Why is Garmin ignoring the GPS issue?

Thanks Carl. Contacted them and going to try the reset they recommended. If it doesn’t work then will need to send it back to Garmin which they said was the likely outcome if the reset didn’t work......

I did an official NYC 1/2 marathon yesterday. After 7th mile GPS started to show strange data , in the end based on Garmin Fenix 5 I did 14.98 miles instead of 13.1 same time

I had Fenix 3 and now have Fenix 5. I carefully examined this forum before buying them. I am happy with my devices accuracy, and I also experienced bad situations described here, even on a sunny day. GPS is not a very accurate system to start with, designed with +/-10m accuracy, that could be taken down to +/-3m and potentially to 30cm in upcoming years, but when this will become consumer-grade standard is a hard thing to say. Problems listed on this forums are typically within the range, but in some difficult areas, the error can be even higher (building, trees).

I mostly run in parks, far from bigger buildings. I see that my track gets shorter during summer (cutting down on the corners), but overall the distance is quite equal from run to run. I use out of the box settings, no GLONASS, no 1 sec recording.

The problem on this site is that people base their opinions on non-scientific observations. Typically comparing one bad track on Fenix with a very good track on the other brand. Quite often those runs are done on different days and taken with different weather conditions. To have a scientific analysis you would need to run with two devices, on the same hand, many times. Other option would be to run with just one device couple hundred times and same with the other, always using the same track. I have not seen such analysis.

All you can say is that "sometimes Garmin is worse then brand X" I am pretty much sure you can also say "Sometimes brand X is worse than Garmin". I also understand that people here are not into the scientific analysis, don't have time, resources or will to do such testing, but then they should trust Garmin, that it works as expected. After all, Garmin has this forum when you can complain about it when other brands don't have it.

Sorry Rafal, but I disagree. I have been running with F5 for almost 7 months now, on average 4 times a week. I run the same course(s) all the time - so I know the distance, the pace, the track and the deviation, and I have comparable info - even if it is just one F5. I run in parks, away from big buildings. Since I use the F5, the accuracy has been a mixed bag, to say the least, and in most cases of a level of variation between runs that makes it simply unusable for tracking purposes. I have had to switch to a footpod to get distance, pace and average pace comparability that made it possible to have a reasonable impression of what my performance is. The deviation of the mapped tracks from the actual run frequently is well over the 10m you mention. Worst of all, the situation has not improved over successive FW versions - in fact, my impression is that if anything, it has worsened. Is this scientific? no. But it is the experience of about 110 runs with the same watch.

The problem on this site is that people base their opinions on non-scientific observations. Typically comparing one bad track on Fenix with a very good track on the other brand. Quite often those runs are done on different days and taken with different weather conditions. To have a scientific analysis you would need to run with two devices, on the same hand, many times.
...

All you can say is that "sometimes Garmin is worse then brand X" I am pretty much sure you can also say "Sometimes brand X is worse than Garmin".

Sorry, I'm also do not think this is so, especially with this assertion. As I wrote I have tested it with both watches on the same run ( and also from the same "Brand", its a 235).
, Ok, ok, that's at least a halfway comparable condition, but If on several runs togehter with my 235, the 235 is "more on the realy track which I'm running" the perception is not wrong for me (First of all, these sawtooth outliers are something unusual.) and its not a statement to make Garmin bad. Especially since it is not my only Garmin device, they usually work very well :D

GPS is not always the most accurate system, I think that is well known and aware by many forum users, so everyone has their right over the personal experience.

I had Fenix 3 and now have Fenix 5. I carefully examined this forum before buying them. I am happy with my devices accuracy, and I also experienced bad situations described here, even on a sunny day. GPS is not a very accurate system to start with, designed with +/-10m accuracy, that could be taken down to +/-3m and potentially to 30cm in upcoming years, but when this will become consumer-grade standard is a hard thing to say. Problems listed on this forums are typically within the range, but in some difficult areas, the error can be even higher (building, trees).

I mostly run in parks, far from bigger buildings. I see that my track gets shorter during summer (cutting down on the corners), but overall the distance is quite equal from run to run. I use out of the box settings, no GLONASS, no 1 sec recording.

The problem on this site is that people base their opinions on non-scientific observations. Typically comparing one bad track on Fenix with a very good track on the other brand. Quite often those runs are done on different days and taken with different weather conditions. To have a scientific analysis you would need to run with two devices, on the same hand, many times. Other option would be to run with just one device couple hundred times and same with the other, always using the same track. I have not seen such analysis.

All you can say is that "sometimes Garmin is worse then brand X" I am pretty much sure you can also say "Sometimes brand X is worse than Garmin". I also understand that people here are not into the scientific analysis, don't have time, resources or will to do such testing, but then they should trust Garmin, that it works as expected. After all, Garmin has this forum when you can complain about it when other brands don't have it.

I completely agree with you. I don't know if there might be a GPS issue but i think the vast majority of complaints raised in this forum about GPS accuracy are based on personal impression rather than objective analysis of data. Of course everyone is entitled to his/her own personal opinion but if we discuss technical matters this should be backed by more than feelings. I noticed that a lot of complaints about accuracy about the distance being not accurate is based on the experience with other devices. In my experience other devices tend to overestimate distances and earn less complaints about tracking accuracy even though they are not more accuratey but err on the other side. I think this also a psychological issue rather than a mere technical one.
Also other trackers use smoothing algorithms and closing in on known tracks such as streets. Does this increase the accuracy of the tracking of your run? Well, maybe, depending on your track but it also rewards cutting corners. Does this make the GPS accuracy better? I guess not.

Considering these facts, i think there is no objective proof that the fenix 5 is better or worse than other devices without knowing the algorithms. If you compare a mere GPS based track with a algorithm enhanced one, i guess its quite clear which one looks smoother and if you add some extra distance for good measure you will end up with less complaints without being more accurate. Some people might prefer this. Personally, i do not.

Well engage your logic and you will find that too many complaints about that and other stuff, as well please check this, this one is done by scientific tests:
http://fellrnr.com/wiki/GPS_Accuracy

If you read "Methodology" you will find that this test is far away from a "scientific" test. He is taking runs and then compare the results so he is disregarding the effect of the different algorithms and how the obtained data is being processed and which further informations are being considered. Looking at the table you will find "precisions" which are simply to good to be archieved by GPS. Setting aside the Footpods you will find an average distance error of 2.9m/Km for the Polar V800. This is way above the precision of a single data point of standard grade GPS under ideal circumstances. Again i am not saying that the GPS of Fenix 5 is better or worse than other devices but there is a lot more behind this than merely the precision of GPS data points.

Fair enough, if you have made the experience that for your tracks and for your usage of this device the GPS is not what you have expected and your opinion is that it performs worse on your courses compared to your previous watches thats up to you and that is something you are perfectly entitled to. No one can second-guess your personal experience and opinion on this device. But this is just yours, other users may have other experiences of this devices and based on their expectations and requirements and a lot of users are more than happy with their fenix 5.