Walking Forum

Main Boards => General Walking Discussion => Topic started by: Wulfsige on 21:03:38, 05/11/20

Title: OS maps app/distances
Post by: Wulfsige on 21:03:38, 05/11/20
I and a friend who loves anything to do with computerised gadgetry find that our OS maps app consistently underestimates distances both in 'create a custom route' if we are planning, or in recording a route we are walking. For example, 2 miles might be recorded as 1.7 miles; much longer walks show greater discrepancies. Can anyone comment on this?
Title: Re: OS maps app/distances
Post by: ninthace on 21:22:15, 05/11/20
Just to be clear, are you saying that if you create a custom route and then walk it, the distance walked is more or less than estimated?
Title: Re: OS maps app/distances
Post by: BuzyG on 21:53:50, 05/11/20
What are you comparing it against? It always gives me a daft smug feeling listening to others in our Ramblers group discussing their GPS distances at the end of a walk.  Never two the same and sometimes a full mile between worst and best. Then I will measure the route, as I recall it, on walking highlands afterwards.  That figure is always on the shorter side of the GPS discussions.  That's the figure I use.  Might be shorter than reality every time.  Might be longer.  But it the same relative to my last walk, for comparison, so I'm happy. :)
Title: Re: OS maps app/distances
Post by: fernman on 08:42:28, 06/11/20
I don't use the app but I'm just wondering if it takes ascents and descents into account?
These would certainly make the distance walked longer than when measured on a flat map.
Title: Re: OS maps app/distances
Post by: ninthace on 09:11:12, 06/11/20
There could be a variety of reasons. Phone used. Accuracy of plot
Title: Re: OS maps app/distances
Post by: Wulfsige on 09:56:52, 06/11/20
We have a road nearby which is called Ty Gwyn Lane. It is ¾-mile long, and I have measured it a number of times by the odometer in the car, which measures distance in tenths of a mile. The OS app gives it as a lot shorter (0.4, I think it was). Also, when I have asked the app how long a walk is that has also been measured by an Omron pedometer and by various GPS gadgets, again the app is always considerable shorter. So I mean that the OS app always gives a shorter calculation than reality.. I can understand it on moor and mountain, where one's actual path wiggles round water-logged patches, undergrowth, &c &c, but on almost straight road (like Ty Gwyn Lane) it ought to be correct, but is still a good deal shorter. I am asking two questions: (a) am I alone in finding this, and (b) why is it happening?
Title: Re: OS maps app/distances
Post by: ninthace on 12:00:23, 06/11/20
I have just measured a route I walk regularly using the create custom route tool.  At first it under measured it slightly, 5.59km as opposed to the correct 5.63/5.64km as recorded by my phone and my Garmin on numerous occasions.  However, I then switched from the the Leisure Map view to the aerial view and refined the plot so it laid exactly over the route in the image. The app now shows it as being 5.64km which is spot on.
The error lies in the drawing of the map and the accuracy of your plotting.  Features such as roads and buildings are shown bigger on the map than they really are if they were drawn accurately to scale.  This leads to errors which can be significant especially when measuring a short distance such as 3/4 of a mile but become less significant over longer distances.
Hope this helps.
Title: Re: OS maps app/distances
Post by: Wulfsige on 12:29:25, 06/11/20
Thank you. I shall try that. BTW I had to look up solvitur - as solvo = to loose/free from restraint, I guess it means you find freedom release and relaxation by walking: ego quoque.
Title: Re: OS maps app/distances
Post by: ninthace on 13:02:46, 06/11/20
Thank you. I shall try that. BTW I had to look up solvitur - as solvo = to loose/free from restraint, I guess it means you find freedom release and relaxation by walking: ego quoque.
https://en.wikipedia.org/wiki/Solvitur_ambulando 
For me, as a scientist, it has a double meaning.
The literal translation "it can be solved by walking i.e. the benefits of walking to think through our problems and to improve or physical and mental well being.
The philosophical meaning, advocating gathering empirical data to support a theory or opinion.
Title: Re: OS maps app/distances
Post by: Eyelet on 23:09:42, 07/11/20

A few questions for you Wulfsige:

Does Ty Gwyn Lane have any bends in it?
Did you record the GPX track on the OS app in the car or by walking the road length?
If recorded in the car, what speed will you have been travelling at?
What does the recorded GPX track look like when you review it over aerial imaging? Is it precisely nailed to the road?

