speedometer or GPS?

[oldjones]

Of course, a GPS that didn't also use map-data—which we can hope was accurately measured on the ground—would be pretty useless. So what you're speculating about is not just the satellite postioning error but the accuracy of the software's translation of that into map positions, and then into distance travelled over time.

Does anyone actually know the map-data source for their GPS? And do you trust it?

 

[raydeon]

.........................................
Does anyone actually know the map-data source for their GPS? And do you trust it?

A US company named NAVTEQ supplies the maps for Garmin and many other GPS. NAVTEQ is now owned by Nokia.

Based on my own experience I trust the map information about 99% of the time. As I pointed earlier it is impossible for any map maker (be it electronic or paper maps) to keep up to date with the constant changes which take place in roads, streets, cities etc... In addition there are detours, constructions, accidents which cannot be shown on any map.

But because GPS are persistent in getting you to your destination, they will always get you back on the right track. It is virtually impossible to get lost.

Map makers update their maps on a regular basis. Navteq offers revisions 3 or 4 times a year for their N.A. and European maps. Of course they are not free updates, often costing as much as the price of a new GPS which comes with a current map.

 

[benstt]

Of course, a GPS that didn't also use map-data—which we can hope was accurately measured on the ground—would be pretty useless. So what you're speculating about is not just the satellite postioning error but the accuracy of the software's translation of that into map positions, and then into distance travelled over time.

Does anyone actually know the map-data source for their GPS? And do you trust it?

I would guess that the velocity calculations don't use the map data at all.

I would also guess that a GPS, while being a little fuzzy on where you are exactly on the planet, can likely tell that you've moved a kilometre reasonably accurately.

Guesses only.

 

[raydeon]

We are losing focus here. The original posters question was, does his vehicles speedometer provide a more accurate representation of his true traveling speed as opposed to that of a GPS? And the answer is, the GPS is more accurate. Period. Is the GPS perfect, no, of course not, but darn close.

Right on OM. Let's hope this is the end of it.
Thank you.

 

[oldjones]

I would guess that the velocity calculations don't use the map data at all.

I would also guess that a GPS, while being a little fuzzy on where you are exactly on the planet, can likely tell that you've moved a kilometre reasonably accurately.

Guesses only.

If they don't reconcile the satellite positions they're calculating velocity between with the map data, then both of them could be off by the maximum amount and grossly over or under-estimate the speed. Seems to me your only hope of accurate speed from a GPS is a) that the map data does show accurate distances between points and b) that the software has figured you can't actually be 30 yds beside the road, and were actually on the bridge not a hundred feet upstream in the river (like the satellites said) when it calculates how fast you drove from the bridge.

Counting wheel revolutions—which is your speedometer's method—is inherently a more accurate method, but the manufacturers have an interest in overstating speed, and no one really keeps them honest. I absolutely don't believe NAVTEQ did their own surveying and map making. It's damn hard—and expensive*—to map the world at the scale and tolerances of driving and calculating speed over short distances. Which is why, when it counts you see guys out there actually measuring the full-scale 'map'—the Earth. And what do they use? Very often a rolling wheel on a handle.

* Mapmaking, like roadbuilding and defence: another fine traditionally 'socialist' project paid for by your tax dollars. And I'll bet a pile of donuts against them that, like the satellites, the maps that underpin your GPS were government-made.

 

[benstt]

If they don't reconcile the satellite positions they're calculating velocity between with the map data, then both of them could be off by the maximum amount and grossly over or under-estimate the speed. Seems to me your only hope of accurate speed from a GPS is a) that the map data does show accurate distances between points and b) that the software has figured you can't actually be 30 yds beside the road, and were actually on the bridge not a hundred feet upstream in the river (like the satellites said) when it calculates how fast you drove from the bridge.

I've used gps software that uses raster maps, meaning that it's just all pixels to the gps. It would show me where I was on the map, but it itself had no concept of roads, etc.

It allowed different map sources to be used - google maps, etc.

It was very accurate, meaning the position that the gps placed me on the map was almost always on a road, even though the software was unaware of roads.

(If I needed to plan a route, it would query a server on the internet that was aware of roads, and use information from it to plot a route.)

I have no doubt that it was accurate enough to calculate a steady-state speed, even though it didn't use map data.

