Devs at GARMIN, do you test software before releasing it?

On my Forerunner 955, I just tried Run activity > Navigate > Use Map and selected a random point 1.54 km away. (I live in the midtown area of a city, and my surrounding area is mostly a grid of North-South / East-West roads). After 4 minutes (which I feel is more than enough time), the watch was still calculating the route. Maybe one of these days I will wait for a route to finish calculating, but not today. Even if it was going to finish within 5-10 minutes, what would that imply for a more realistic route (e.g. to a point 5 to 10 km away)? Would that take hours? Days?

Maybe it works for other ppl, in other scenarios, but it doesn't work for me.

Imo, the average user will either:

- not use this feature at all

- try the feature and give up immediately if they have a similar experience to mine, telling themselves "ig it's not supposed to work" or "I probably did something wrong"

You "probably did something wrong" by posting about your FR955 in a fenix 7 Series forum...

Did you have a GPS fix when you started route calculation? Because not having a GPS position will indeed make it calculate forever because it doesn't have a starting point. I am asking because that was also my mistake when I first tried this feature. 

Also I have never tried it in a city, could be that the many intersections make it so slow. I will give it a try next time I am in a city (might take some time though).

I would not expect that there is a difference between forerunner 955 and Fenix 7. 

gaijin said:

You "probably did something wrong" by posting about your FR955 in a fenix 7 Series forum...

It shouldn't matter. The code is most likely shared between the two watches. 

By the way, I have tried this feature on Fenix 7X earlier today - a 10 mile course generation - it took several minutes.

silentvoyager said:

It shouldn't matter. The code is most likely shared between the two watches. 

Yeah, that’s exactly my rationale. It’s been known that since 935 was released, Fenix and Forerunner watches have used a similar codebase and UI. As a matter of fact, this is basis of complaints when Fenix has some behavior that Forerunner users want, but don’t get. (One example is the “map only” activity touch mode, which, until recently, was on Fenix/Epix but not Forerunner. Another example is haptics on Fenix/Epix, which have recently come to Forerunners.) A counterexample is that the 945 LTE got some Fenix behavior in an update which actually made it worse. (I won’t go into details, but it has to do with the fact that Fenix 7 supports transparency and touch while 945 LTE does not.)

Plus this is obviously a thread about the perceived quality of garmin software. Should we create the exact same thread for every single Garmin product? Bc obviously it’s a company wide problem and not confined to any single device.

#Matthias# said:

Did you have a GPS fix when you started route calculation? Because not having a GPS position will indeed make it calculate forever because it doesn't have a starting point. I am asking because that was also my mistake when I first tried this feature. 

To be fair i was indoors when I most recently tried it, but it obviously used my last known GPS location to give me the initial position on the map which was close to my current location. If it needs a real GPS fix to calculate the route but not to show me an initial position, I would also consider that a bug. It obviously has a starting point in mind since it’s able to tell me that the destination is 1.54 km away.

It’s not like Connect (the site or app) needs a real GPS fix to calculate a route, it just uses whatever starting point you ask for.

I’ll give it a shot with a fix outdoors though, thanks! I’m curious to see what will happen. (I don’t remember if I ever tried it outside before).

Even so, I have heard other complaints about route generation taking an unreasonably long amount of time. Either way, my general point is that there’s more than one Garmin feature which looks great on paper but is not worth using in practice. Contrast that with Apple’s supposed policy of not adopting a technology or introducing a feature or product until they can get the usability right. They’re not perfect, but they def have a different philosophy than Garmin.

#Matthias# said:

that was also my mistake

Not really your mistake if the software can't do it it should tell you "please select starting point" and not calculate indefinitely regardless.

FlowState said:

it obviously used my last known GPS location to give me the initial position on the map which was close to my current location.

Yes that was exactly my thought when I first tried calculating a route while sitting in the living room when I first got my Fenix 6x back then. But when I was outdoors, it magically started to work quite well. Not sure if it would qualify as a bug, but what it is for sure is a bad user experience, because the user does not know that the watch is actually waiting for a GPS fix and not actually calculating the route. 

FlowState said:

"I probably did something wrong"

KrapfenKringel said:

#Matthias# said:

that was also my mistake

Not really your mistake if the software can't do it it should tell you "please select starting point" and not calculate indefinitely regardless.

There it is. Software doesn’t work right and/or fails to guide the user properly, and the user blames themselves (or other users blame the user).

We shouldn’t blame ourselves when a product we paid good money for doesn’t work as expected.

Displaying “Calculating Route” forever is not a substitute for failing gracefully. I didn’t know why it was doing that, I just assumed it was broken. As far as I was concerned, it was broken.

I still think 10 minutes for a 10 mile route is ridiculous. Elsewhere somebody claimed that the CPU used in Garmin watches is irrelevant, as they claimed Garmin CPUs are underpowered/underclocked on purpose (which is true), garmins don’t need fast CPUs for anything (hard disagree), and the only significant performance bottlenecks come from storage. 

Either the algorithm Garmin uses is lacking and/or the CPUs are underpowered. I also think the very welcome UI responsiveness/speed improvements over the years have to be at least in part due to faster CPUs (especially since Garmins haven’t even had GPUs until the most recent generations).

#Matthias# said:

Not sure if it would qualify as a bug, but what it is for sure is a bad user experience, because the user does not know that the watch is actually waiting for a GPS fix and not actually calculating the route. 

I would argue that:

- user-hostile philosophy = most UX issues aren’t bugs, and we won’t fix them unless we’re forced to

- user-friendly philosophy = most UX issues are bugs, and we’ll fix them if we can

Guess which one I think Garmin implicitly subscribes to. Obvious “good user experience” can be subjective, but I think some cases are egregiously bad.

This is why I think we as users must demand better. Unfortunately we have no power to effect change.

#Matthias#

I tried a 5km route after getting a fix and it calculated in 20 seconds. Thanks for the help! I take back what I said, the feature isn’t useless.

I don’t like that it requires a GPS fix to work although I get why. I def think it’s a problem that it doesn’t actually tell you that. Another annoyance is you have to hold START to actually select a location to go to (at least on 955), but there’s no indication in the UI. Yeah, I obviously figured it out but I don’t think it’s great design, especially since holding START isn’t a common interaction on these devices.