One of the issues when a track is loaded into the OS Maps app is that it converts it into a route and filters the track points before it is displayed. For a clear explanation of filtering visit the tutorial section of the GPSVisualizer website: https://www.gpsvisualizer.com/tutorials/track_filters.html (https://www.gpsvisualizer.com/tutorials/track_filters.html)

I just loaded a GPX track recorded on a Garmin GPSMap66 with 1654 track points and a distance of 22.7 km. Once imported into the OS app, the displayed distance was reduced to 22.2km. When the route was exported out of the OS app back into Garmin BaseCamp, the route only contained 533 via points. When the original and the OS app tracks were compared in detail, I could see where the filtering had slightly converted many slight walked arcs to straight lines ("cutting corners"), thus reducing the distance. As far as I know, OS haven't disclosed their filtering algorithm, but any applied filtering will shorted the total distance shown. I think this may account for the systematic shortening the OP is finding.

Interestingly, if the OS exported route is converted back into a track in BaseCamp, all 1654 track points were preserved and the distance went back to the original value. Hence the OS app is not modifying the recorded data in the GPX file, just displaying it with some track filtering applied.

Also I agree with Ninthace that plotting a route over aerial imaging will generally give a more accurate distance than plotting over mapping due to cartographic generalisation.
Title: Re: OS maps app/distances
Post by: Wulfsige on 09:20:25, 08/11/20
Thank you. I tried the aerial suggestion which Ninthace gave, and it seemed to solve the problem. I think we've cracked it. Even with the aerial view as the basis for the calculation, there is bound to be some straightening out - for example, on the Ty Gwyn Lane/Pandy/Rhosddu walk (only two miles) there are four or five road junctions which are, as it were, shaped like the letter delta, and to avoid crossing on the wide base of the delta where the road is wide and cars sweep round, it is better to walk up to the tip of the delta, cross where it is narrow, and walk back to the base (= the bigger road which one is walking along). On mountains, riversides, &c this wiggly characteristic of where you actually walk is of course much enlarged, partly because of the nature of the terrain, and partly because such walks are a lot longer. So it seems to me that the aerial view brings a route up towards the real distance walked, and if the wiggly bits are notionally added, it becomes very close to reality.


So - many thanks. I think we've identified and solved the problem.  :)
Title: Re: OS maps app/distances
Post by: Annie oakley56 on 13:52:22, 29/11/21
The problem I have , is that on my list of routes I have an estimated time. say 4:50 Hrs. If I open it t have a look at the detail etc, the time increases to 7:30Hrs. This happens with every route i have. Even Ordnance survey were unable to offer an explanation! Any ideas from anyone?
Title: Re: OS maps app/distances
Post by: forgotmyoldpassword on 14:59:36, 29/11/21
The problem I have , is that on my list of routes I have an estimated time. say 4:50 Hrs. If I open it t have a look at the detail etc, the time increases to 7:30Hrs. This happens with every route i have. Even Ordnance survey were unable to offer an explanation! Any ideas from anyone?

Well, since time in this case is calculated by distance and speed, I'm assuming the app is saving a variable of your average pace when you're walking then applying it to get a customised 'estimated time to complete'.  In some respect that's quite sensible since some of the '4 hour routes' I'd struggle to complete without bursting into a jog, and I consider myself a fast walker already, whereas others I'd complete in 90 minutes and have a leisurely hour spare to enjoy the views. 


