New 2021 GPS Accuracy Issue Impacting Some Garmin, Suunto, other GPS Devices

Update: As of Sunday morning this should be resolved, however, you’ll want to ensure your GPS has fully synced before heading out, and it doesn’t hurt to give it a few extra mins outside before pressing start.

Just a super quick post to provide a tiny bit of context on an issue some (but not all) folks are seeing where your recorded GPS track is offset by upwards of a hundred meters or more. In general, the pattern of your route is correct, but it may be displaced to one side or the other. However, in many cases by the completion of the workout, it sorts itself out. In other words, it’s mostly a one-time issue.

The issue largely appears to be impacting companies that leverage the Sony GPS chipset in their devices, which is basically all new Garmin devices in the last few years, most new Suunto devices in the last few years, all Polar devices in the last few years, all COROS devices, the Wahoo RIVAL, and more. More or less everyone except Apple.

Very rough list of impacted devices (I’m missing a ton here, these are just the most popular ones)

– COROS GPS watches (need to double-check original Pace though)
– Garmin Forerunner 45/245/745/945/Fenix 6/MARQ/Vivoactive 3/4/Venu
– Garmin Edge 130 Plus 530/830/1030 Plus
– Polar Vantage V/M/V2, Grit X, Ignite
– Suunto 5 & Suunto 9
– Wahoo RIVAL

Again, I’m missing a boatload here, but that’s the big-ticket ones off the top of my head.

All these companies have switched to the Sony GPS chipset since around 2018, with Polar and Suunto initially jumping onboard, followed quickly thereafter by Garmin, COROS, and more. The Sony GPS chipset is widely used by these companies because of the power savings delivering longer battery life.

But – that’s all besides the point, and not the cause of the issue. The issue has to do with the ephemeris data file, also called the EPO file (Extended Prediction Orbit) or Connected Predictive Ephemeris (CPE). Or simply the satellite pre-cache file. That’s the file that’s delivered to your device on a frequent basis (usually every few days). This file is what makes your watch near-instantly find GPS satellites when you go outside. It’s basically a cheat-sheet of where the satellites are for the next few days, or up to a week or so.

Your watch or bike computer automatically gets this file via Bluetooth Smart from your phone, WiFi, or USB, depending on how you connect your watch. Most companies deliver it anytime your watch syncs and needs a new version. So from your side, you never do anything – it just quietly happens in the background.

Right now, the data in this file is wrong, and thus the data your watch uses for those first few minutes is also wrong – leading to the offsets. You can see an example of this below, from one DCR reader:

And another:


As the watch sustains its connection with GPS satellites over the length of the workout, it’ll generally re-correct itself as it pulls in updated data from above. Most people are reporting that it’ll resolve itself either by the end of the workout, or the next workout. You can see that here from yet another DCR reader:

To help resolve this, you can simply stick your watch/bike computer outside and let it record an hour workout, then just discard it after that. It’ll have received the required satellite information to likely correct itself.

But it won’t impact everyone. For example, I tried today with two Sony-based watches (Garmin FR745 and FR945) to get it to lay down an incorrect track – and it wouldn’t. I suspect this is because in my case the watch had been bumped on a few times while playing with my kids yesterday and today, so my guess is it had already re-downloaded what it needed to be correct out of the gate. I also tried with two non-Sony chipsets (Fenix 5S and Apple Watch SE), and didn’t see any problems either.

In talking to Garmin today, they’re working on a fix for the issue, that they believe will most likely be that they just update the satellite pre-cache data from the server, which in turn your watch will grab just as it always does. But first they (and others) are working to sort out what exactly is wrong with the data that’s causing this.

In talking to another person in the industry dealing with the issue, they noted that technically 2020 had 53 weeks, and this is the 53rd week. As such, the suspect Sony data file issue might actually be tied to that complexity.

See…just when you thought you escaped 2020, it’s still there waiting for you at the start of your run.

With that – thanks for reading!

DC Rainmaker:

View Comments (215)

  • Hi there,
    thx for this information. I had it today, that my fenix 6 recorded the track a few hundreds of meters off my real track. Hope this will be fixed soon!
    br Michael

  • (Retired military, used GPS since the early 90's)

    We had on our devices something called the "almanac" that was received via a very sloooowww transmission rate direct from the GPS satellites. This was the EPO/CPE you are referring to. We'd have to leave the unit out-and-on for an hour or so and it would receive the data (early 90's). This was done every few months as I recall. Modern units, it was never an issue. I thought that process was part of the basic GPS standard. I'm surprised that is not incorporated into all these consumer devices.

    • It is, It can get it from the GPS signals, but it still says that it's Expired.

      I tried deleting CPE.bin, power cycled the watch, put the watch on hike, waited 6mins.. Now I got new CPE.bin file, which was still "Expired", and it's pretty small 2kb, compared to 935's EPO.

      Which makes me think that it's some bug in the Sony's GPS firmware.

