Zumo User Forums

Hello All

I have been coming gradually to the conclusion that this device cannot tell the difference between major and minor roads which have the same speed limit. For instance in the uk you may have a major trunk road and a small single track road which both have 60mph speed limits though you may struggle to achieve 20 on the minor road in some cases.

I have no unusual settings and all the map updates.

Using faster time - with custom route or just setting destination, the device will often choose very small roads in preference to major ones. ( With custom routes this can happen between via points ).

The overall journey time can then be much longer and possibly dangerous as you struggle to make up time on minor roads.

I often travel together with another rider using the same device and routes and we get the same results.

Garmin try to blame this on mapping errors ( whatever that means, what data is actually missing and how can there be so many)?

From the various forums around there seem to be very many cases of this. Its an expensive device and faster time is a fundamental aspect of the functionality. Im close to suggesting the device is not fit for purpose and should be refunded.

Any thoughts welcome.

I'll attach the full text of my dialog with garmin support.

Cheers

William

I came to the same conclusion many years ago but also that common sense has to prevail. I generally do not trust the device to be smarter than me. For me it's usually just a device to take me how I want to go (via Base Camp routing) and a recorder of where I've been. Even then I often just ignore it and choose my own way. Other than that my familiarity of the roads far outweighs the device's capabilities of deciding which is best. I believe that just because it's on doesn't mean I must blindly follow it's instructions.

Thanks for sharing this information.

Garmin may be right about the mapping errors. But I don't find their responses to be very convincing.

The Zumo 590 introduced a feature called trafficTrends which you could turn on and off. The maps are coded with all sorts of data which I assume allows satnavs to decide things like whether or not to route a high sided vehicle along a road which has a low bridge.
In that coding (I assume) is data that relates to the historic traffic flow along roads for days of the week and time of the day.

Information about that can be found here https://support.garmin.com/en-GB/?faq=4 ... ryKlkrfoh5

I have long suspected, but cannot prove, that later versions also use this data - but unlike the 590, they do not provide a facility to turn it off. Neither the 595 or the XT make any mention of trafficTrends. I started investigating why, when set to faster time, both my 590 satnav and a friend's 595, altered the route in order to take a quiet, extremely narrow back road which is often gravel strewn and muddy.

When I tried to reproduce the 'error' by using my same 590 riding the same route, it refused to recalculate the route in the same way. But it was a different day and a different time.

I wasn't entirely convinced by whatever data it uses for traffic on different ,days of the week as it never seemed to know that Friday morning commuting traffic is significantly lighter on a Friday, and the above example was late Sunday afternoon. Weekday traffic would be quite busy at the time that we were riding. I turned off trafficTrends on my 590. It was clearly using traffic data, but it wasn't particularly useful to me on the roads around here.

Logically, such data will only serve to highlight that traffic is moving slower on a particular road at a particular time. If there is no data, then the expected speed will not be reduced. I suspect that the quiet back roads have insufficient data to have anything recorded - so the calculation will use the unmodified speed limit. So the route will be faster than the main road that historically had traffic flowing at half normal speed at this time of day on this day of the week.


The later models also build up a profile of how YOU ride on particular roads and (I believe) on particular types of road. My evidence for this doesn't bear scrutiny - its observation, thats all - but if you note your ETA for a long trip, and then start to ride slowly on a road with the National speed limit of 60mph for a period of time, the ETA gets later, much more so than the 10 minutes that you have taken to travel a road that would normally take 5 minutes. It seems to assume that on that type of road, you prefer to ride at 30 mph, and calculates the ETA based on that information. Speed up again and it takes the ETA quite a while to go back to the original ETA (+ the 5 minutes you added by travelling slower). If it wasn't using historic riding data to calculate the time for the roads ahead, it would go back as soon as your speed increased to match the NSL.

As I said, its a notion that I have, but it is not evidence that stands up to much scrutiny.

But I have noted a few times that the satnav is always keen happy to shortcut a corner on a poor quality back road which has the same speed limit. You just have to say - I am not going down there, and the satnav will recalculate. I followed up a few such observed route changes. What I have not discovered is exactly when the original route was altered by the satnav to take that short cut. It wasn't there on the original, it didn't alter it on loading the route. I can't make it happen on simulation.

When this started happening, i noted that the Zumo no longer says or displays 'recalculating' - so it is hard to work out when it happens if you never actually go off route. But it does recalculate the route at some point.

Recent experiments with some odd routing behaviour with the XT have suggested that routing prefers main roads to obtain a faster time, rather than taking a side road. I haven't checked whether this prevents the backroad shortcut from happening. Oddly, changing unrelated avoidances - eg ferries - seems to affect this routing behaviour, even if there are no ferries on the route. The same route using the same map on a newly reset 590 calculates the correct route regardless of ferry avoidances. The XT's behaviour is odd on the tests that I did. Summat's up wi' t' XT. All reported in detail, all acknowledged.

The 595 has a lot of similarities in routing behaviour to the XT. My ' 595' is a 590 with 595 software - but not any longer. I've reinstated it to a 590 - the software was more reliable for me. Although it is on my desk at present, and the wiring for it has been removed from my bike. Perhaps prematurely, given the various issues I am having with my XT. But not enough to go back to the 590/595 appalling battery life and dim screen. Yet.

I agree with the comment on devices appearing to learn how you ride particular types of roads, I used to be able to beat the Garmin ETA on certain roads that I use regularly but now those times are amazingly accurate. Indeed Garmin do say their devices do this, see

https://support.garmin.com/en-GB/?faq=S ... xBFWKu9tb6