Is GPS MPH accurate

[hindle8907]

Just been for a ride out and tracked the route on my GPS device.
it shows me where i did my max speed on the software and this is 2 mph over what was indicated on my speedo . but i thought speedos over read ? not under ?
cheers.

[Wideboy]

no they are not accurate

[DarrenSV650S]

How long were you riding at your max speed? You need to be at that speed for a few seconds for the gps to register it I think

[Thingus]

Sometimes read more sometimes read less. I'm sure there's a use for it for average mph / miles covered but i wouldn't trust it for speed cameras or whatever.

[Pingu65]

My GPS Tracker is always reading less than the speedo.

e.g If i could get too 100MPH on the speed then the GPS would probably show about 93-94 MPH ;)

[hindle8907]

you would think the GPS would under read yeah ... i didnt think it would over read .

[DarrenSV650S]

What gps do you have? Also, you didn't answer my first question :)

[Specialone]

Just been for a ride out and tracked the route on my GPS device.
it shows me where i did my max speed on the software and this is 2 mph over what was indicated on my speedo . but i thought speedos over read ? not under ?
cheers.

I used to work with an ex- ops manager for VDO, who make speedos etc for loads of manufacturers, he told me most clocks were approx 5 mile over when they sent them out prior to fitting on the track.
You think about it, if everybody had a speedo reading over its the best way of reducing speed.
Ive tested this theory, the signs on the street that flash your speed up are always 3-5 mph out to what my speedo says.
Things like tyre wear can affect the true accuracy of the speedo as they wear, the circumference / diameter of the wheel reduces in size meaning every revolution of the wheel is not travelling as far.
Dont police have to have their speedos calibrated regularly?

My tom tom seems more accurate to these speed signs when i have it on while going past them than my speedo

[Rocket]

What is speed? Distance divided by the time to travel that distance. A GPS is very good at measuring location (distance) and time so therefore it should be accurate. I have used TomTom on my iPAQ since 2003 and it's always been accurate for me. Modern speedos seem to all lean towards the top end of 10% over.

[Geoffrey]

the royal air force do not believe gps systems are accurate, and their technical department is bae systems

i think there have been a few threads on here about this subject, the majority have the view and knowledge that they are not accurate.

[Spanner Man]

Good morning all.

My 'Cab Nag' consistently reads around 10% less than the speedometer in my van.
Upon passing the YOUR SPEED IS? Type signs, my Satnav is bang on whatever the sign says.
Howver, once I was out collecting a bike in deepest darkest Derbyshire with a couple of mates, who had bought along their Satnavs, so just for a giggle we had all 3 going at once, & guess what? They all read differently. At 50 mph on my van speedo my satnav read 45, whilst the others read 48 & 51 respectively.

So in short, I would say you can't rely on them to be any more accurate than your speedometer.


Cheers.

[Alpinestarhero]

I suppose it depends on how many satellites are used to locate and track your position, right?

You've opened a can of geek-nerd worms now stewart :lol:

[Spanner Man]

I suppose it depends on how many satellites are used to locate and track your position, right?

You've opened a can of geek-nerd worms now stewart :lol:


OH BLAST! I didn't think of that when I wrote it! I'd better hide before I get smothered in Anoraks!:shock:


Cheers.

[xXBADGERXx]

SatNag`s are not accurate . Best way to get the most accurate reading is to take your glove off , lick the end of your finger and point it upwards whilst riding and have a guess ....... you stand just as much chance of getting it right .

[Gerry]

At www.mkivsupra.net (http://www.mkivsupra.net) (who have a thing about speed) the feeling is that a GPS readout is usually more accurate that a speedo reading.

I have two GPS units (one TomTom satnav and one Snooper) and they both read exactly the same, about 10% under my old stock speedo.

Saying that, under acceleration both units are about 1-2 seconds behind what you are doing.

Just my 2p's worth... :)

[hindle8907]

What gps do you have? Also, you didn't answer my first question :)

Well its my Iphone ... And i was going 102 MPH (On my private Runway ) on the speedo for about 1 minute but if a speedo over reads by 5-10% and my tracking software is showing max speed of 104.4 MPH that means that my iphone is well over reading.

as some one just said gps does time vs distance so you wouldnt think it would be that much out .

