Most accurate mapping software

mickmcl09

I've been using map my ride to plot spins. The accuracy seems to leave a little to be desired in terms of gradients & total elevation. What other sites are people using and do you find them more accurate?

AstraMonti

I mainly use Ridewithgps.com, it's faster and better for my taste. I have problem with syncing it with garmin but that's another issue.

mickmcl09

Much easier to use, faster to plot routes and looks alot more accurate in terms of elevation.
Can you plot feed stops etc. like you can with Map my Ride?
Overall I'd say I'll be using ridewithgps, it far easier and quicker.

Thanks

smacl

In terms of accuracy, I think the issue is the data used rather than the software. AFAIK, most free site use the NASA SRTM for height and hence interpolated gradient. The ordance survey have better data available, but not for free. I use OSi trailmaster, which has since been replaced by Geolives, that I've yet to try. If you're handy with GPS, the scottish mountaineering club has much of Ireland available as contours for garmin.

Dr_Colossus

I did a 150km route yesterday through the Wicklow mountains including Slieve Maam and Shay Elliott, the garmin 705 gave the total ascent as 2,185m. Mapmyride gave the ascent as 1,669m while RidewithGPS gave it as 2,390. I don't know how accurate the garmin is but would have to presume it's pretty close since it's graphing it as you cycle? RidewithGPS seems alot more accurate anyway in terms of altitude plus it's alot easier to graph and modify.

One problem I had though with the gpx file downloaded from RidewithGPS is that the route planned wasn't what was showing exactly on the mapping. ie it was giving a different route through the city to that panned. Out in the wicklow mountains it was exact though. Anyone have this problem, it was my first time using the RidewithGPS gpx file.

ROK ON

Ridewithgps is a very nice interface, however its elevation counting is insane. I have done a ride with 2 300m ascents and i get back load my Garmin date into the pc. Presto ridewithgps will tell you that you did about 1500m.
I stopped using it after that. Mapmyride is a very poor interface (fails a lot on the various pcs that I use). But seems much more accurate in terms of elevation gain.

Inquitus

ROK ON wrote: »

Ridewithgps is a very nice interface, however its elevation counting is insane. I have done a ride with 2 300m ascents and i get back load my Garmin date into the pc. Presto ridewithgps will tell you that you did about 1500m.
I stopped using it after that. Mapmyride is a very poor interface (fails a lot on the various pcs that I use). But seems much more accurate in terms of elevation gain.

mapmyride is just as bad, had the SKT 90 at 432m and it was 1000m on the 4 gps I have seen since, I guess both can be more or less reliable on different routes, but neither can be relied on.

serendip

GPS does a rather good job of latitude and longitude, but I wouldn't trust it too much on elevation. I've been climbing in the Alps and seen it out by more than 100 meters. And watched myself gaining and losing height while eating my picnic.

Nietzschean

serendip wrote: »

GPS does a rather good job of latitude and longitude, but I wouldn't trust it too much on elevation. I've been climbing in the Alps and seen it out by more than 100 meters. And watched myself gaining and losing height while eating my picnic.

GPS as a technology can be very accurate for elevation, its mostly down to the reciever's accuracy. And most bike GPS's are as you said good at lat/long not so good at elevation. Though i'm curious as to which one you had? Have people managed to compare accuracy while stationary across a few of the higher end ones?

Is 705 > 305? are the next gen 800/500's any better?

blorg

The bike GPSes are far better than that at elevation, at least the ones with barometric altimeters (305/705.) When working they are not more than 10 meters out, certainly not hundreds, and will give accurate elevation gains over a ride. The main issue in Ireland is what sort of smoothing level you apply- this is how you get wildly different accounts. E.g. the road imperceptibly goes up 2m and down 2m, up 3m, down 1m. Do you count this as 5m elevation gain, or ignore it? That depends on the smoothing. In the Alps the numbers tend to be far closer between sources.

serendip

blorg wrote: »

The bike GPSes are far better than that at elevation, at least the ones with barometric altimeters (305/705.)

Yup. For mountaineering, I stick to good old fashioned barometric pressure for elevation. Even with gradual changes in atmospheric pressure, I still find it more reliable than the GPS.

(My GPS is an old Garmin ForeTrex 201. Perhaps they do better GPS chips nowadays.)

