This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Firmware 5.1 vs 5.3: Route Calculation Tests

A big thank you to forum participant ‘8429046’ for providing the link to FW 5.1.

There’s been much back and forth on these forums regarding firmware versions and responsiveness of the Explore’s UI. In the interests of determining the extent that firmware versions affect initial route calculation times a side by side comparison was in order.

If the details don’t interest please skip to the Conclusion.

Parameters:

  • Unit is Edge Explore North American version, updated with most recent North American maps (2020.20); Turn by Turn Navigation is enabled in the Rider Settings; Bluetooth and peripheral connectivity all disabled; battery saver mode is enabled; Navigate to course start refused; GPS locked on. There was no interaction with the unit while it was calculating.
  • The same saved course was used throughout: 192KM long, created in Garmin Connect, then sync'ed to the Explore via Garmin Express with many TBT points. Edit: Here's a Garmin Connect Link to the course used. [https://connect.garmin.com/modern/course/42948515]
  • Three map configurations were used (with the aforementioned route) and each configuration was run twice for each version of firmware for a total of (3 x 2 x 2) 12 results.

The map configurations were the following:

  1. Map Config 1:  Two Stock Garmin Maps Enabled: a) AMR Standard Basemap, NR; b) Garmin Cycle Map Amer, North
  2. Map Config 2:  Four Stock Garmin Maps Enabled: a) AMR Standard Basemap, NR; b) Garmin Cycle Map Amer, North; c) Garmin DEM Map North America; d) Garmin Geocode Map North America
  3. Map Config 3:  One Third Party OSM Map Installed and Enabled: A Routable Bicycle Map (Openfietsmap Lite) for Ontario, Canada (downloaded from garmin.openstreetmap.nl)

RESULTS:

Firmware 5.1:

  • Map Config 1: a) 7 min: 23 sec; b) 7 min: 20 sec
  • Map Config 2: a) 7:12; b) 7:11
  • Map Config 3: a) 7:48; b) 7:40

Firmware 5.3:

  • Map Config 1: a) 9:08; b) 9:02
  • Map Config 2: a) 9:08; b) 8:53
  • Map Config 3: a) 8:50; b) 8:55

CONCLUSION: Across the board, the older 5.1 FW is quicker at route calculation than the most recent 5.3 FW. In the case of the stock Garmin Maps 5.1 accomplishes the calculation in 80% of the time req’d by 5.3; for the third party OSM map, it’s 88% of the time.

Clearly in this area, Garmin’s FW updates are making for a worse user experience.

  • Wow! You performed a deep dive into the subject of the long initial Route Calculation Times... However, 192km is a rather long distance and honestly 7 - 9 min calculation time feels as 'normal' for the Edge Explore. There have been reports about calculation times of 15 minutes and more, but I guess that something is wrong then with the settings / configuration.
    Let's hope that Garmin notices this too and puts big effort in better Route calculation algorithms. Or quicker hardware, but without a price tag...

  • Agree on your points. The conclusion that should concern us though is this: performance is getting worse and it's because of Garmin's updated code. 

    I too feel I can live with these calculation times -- 190KM is not a typical ride but I chose it underscore the point.

    But there's an aspect of the tests I didn't delve into and it's how user interaction and/or peripheral connections bear on the calculation times. For instance, in one of the tests, I was panning across the route map for less than a minute (while the route was calculating). That calculation went from 9 minutes to over 11 minutes -- a result which I excluded. 

    Now consider a rider on the bike with activity timer, IQ apps and Bluetooth all active and wonder that the 15+ minute route calculations--as some commenters have complained of--would be normal. My scenarios reflect best case baselines, that is, zero connectivity (besides GPS) and no user interactions to distract the CPU.

    I wish Garmin would narrow the focus of the Edge Explore to an efficient and capable navigation tool; they already offer models featuring a comprehensive suite of telemetry. KISS Garmin. 

  • Im glad there is an empirical confirmation that the updates are impacting the calculation times.

    Given that each edge model has its own firmware, and that Explore is on the market long enough that no new features will be added, I don't see the reason why code can not be optimized.its not like Windows, developed for multiple devices where some of those become obsolete or unable to run the new version.

    Aother thing which would be interesting to see, it repeating the above tests with older versions of maps (from 2018-2019-2020), but unfortunately, I don't know where to find those.

    I think that map upgrades could have much more impact (because new poi, streets, "layers" if you will, are added, while Explore HW stays the same).

    If you manage to get those old maps, it would be great to see the times then... 

    Oh yeah, I changed my avatar name from 8429046 to this :)

  • As well as earlier maps the same tests can also be conducted with earlier versions of firmware than 5.1 but it's all academic. I mean, we all bought the device to enhance our cycling not complicate our off-hours by devoting time to diagnosing poorly performing gadgets. I'm done with tests. if the next firmware iteration doesn't improve performance I'm going to revert permanently to the 5.1 firmware I've put aside (thank you) for the purpose and that's that. 

    As far as what’s happening with software development at Garmin, who knows? The developers obviously know more about the Explore than us and they certainly have the means to replicate my simple tests and so many more. I suspect their mandate is to implement a feature set that the modestly spec'ed hardware cannot tolerably deliver. Perhaps they’re recycling software elements of their flagship devices, the Edge 1030 and such, that function well in those contexts but flounder here.

    Here’s what I’d like to see: a lite version of the firmware that’s all about navigation and nothing else. No sensors, bluetooth, weather widgets, IQ Apps and all the other CPU cycle and battery hogging add-ons. That’s what I thought I was buying when I ponied up for the Explore.

  • I agree completely. My second navigation device is the Edge Touring Plus. It has all the features you need for reliable (bicycle) navigation, maps fexibility and basic sensor support - no IQ Apps, no widgets, no BT connectivity, no training add-ons. However, the screen is small en low-pixeled so what we need is a device like the Edge Explore with its very nice screen but with the functionality of the Edge Touring Plus, and a SD-card slot please.

  • I would agree to ditch IQ and BT support, weather, notifications, etc. But would like to leave ANT+ support for personal analytics purposes.

    So yeah, to track basic data like speed/distance/elevation, and have navigation.