[the_lone_wolf]

The iphone's GPS receiver is never going to be a match for a dedicated satnav

My Zumo550 reads consistently ~5% below the speedo on my ADV and a little more when in the car

[Geoffrey]

as some one just said gps does time vs distance so you wouldnt think it would be that much out .
but look at the delay gps systems have in sending and receiving information, you will never get an accurate system

[ThEGr33k]

Its based on triangulating from satellites miles ahead. The angles involved are tiny! I think that it wont be as accurate as some people claim it is...

BTW GPS devices dont send info. Can you imagine the antenna you'd need for that?

I dont think the Iphones GPS is GPS, I might be wrong as ive not really researched it but I think its actually something to do with the actual phones reception from different masts.

[hindle8907]

oh ok .

so if my speedo over reads by 5-10% my GPS device is over reading more.

[hindle8907]

the iphone does have gps ... its a verry good gps imo ... verry acurate on the mapping/tracking side of things .. and gets a connection faster than my sat-nav.

the first gen iphone didnt.

[the_lone_wolf]

the iphone does have gps ... its a verry good gps imo

except it isn't...;)

It's pretty weak, simply because you can't get the hardware required for a good gps signal into the space available in a phone

once it gets going the Navigon app has no problems with navigation or losing satellites – providing the iPhone was positioned right on the dash. Taking it away from the windscreen to use the iPhone for other purposes meant that the GPS signal was lost.
Source (http://www.techradar.com/news/portable-devices/satnav/a-week-with-the-iphone-as-a-sat-nav-623238)

The zumo will work indoors on the ground floor of my house after taking a while to find satellites...

[ThEGr33k]

oh ok .

so if my speedo over reads by 5-10% my GPS device is over reading more.

Basically it sounds like it... Jap speedo's seem pretty terrible for accuracy :(

[-Ralph-]

My GPS is accurate to within 1 mph, I did my own distance time calculations over varying distances, on a dead straight road, using cruise control to regulate speed and OS maps to measure distance. It's the kind of cheap sat nav that was £120 quid 2-3 years ago and can now be bought in the supermarket for 60-80 quid.

No, I'm not normally that sad, but I do a lot of mileage, much of it through average speed cameras on the motorway, I wanted to know what device I should be using to set my cruise control - it was a license preservation exercise.

My car speedo over-reads by 5% and my SV by 10%.

[hindle8907]

except it isn't...;)

It's pretty weak, simply because you can't get the hardware required for a good gps signal into the space available in a phone


...


well i have never had a problem with mine for satnav/ or Tracking its worked flawlessly and accuratly every time ... its just been this one queery aabout the MPH accuracy that made me ask my question ..

Where do you get that from ?
that its too small to get a good GPS Signal
i can get a GPS signal on my iphone in my house ... but if i get the sat nav out and try it doesnt ... it needs to be outside with a clearview of the sky ?? so this iphone thats really weak works better than a 190 pound Tom tom ?

[the_lone_wolf]

Where do you get that from ?
that its too small to get a good GPS Signal
Seriously?:confused:

Not even steve jobs could make a phone like mary poppins' handbag:smt003

[-Ralph-]

And Bluepete's is accurate too.

My TomTom one gives the same speed readings as the calibrated speedo in my job cars and the same speed as the provida video. That's good enough for me.

Both bike and my own car's speedo over read. I prefer to use the TomTom in my own car for setting the cruise control speed on long trips.

Dunno what my bikes top speed is, certainly no more than 70mph surely?

Pete

[Dicky Ticker]

The transmission waves travel in a straight line plotting a straight course for the reciever between signals. You on the other hand are going round corners or deviating your course slightly,up or down hill, which could account for the difference.It is a miniscule difference but can account for the error between a 100% calibrated speedo and the GPS signal.
Another point being that your GPS will show a positioning error and even an expensive ships GPS can be a couple of feet out most of the time even with twelve satalites

All speedos are manufactured to under read but if you want to spend money they can be calibrated to within +/-1% fairly cheaply

[robh539]

The transmission waves travel in a straight line plotting a straight course for the reciever between signals. You on the other hand are going round corners or deviating your course slightly,up or down hill, which could account for the difference.It is a miniscule difference but can account for the error between a 100% calibrated speedo and the GPS signal.
Another point being that your GPS will show a positioning error and even an expensive ships GPS can be a couple of feet out most of the time even with twelve satalites

All speedos are manufactured to under read but if you want to spend money they can be calibrated to within +/-1% fairly cheaply


Hi gent while I agree sat navs are not accurate that we have. The military do use them. on ships, planes. commercial services use them. They are accurate to within a 1/2 mete. Though the stuff we can buy in the high street are crap in comparison, as there made not to be spot on. There is a mathematical formula to compensate for this.