And it would throw in height above sea level too.

 

[simon482]

Simon, you shouldn't talk about things you don't know about my friend. I never reply to posts unless I'm sure and educated on the topic at hand. Seems you should consider taking the same steps before you offer your "valuable" opinion.

there is no way in hell you are going to convince me that gps is more accurate then my spedo.

 

[oldjones]

I've used gps software that uses raster maps, meaning that it's just all pixels to the gps. It would show me where I was on the map, but it itself had no concept of roads, etc.

It allowed different map sources to be used - google maps, etc.

It was very accurate, meaning the position that the gps placed me on the map was almost always on a road, even though the software was unaware of roads.

(If I needed to plan a route, it would query a server on the internet that was aware of roads, and use information from it to plot a route.)

I have no doubt that it was accurate enough to calculate a steady-state speed, even though it didn't use map data.

And it would throw in height above sea level too.

And because of that one word, I'd only admit the GPS speed was almost always accurate. So we agree. Now if we only knew when it was almost accurate and how big an error 'almost' allowed for …

 

[benstt]

And because of that one word, I'd only admit the GPS speed was almost always accurate. So we agree. Now if we only knew when it was almost accurate and how big an error 'almost' allowed for …

No, we don't agree. Thanks.

In the cases where the GPS showed me off the road, the map was wrong. It's not like it bounces me randomly all over the road; the placement on the map is smooth.

 

[Old Milwaukee]

there is no way in hell you are going to convince me that gps is more accurate then my spedo.

Obviously the majority of posters on this thread, including me, must be wrong, according to you. These people http://www.flyertalk.com/forum/travel-technology/892338-gps-vs-speedometer.html must be wrong as well.

 

[simon482]

Obviously the majority of posters on this thread, including me, must be wrong, according to you. These people http://www.flyertalk.com/forum/travel-technology/892338-gps-vs-speedometer.html must be wrong as well.

how did you tell your speed accuratly before the invention of the gps? did you lick your finger and stick it out the window and judge the windspeed? speedometer is a proven technology that has been around since the invention of.......THE CAR.

 

[benstt]

Obviously the majority of posters on this thread, including me, must be wrong, according to you. These people http://www.flyertalk.com/forum/travel-technology/892338-gps-vs-speedometer.html must be wrong as well.

In reading up on this, it looks like GPS uses doppler shifts of multiple satellites to determine position and velocity estimates.

Ie they can estimate velocity just on one point in time sample; they don't need to wait and compare multiple positions.

 

[Old Milwaukee]

how did you tell your speed accuratly before the invention of the gps? did you lick your finger and stick it out the window and judge the windspeed? speedometer is a proven technology that has been around since the invention of.......THE CAR.

Now that is a funny argument. I'll pretend I never read that and go on with my day.

 

[simon482]

Now that is a funny argument. I'll pretend I never read that and go on with my day.

see how funny works.

 

[Old Milwaukee]

see how funny works.

Simon, you don't know what you are talking about regarding this topic. You are wrong and you clearly can't admit it. Now you are just looking foolish. Best be done with this post, cause I am.

 

[OddSox]

how did you tell your speed accuratly before the invention of the gps? did you lick your finger and stick it out the window and judge the windspeed? speedometer is a proven technology that has been around since the invention of.......THE CAR.

There used to be mileposts on the 401, complete with signs that said "Check Your Speedometer".

Edit: The mileposts are still there, except they are in kilometers now.

 

[simon482]

Simon, you don't know what you are talking about regarding this topic. You are wrong and you clearly can't admit it. Now you are just looking foolish. Best be done with this post, cause I am.

i always admit when i am wrong. in this case i don't believe i am and it will take a lot more then some website to tell me otherwise.

 

[Thunderballs]

My Tom Tom matches my speedometer reading within 1 km most of the time. I was actually surprised that it was that accurate.

 

[Brandon123]

I wouldn't rely on a GPS, stick to your speedometer, not worth getting a ticket over it...

 

[Moraff]

I wouldn't rely on a GPS, stick to your speedometer, not worth getting a ticket over it...

Unless you are going to go to the time and effort of determining the accuracy of your speedometer I would suggest going by the lower of the two might be the best alternative (if you drive in the realms of speed where a few km/h difference in actual speed will be the determining factor in ticket or no ticket)