Iphone 4 GPS accuracy vs Cycle computer

2»

Comments

  • Bar Shaker
    Bar Shaker Posts: 2,313
    Perhaps the way GPS works needs to be explained for you to understand the inaccuracy and why its not important.

    The satellites carry atomic clocks. The satellite sends out a signal saying which satellite it is and what its atomic time clock says is the time. There are many satellites around the Earth doing the same and your GPS unit will take the nearest one and use it as a reference point. From a data table, it knows where it is. The second satellite that is picked up will have a different position and this is cross referenced against the time signal received and the time signal that the GPS unit is expecting. There will be a slight delay and the length of this delay determines how far the unit is from the satellite. The third satellite signal means a 2D fix can be established and the fourth satelite means a 3D fix can be established... all by looking at the delay between the time stamp and the time it was received.

    a 15m error is quite small when we are timing something that is travelling at 299,792,458m/s but it is significant and the equipment is actually more accurate than that if it has more satellites within view. It uses the averages between the signals and refines the position of the unit. The more the unit can listen to at anyone time, the more accurate it will be. But it will never be 100% accurate because of som physics that was only theoretical when the GPS system was launched. The clocks had to be slowed down because the satellites are travelling (a small bit) closer to the speed of light than we are. What wasn't understood is they must then be sped up a bit again as gravity slows time down. The Earth has different gravitational strengths at different parts of the surface and this is one of the reasons why the accuracy isn't the 90mm that the precision of the clocks should provide. The point about this inaccuracy is that it is reasonably constant for where you are.

    By this, I mean you are always a set amount away from your expected spot and in the same direction away, as the time stream from one or more satellites is being slowed more than from the others. So whilst you may be up to 15m out, your position will be offset by the same amount and same distance rather than the unit thinking it is ripping across the field at close to the speed of light... whilst you are standing still.

    In the UK, the 'dilution of precision' as it is known, from a good spread of satellites, is normally around 3m-5m. This final difference is built up from receiver clock speed, gravity and error in the satellites' clocks.
    Boardman Elite SLR 9.2S
    Boardman FS Pro
  • Anonymous
    Anonymous Posts: 79,667
    Bar Shaker wrote:
    Perhaps the way GPS works needs to be explained for you to understand the inaccuracy and why its not important.
    People. Glass Houses.

    The 15m is not a constant offset, it is a random variable due to signal scattering by the atmosphere and geography.
  • bails87
    bails87 Posts: 12,998
    To be fair to yeehaa, I've left my GPS on while stood/sat completely still and ended up with what looks like a spirograph of trails around me.

    I don't know what causes it, but it happens. Probably a combination of all the factors. Although I did hear they were going to start replacing satellites with ones containing more accurate clocks, which would increase accuracy of positioning.
    MTB/CX

    "As I said last time, it won't happen again."
  • Bar Shaker
    Bar Shaker Posts: 2,313
    Bar Shaker wrote:
    Perhaps the way GPS works needs to be explained for you to understand the inaccuracy and why its not important.
    People. Glass Houses.

    The 15m is not a constant offset, it is a random variable due to signal scattering by the atmosphere and geography.

    Radio waves don't get scattered.

    If you don't have enough satellites in view, the unit will 'move' but its down to bad reception, not a failure of the technology.
    Boardman Elite SLR 9.2S
    Boardman FS Pro
  • thel33ter
    thel33ter Posts: 2,684
    I compared my HTC Hero's GPS track with another riders Cateye computer and the average was within 0.1mph, and the distance with within 20 metres, but I did do some extra little loops and stuff. (He was on a roadie, I was on my Hardrock with slicks)

    So I reckon its pretty accurate.
    And now you know, and knowing is half the battle
    05 Spesh Enduro Expert
    05 Trek 1000 Custom build
    Speedily Singular Thingy
  • Anonymous
    Anonymous Posts: 79,667
    Hang on, you actually rode a greater distance than your mate, they both came out the same, so you reckon they're accurate?
    And Bar shaker has the audacity to question MY intelligence?
  • thel33ter
    thel33ter Posts: 2,684
    As in riding in a few circles while stopped, nothing major, just wasn't very clear.
    And now you know, and knowing is half the battle
    05 Spesh Enduro Expert
    05 Trek 1000 Custom build
    Speedily Singular Thingy
  • Anonymous
    Anonymous Posts: 79,667
    Ah! :lol:
    Fair enough. I thought something properly odd was going on there.
  • bails87
    bails87 Posts: 12,998
    I think on a road type ride the difference between the two (whichever is 'right') will be smaller. Whereas on an off road, technical ride you'll have all those little tweaks of the front wheel, that might not be picked up by GPS, because they don't move the rider to the side very much, even though the front wheel is covering ground.
    MTB/CX

    "As I said last time, it won't happen again."
  • Anonymous
    Anonymous Posts: 79,667
    Off road you tend to get a load of tree cover too, which again can play havoc with GPS.