They are getting better

[embee]

This gets debated regularly, "Hitler's Speedo" syndrome.

2 things, first vehicle speedos are required by law to "not under-read" speed, so by definition they will all over-read to some extent. It's not a great reflection of whether they are "crap", it's simply a case of how much margin the manufacturers choose to build in to allow for all the variables and still be legal. Up to 10% optimistic is typical. Odometers are usually a bit nearer because they are not subject to the same legal requirements.
Second, GPS is "accurate", depending on what you mean by "accurate". There is no such thing as perfect accuracy, so it's a case of how close is good enough for the purpose. GPS/SatNav manufacturers have no reason to want to build in deliberate inaccuracy, so it's down to sattelites, signals, times, intervals, algorithms, (altitude?), etc. They certainly aren't going to be "accurate" on an instantaneous basis, but over a reasonable distance in a reasonably staight line at reasonably constant speed they'll be reasonably accurate, and very probably nearer the truth than the speedo.

[Geoffrey]

The military do use them. on ships, planes. commercial services use them. They are accurate to within a 1/2 mete.
being a former raf loadmaster with 33sqn, 18sqn, 27sqn, and 47sqn (and still keeping within the circles at raf Lyneham); i can confirm the statement above is not true in that they are not that accurate!

[the_lone_wolf]

They are getting better
Half a metre?

The GPS equipment we use can accurately pinpoint to within 5mm in 3 dimensions given a good view of the sky, ~25mm under tree coverage:thumleft:

With EGNOS enabled my zumo can get down to ~5m accuracy, but I don't know how accurate it's own assessment is...

[Sir Trev]

My cab-nag (must remember that one) does not provide a speed reading :cry:. But then it does fold down to a very flat and easy to pack size.

Shame it's such a faff to open it out to read from it. And if you try and do it on the move in the car it's bad enough (fills the windscreen) but on the bike it would probably rip or just blow away...

[simesb]

GPS does not "measure" speed. It can only calculate it as a function of the time taken to move from one set of co-ordinates to another. As these co-ordinates are never accurate (for civilian units) and are a the speed is therefore not necessarily accurate.

[boot]

There are plenty of stories about (http://arstechnica.com/gadgets/news/2008/07/nabbed-for-speeding-gps-data-could-get-you-off-the-hook.ars), where people have refuted speeding fines and points using GPS data. I often record my GPS data while on the move, and would use it to contest should the event occur and I believe to be in the right.

From personal tests over long distances, I'd say the GPS in my SEX1 is pretty damn accurate.

I don't calibrate my GPS, and I'd put good money on it that my bike's speed hasn't been calibrated in the last 9 years - I know which I think is more accurate.

[Spanner Man]

My cab-nag (must remember that one)


Just make sure you tell the missus you're not refering to her :D


Cheers.

[trumpet]

my gps read 10 mph slower held in top for around 30 secs

[MattCollins]

Pearler of a subject... :rolleyes: Manufacturers of the current crop of receivers claim accuracies of around 0.5m/s (a smidge over 1mph) in ideal conditions. The real world is more like up to +/- 2 or 3mph, but it is possible to get errors well outside of this range.

GPS does not "measure" speed. It can only calculate it as a function of the time taken to move from one set of co-ordinates to another. As these co-ordinates are never accurate (for civilian units) and are a the speed is therefore not necessarily accurate.

Not quite. :) You're right about the accuracy but GPS receivers (anything reasonably modern) are heavily weighted towards Doppler to measure speed (edit: and direction).
Also, Doppler (if available) is used to extrapolate position when a reliable fix is not available which is more common than most would guess - it is often possible to get good Doppler measurements when a good fix is unavailable . Consumer gear does a lot of guessing, rounding, fiddling, whatever (which can produce some "interesting" errors), so that it displays continuous, usable, but not entirely accurate data.