    • Your supposition about the Sony firmware being the culprit seems legit. I noted you mentioned the 6-minute cycle for the modern download so I've no doubt your refreshing the file. Interesting.

    • @cybirr - just to make you feel better about consumer devices, they can download the almanac via GPS. They just do it through the phone so you don't have to wait the whole 12 min to download it via GPS.

    • Thanks for that link. A couple firmware updates ago Garmin removed the GPS accuracy data field. The signal strength is still there but not the accuracy. I thought it was a nice-to-have data field. Thanks for offering this app.

    • Accuracy is still there,
      Press the activity circle,
      then press the screen to see the house and the 2 arrows,
      then press the right arrow until the page opens from which you will have access to the satellite view

  • Thanks! Apart from today and yesterday, I had this problem one time before with my FR 245, on 20 September 2020. One of my Strava contacts had the same issue around that time, but I didn't read anything about it anywhere else so may be purely coincidental.

    • Thank you so much for posting this! In the last 16 months or so using Polar Ignite, I had this issue exactly once, which puzzled me. Just checked after reading this, and it was on Sept. 20th!!! Starting at 3:38pm (Brussels time), and "was back on track" after 14min30sec (ie 3:52pm).

    • Thank you too! That day, my 30 minutes run started 2 minutes before yours but didn't correct itself. It was near Amsterdam so probably very bad gps reception in the entire Benelux :). My friend's faulty run was on 16 September, however. He also has a FR245.

  • Suunto 9 user here - first occurrence for me was at 2am UTC on 31/12. I've been leaving my watch on a window sill for 15min before every run the last couple of days.

    • Realistically, no.

      I mean yes, technically one could painstakingly re-construct it manually second by second with low level file editors. But it'd take you a long time and there's a good chance it's be flagged by Strava's file checks anyway.

    • I exported to GPX and imported into QGIS to see what was going on before I realised this was a wider issue.
      I found the start of my run was completely wrong, couldn't even find an offset that matched. The middle of my run was offset to the north west. The last part of my run was offset to the north only. There were also a few bits between the different offsets that were pretty weird, i.e doing little u-turns and loops.
      So in my case it wasn't really worth the effort to fix especially since the distance seemed about right.

      If your track just has one offset, it's pretty easy to fix in QGIS and reimport to Garmin.

    • That script worked for me. My run was out but the same offset for the entire activity so the script was perfect once I calculated the right offsets. Had to delete the original from Strava to avoid getting flagged as duplicate.

    • FYI - if anyone has an activity that they need to move (i.e. the whole thing is just offset), I was able to drag mine to the right place using the WTracks tool. Link here https://opoto.github.io/wtracks/ - it's a simple load up of your gpx file, drag to the right location, and save. You can also do other editing etc if needed, but it was perfect for my needs.

  • Yeah I had an "offset" track for about 15 mins yesterday (1st Jan) on a 945. I stopped for a moment then to check where was heading and maybe that helped correct things and was OK again after and fine today.

  • Just to feaffirm what you already stated, i had no issue with my 2016 Suunto Spartan Trainer. It does not have the Sony chipset.

  • Not understanding how calendars work is a pretty weak excuse for a company that’s been producing GPS units for 30 years. I’ll *maybe* give Suunto, COROS a pass. But that’s why I’m supposed to feel good about paying more for a Garmin device, right? That they understand how GPS works?

    Another example of “Garmin’s Biggest Competitor Is Their Own Software Instability”.

    • Except Garmin had no control over this. This is literally the file from Sony. Just like everyone else (Wahoo/COROS/Polar/Suunto/etc...), they just pass it along.

      There are many things one can assign blame to Garmin for, I think this one is probably a stretch.

    • Without starting a whole big thing, I’m not so sure Garmin is blameless. Why isn’t the FR 745 affected? And more to the point, they made the decision to switch to the Sony chip, so they bear some responsibility for the downstream consequences. I have been really unhappy with the Sony GPS accuracy even before Jan 01. Today I took my old FR 230 for a run and got the cleanest GPS track in years.

    • The 745 is affected, as noted, it's not a comprehensive list. Just skipped my mind on the Garmin side, there are boatloads more there that use it (like, dozens).

    • FYI: Not a single user report of this problem on the FR 745 Garmin Forum. And a lot of people positively stating no problems with it on the Reddit Garmin forum.

    • I think the Garmin should have control of the components it uses though. In the end if the GPS watch does not work well, users do not care whether the problem is with the Garmin or the Sony chipset. They bought a Garmin smartwatch and Garmin should ensure that the components it uses works well

  • Hi there, this definitely affected my COROS Pace v1 this morning and to a lesser extent yesterday. No satellite updates in the app so it's likely the bug you noted. Here's hoping tomorrow's run is better. Did result in some Strava CR I definitely didn't earn though! :)

1 2 3 15

By continuing to browse the site you are agreeing to the use of cookies