Oh. And note I said that I "have seen" it about by 100 meters. Being out by up to about 10 meters vertically would be more typical. And it's certainly influenced by how much of the sky you can see.

markcroninbsc

the reason why mapmyride, and ridewithgps and all the other ones give different total ascent values is because they all use different algorythms to calculate it. I'm not sure what ridewithgps uses but i think mapmyride only count climbs with an elevation gain of more than 80 meters, and it seems a bit buggy on it calculations anyway. I think the garmins and ridewithgps seem to count every evevation change which isnt really much good either because you dont really want to be counting speed ramps on your total ascent.

ridewithgps seems to be the best one tho.

Theres an app for blackberries (BBSpeedOMeter http://www.mobihand.com/product.asp?id=69752&n=BBSpeedOMeter) that i made. puttin up a new version this evening that'll integrate seamlessly with ridewithgps.com if anyone with a blackberry is interested :P

My app calculates total ascent like this:

hill must be 35 meters to count as a hill and after the first 35 meters it must maintain a gradient of over 1% to keep counting.


the reason why gradient seems wrong on most of these sites is because they average out the gradient over a distance. so for example if you map a spin on maymyride thats 10 miles long it'll average out your gradients over every 1/4 miles but if you map a 100 mile spin it'll only give you average gradients over 1 miles.

CheGuedara

Finding minimal differences between elevation data returned for rides on Garmin Connect and RwGPS e.g. 2006m vs 1983m for the SKT, so not convinced its badly out in general (GPS is Edge 500 if it make a difference)

Have recently given strava.com a go as well. Seems OK (SKT ele = 1993m) , nice to look at, auto detects and categorises climbs with function for comparing ascents between your own rides and against club mates that use the site.

Some silly omissions though (cant seem to graph elevation vs speed) and estimated power output...hilarious. Good potential in it though

Blowfish

markcroninbsc wrote: »

the reason why mapmyride, and ridewithgps and all the other ones give different total ascent values is because they all use different algorythms to calculate it. I'm not sure what ridewithgps uses but i think mapmyride only count climbs with an elevation gain of more than 80 meters, and it seems a bit buggy on it calculations anyway. I think the garmins and ridewithgps seem to count every evevation change which isnt really much good either because you dont really want to be counting speed ramps on your total ascent.

I'd disagree with that, there's a vast difference between the effort needed to maintain a constant speed on the flat and maintaining the same speed on 'rolling' terrain, even if the lumps you are powering over are only 20m high.

martyc5674

I have a garmin 305...IIRC i think it has a barometer(or did i imagine that!!)

Cheers,
Marty.

smacl

Just out of interest, I decided to compare an elevation profile produced from an OS map versus MapMyride, for a section from Enniskerry to Glencree. See results attached. The OS reports the height at the junction with the Military road as ~370m, whereas mapmyride has it as ~345m. Given the OS data is based on a 10m grid, whereas the mapmyride is based on a 90m grid (SRTM 3 arcsec), you also see the ground undulations much better. For those using GPS, and interested in such things, it would be worth checking your heights against OS heights or contours (which are good to ~5m on discovery series).

I wasn't too impressed with mapmyride, as it wouldn't let me take the route I wanted from Johnnie foxes in glencullen down to Enniskerry. Could well be user error though, as I haven't used this app before.

Blowfish

smacl wrote: »

I wasn't too impressed with mapmyride, as it wouldn't let me take the route I wanted from Johnnie foxes in glencullen down to Enniskerry. Could well be user error though, as I haven't used this app before.

That's actually a google maps issue, you can see it here when zoomed in, it shows the road as not being continuous.

blorg

Blowfish wrote: »

That's actually a google maps issue, you can see it here when zoomed in, it shows the road as not being continuous.

Google Maps has that issue a lot where a road crosses a border, whether a border between countries or in this case the Dún Laoghaire-Rathdown/Wicklow border.

Limestone1

smacl wrote: »

I wasn't too impressed with mapmyride, as it wouldn't let me take the route I wanted from Johnnie foxes in glencullen down to Enniskerry. Could well be user error though, as I haven't used this app before.

Untick the 'follow road' check box for that section. It will allow you to create the route point to point .

Mr Velo

Try Runkeeper (www.runkeeper.com)
I've been using it for a few weeks now and find it very impressive for route planning and ride mapping.