it did point out that the watch would likely handle the actual rollover fine. Not that it made up for the amount of time it took me to manually fix the GPX file!
It’s gonna be interesting to see what happens to the Motoactiv!
does yours still work? I can’t get mine to charge-looks like it’s a plug issue not a battery issue-the light blinks when I plug in the usb, and comes back on if I wiggle a bit, but won’t charge long enough for me to turn mine on. I would like to see my old runs on it. I stopped using it when I got my Fenix2
I had no idea this was even a thing.
That’s why we keep Ray around…
On the upside, Ray can finely clean out his multiple drawers of old GPS devices!
Let’s not get crazy now…
Yes, but it will only have an issue if you are using a 20 year old sat nav. Everyday the atomic clocks on the satellites are updated twice with reference to the master clock. This means they are all at the same time. Because they are in space and travelling at speed, time runs slightly slower for them. This causes position error. Because older GPS devices have their own week counter, they could return to week 1 tomorrow which will cause them to be a bit out of position. Most devices now take the information directly from the GPS signal, rather than counting the weeks.
Um, no, I think you may be confused about how the satellite time is conveyed to the receiver. The GPS signal (the original L1) contains the atomic clock time encoded as the GPS week and an offset within that week. The receiver does indeed take that information while it’s locked to GPS, only reverting to counting for itself when the GPS reception is off, but the problem is precisely that the time sent by the satellites rolls over.
Most units resolve the ambiguity in the week by assuming that the actual week must be later than the firmware compilation date, so yes it is generally only an issue if you have a very old unit.
Also, if you have a receiver that handles L2 as well, the week number in that message takes more than 100 years to roll over, so there is no issue, but we’re talking here about watches where there are no dual-band units available yet, they’re all exclusively using L1 for GPS.
Galileo has a 12-bit week counter which takes nearly 80 years to roll over, so if you’re using Galileo+GPS the device’s job should be a bit easier.
Obviously not everyone designs their products to last, but it seems like the fact that this has already happened once before would mean that most designers are aware of it and can mitigate it one way or another.
Now, the 2038 problem will be interesting, but again it seems like enough people are aware of it that it will mostly be theoretical.
Using my Garmin Edge 520+ I can select the system/mode I wish to use (GPS, GPS + GLONASS or GPS + Galileo). Which system(s) works best for N. America? Western Europe? In the past, I’ve selected GPS + GLONASS. But given all the news, I’m a bit worried the Russians have begun following me 😉
Ignore the propaganda about Russia and GPS in the news. It’s just propaganda. The USA and UK certainly do the same things with regards to messing with GPS signals, it just sells more newspapers and generates more fear if they say Russia.
This is a good strategy for news in general – think of it from the perspective of the other nation and you’ll quite often realise it’s all propaganda BS.
All of them are worldwide, so it doesn’t matter. (You never know whose window you want to send a cruise missile through.) I’m not sure that Galileo is completely operational yet.
great info as always, Thanks!
I’m pretty sure we should be OK unless there’s some old Cobol code in my 935. 🙂
No Galaxy watch active review today?
The day isn’t over yet. 🙂
today – first time ever – my garmin forerunner – got wonky and gave me a strange watchface. after a restart everything is fine. coincidence?
Couldn’t say if it’s coincidence or not, but that face looks like part of the self-test routines. Long time since I did that on what looks like a 620, but there’s a key sequence that brings them up, which I don’t think you’d do by accident (it’s the same as in this link for the 220
Well, maybe coincidence too, but I had a lengthy fight with my Fenix 2, to get it back to work after connecting it to a computer.
It froze with some wonky display and after reboot via light button stayed on start screen forever.
Well, connecting with the “software loader” (down, back, start pressed) and repair of the file system with Windows sortet out the problem, so it could have been really a coincidence…
So one strange thing after the change-over
On my Garmin 935 – All my records were reset.
So my Run on the 6th now has my ‘fastest 1km run’ (about 90secs slower than training Peaks) and my ride on the 7th had new everything records.
That seems to be coincidence only.
Had this strange behaviour after last FW update.
Several times I had new fastest mile / 1 km despite not becoming any faster. 🙁
Old Forerunner 305 still works fine. 🙂
no issues at all on my Fenix3 last Sunday.
Right now i’m reading about tv signal blackouts due to repeaters gps chipset update. just an italian thing?
My Strava recorded activity (cycle riding) with Strava GPS shows the date 09-07-1963 on my mobile phone for last 2 days. I have to manually fix it.
But at least the fitness devices were ok.
link to arstechnica.com
Own an older Auto GPS unit that was struggling recently to latch on to satellites and maintain a connection. Had been updated thru Garmin so I figured it was due to both being old and in a newer vehicle with possibly less direct sight to satellites. Monday it fired up and has locked on perfectly since. Go figure.
I gave my old Edge 500 to a buddy when I upgraded to my 1000. His first ride this April showed as happening in 1999!