Cheers

[muffles]

the royal air force do not believe gps systems are accurate, and their technical department is bae systems

i think there have been a few threads on here about this subject, the majority have the view and knowledge that they are not accurate.

I presume you mean on planes, in the air? I can see them being inaccurate cos what GPS would measure is of no relevance - it would be absolute air speed or some term like that.

I thought planes only cared/measured ground speed (which would be measured incorrectly by a GPS on the plane cos the plane isn't on the ground and the earth isn't flat) and air speed - which isn't to do with absolute position so is affected by head/tailwind. Correct me if I'm wrong of course, but it would seem logical GPS cannot measure something a pilot is interested in.

[MattCollins]

Muffles,

That'd be true of the average consumer model GPS which measures speed but does not calculate ground speed, but it would only be noticeable at higher altitudes and speeds. Aviation GPS receivers are an entirely different breed.

Cheers

[Geoffrey]

I presume you mean on planes, in the air? I can see them being inaccurate cos what GPS would measure is of no relevance - it would be absolute air speed or some term like that.

I thought planes only cared/measured ground speed (which would be measured incorrectly by a GPS on the plane cos the plane isn't on the ground and the earth isn't flat) and air speed - which isn't to do with absolute position so is affected by head/tailwind. Correct me if I'm wrong of course, but it would seem logical GPS cannot measure something a pilot is interested in.
the raf is more than just planes mate, we do not rely solely on our gps systems in any aspect. what do you think causes all those friendly fire and near miss incidents, despite what the commoners think it is not the americans fault (no i am not american)

[-Ralph-]

despite what the commoners think

Well that's me charmed Geoffrey! Now go warm up the Bentley old boy! ;)

[Geoffrey]

Well that's me charmed Geoffrey! Now go warm up the Bentley old boy! ;)
i was an nco so i worked for a living, being an officer did not appeal to me so i did not take the charm classes

[muffles]

the raf is more than just planes mate, we do not rely solely on our gps systems in any aspect. what do you think causes all those friendly fire and near miss incidents, despite what the commoners think it is not the americans fault (no i am not american)

Fair enough - out of interest what do they use to measure position/speed in the RAF (planes or otherwise)?

I actually used to work for BAE SYSTEMS (btw for anyone not 'in the know' that's supposed to be all caps lol) and I worked on a classified (forget the term - not the full security clearance, only required the 'job level' one I got when I joined) project that involved UAVs and thus would have some aspect of location/speed to it. I'll see if I can find out any details on what was used, but it's 5+ years ago now and I'm not sure anyone I am still in contact with, still works there...

[davepreston]

i'll ask the head engineer of the uav programe when im in work next if ya want as i see him every day and have the top security clearance their is for BAE (being security there and all that lol)

[muffles]

i'll ask the head engineer of the uav programe when im in work next if ya want as i see him every day and have the top security clearance their is for BAE (being security there and all that lol)

Lol yeah go for it, I'm not sure if there are mutiple programmes or anything though. If it helps I used to work at the ATC Great Baddow and they did some testing in Woomera in November 2003. I'm trying to think what I can and can't say (not much chance of anything happening but I don't want to get arrested lol). Do the UAVs (well, the ground systems) have a red box with two buttons on top?

[redshift]

GPS is getting a lot better - I remember doing a project ten years ago with a Rockwell PLGR (military issue GPS handheld receiver) that would register 25mph when sat on my desk.

Speed in aircraft is still measured the old fashioned way - compare static pressure with pitot pressure, apply a few cunning calculations depending on height, temperature etc etc then display the result.

[flymo]

of course they are accurate, its just a question of the degree of accuracy.

[muffles]

of course they are accurate, its just a question of the degree of accuracy.

Lol that's both obvious and pointless :lol:

I could try and shoot you and actually aim completely the other way, then say I was accurate, it's just a question of how accurate - I was only 180 degrees off! :lol:

In the cases that are being put forward of the RAF they are saying they are not accurate enough for their purposes (adding the bit in italics as the clarification) but that's pretty much implied/assumed as otherwise the statement means nothing (as per my shooting example above).