In theory if I were designing something like this, I'd like to see these apps log an average ascent speed, average descent speed and then speed on flat - because many struggle with one, both or all and whereas many walkers can motor on quite nicely on the flat, once terrain features occur then slowing to a crawl isn't uncommon.  Maybe this is the future of walking apps, especially since almost everyone has a GPS device on them these days.
Title: Re: OS maps app/distances
Post by: Southwester on 16:14:50, 29/11/21
This link may be of interest about the accuracy of various methods of measurement.


https://www.walk1000miles.co.uk/articles/app-smartwatch-or-fi tness-tracker-whats-the-best-way-to-record-your-miles

 
I am old school, use paper o/s maps and wheel!  
Title: Re: OS maps app/distances
Post by: Southwester on 16:15:59, 29/11/21
Something went wrong in posting link:


 https://www.walk1000miles.co.uk/articles/app-smartwatch-or-fitness-tracker-whats-the-best-way-to-record-your-miles



Title: Re: OS maps app/distances
Post by: ninthace on 16:30:07, 29/11/21
Something went wrong in posting link:


 https://www.walk1000miles.co.uk/articles/app-smartwatch-or-fitness-tracker-whats-the-best-way-to-record-your-miles (https://www.walk1000miles.co.uk/articles/app-smartwatch-or-fitness-tracker-whats-the-best-way-to-record-your-miles)


OS maps 0% accurate, presumably therefore wholly inaccurate.  Confirms a deeply held suspicion.
Title: Re: OS maps app/distances
Post by: BuzyG on 17:53:17, 29/11/21

OS maps 0% accurate, presumably therefore wholly inaccurate.  Confirms a deeply held suspicion.


Nice try.  ;D
Title: Re: OS maps app/distances
Post by: Eyelet on 19:00:27, 29/11/21
The problem I have , is that on my list of routes I have an estimated time. say 4:50 Hrs. If I open it t have a look at the detail etc, the time increases to 7:30Hrs. This happens with every route i have. Even Ordnance survey were unable to offer an explanation! Any ideas from anyone?

I can think of a couple of possibilities:

1. If the route is a recorded GPX track which contains distance, time and elevation data, when it is uploaded, the initial "read" by the OS app displays the total time taken from the GPX file. When you open the route to look at the detail, the OS app loads all the track points (with some filtering for long routes) taking the distance and elevation data from the GPX file. The OS app algorithm for estimating time recalculates this using the speed for that type of activity that you have saved in Preferences and uses this data to calculate the time estimate. OS used to use Naismith formula, but back in May 2019, they said that they intended to adjust Naismith Rule in their app: https://www.ordnancesurvey.co.uk/newsroom/blog/why-were-adjusting-naismiths-rule (https://www.ordnancesurvey.co.uk/newsroom/blog/why-were-adjusting-naismiths-rule). I have not so far been able to find out what the end result was, but the OS algorithm currently seems to be very conservative and overestimates times.

2. If these routes are being imported into the OS app from another source (rather than being plotted using the OS app) it may be that the GPX file contains no elevation data. When you import a GPX with no elevation data, the initial "read" by the OS app displays the time based on the distance of the route only (i.e. it initially reads it as flat). When you open the route to look at the detail, the OS app takes the elevation data from its internal digital elevation model so you can see the elevation plot and the algorithm that calculates the time estimate adds this ascent into the calculation, which adds more time. 
Title: Re: OS maps app/distances
Post by: Mel on 19:08:58, 29/11/21
The problem I have , is that on my list of routes I have an estimated time. say 4:50 Hrs. If I open it t have a look at the detail etc, the time increases to 7:30Hrs. This happens with every route i have. Even Ordnance survey were unable to offer an explanation! Any ideas from anyone?


Based on absolutely nothing other than "a thought".. is it possible that, if you're opening the route at home, the app is assuming you've started to follow it and recalculates the estimated time adding on the extra time it will take to walk to the route's start point from your house?


 
Title: Re: OS maps app/distances
Post by: Eyelet on 19:14:48, 29/11/21

Based on absolutely nothing other than "a thought".. is it possible that, if you're opening the route at home, the app is assuming you've started to follow it and recalculates the estimated time adding on the extra time it will take to walk to the route's start point from your house?


An interesting thought, but if I open a route in northern Scotland, the extra time added is not proportional to the distance from my house to the start point, so I don't think that's the answer.
Title: Re: OS maps app/distances
Post by: Annie oakley56 on 07:28:31, 01/12/21
In 'terms of OS routes, it appears that when they are just a list of routes on your phone , the 'Naismiths' rule for 'elevation  relevant to that route is not applied. As soon as you 'open' the route , it adds it on.