[flymo]

Lol that's both obvious and pointless :lol:

I could try and shoot you and actually aim completely the other way, then say I was accurate, it's just a question of how accurate - I was only 180 degrees off! :lol:

In the cases that are being put forward of the RAF they are saying they are not accurate enough for their purposes (adding the bit in italics as the clarification) but that's pretty much implied/assumed as otherwise the statement means nothing (as per my shooting example above).

A little tongue in cheek perhaps but with an element of serious point there too. Accuracy is a relative term.

Under most conditions I would expect a vehicle GPS unit to be more accurate than a typical speedo in terms of what is displays to you. A speedo is a pretty accurate device especially in a modern vehicle but as has been described, they are intentionally configured to display higher than actual speed within a set tolerance.

Both are affected by different conditions that can reduce their accuracy though.

[AndyW]

A little tongue in cheek perhaps but with an element of serious point there too. Accuracy is a relative term.

Under most conditions I would expect a vehicle GPS unit to be more accurate than a typical speedo in terms of what is displays to you...

The GPS needs to take account of height. If you are going up a 1:4 on a twisty switchback road, distance driven != distance between points.
For an (unfortunatly impossible) example to illustrate, if you drove vertically upwards, the GPS would be showing a speed of 0, whereas you'd be exceeding escape velocity at 11.2 km/s

[flymo]

The GPS needs to take account of height. If you are going up a 1:4 on a twisty switchback road, distance driven != distance between points.
For an (unfortunatly impossible) example to illustrate, if you drove vertically upwards, the GPS would be showing a speed of 0, whereas you'd be exceeding escape velocity at 11.2 km/s

That may be true but its not real world most of the time in the UK is it?

[AndyW]

That may be true but its not real world most of the time in the UK is it?
Lots of hills in Wales and Lake District sufficiently steep to affect reading, though by how much is open to the more mathematically minded
Scratches head and tries to remeber math from 30 years ago:
sketch as right angled triangle, distance travelled = hypot, distance between points ignoring height = adjacent, theta = angle of incline
cos theta=a/h
we want difference between a and h as a fraction to multiply the GPS speed reading by...
erm, I give up I need more coffee to try and work this out without drawing it out.

[flymo]

ok, so putting aside this as a suitable measure for mountian rescue rovers :-), generally speaking it would be a fairly accurate system in comparison to a typical speedo that hasnt been calibrated.

[-Ralph-]

Vehicle GPS is only really any good at steady speed in a straight line. Try measuring your speed with one when braking from 60 down to a 30 limit and see if you arrive at the sign knowing for sure what your speed is. They will not show the right speed until you've been at 30 for 50-100yds. Once you are steady and straight though, they are accurate to 1 mph or so, which is good enough for me and better than a speedo.

[hindle8907]

ok well i was on the motorway in a straight line for around 2 minutes and my speedo showed 102 and my gps showed 104 so this means my speedo is nearly bang on .

[MattCollins]

The GPS needs to take account of height. If you are going up a 1:4 on a twisty switchback road, distance driven != distance between points.


Considering GPS is 3 dimensional that'd be an interesting one to put to the test.

[AndyW]

Considering GPS is 3 dimensional that'd be an interesting one to put to the test.
Just because it can doesn't mean it does. I'll bet loads of someone elses money everything gets turned into latitude and longitude and that is what is used to calculate positional info, not the raw positional data.

[keith_d]

IIRC the Americans insisted on adding +/- 10m of positional inaccuracy to the civilian GPS signal, which doesn't bode well for measuring low speeds.

However, my Speedo is legally required to be wrong to allow for variations in tyre size and drive train components.

So I'd be willing to bet my Tomtom is more accurate than my speedo for straight line speed, though at a track a regular GPS wouldn't stand much chance.

Keith.

[MattCollins]

AndyW, That's why I said that it 'd be an interesting one to put to the test... I don't know... and wouldn't bet against you.

:)

[MattCollins]

Keith, if you are referring to SA, those days are long gone. There are however restrictions on speed and altitude (1000mph and 60000ft iirc), but somehow I don't think that is going to worry us too much.

Cheers

Edit: There are different ways to use the signals... http://en.wikipedia.org/wiki/Gps#Carrier_phase_tracking_.28surveying.29