>"Petert" googlemail.com> wrote in message
>news:oqbad4tqqse5u15etr5v1sac9c4hpt2cog@4ax.com...

>> On Sat, 20 Sep 2008 13:09:53 +0100, "Mortimer" privacy.net> wrote:
>>
>> Someone explained to me once that the speed shown by a GPS receiver
>> isn't necessarliy accurate as you may be driving up or down a steep
>> hill - the GPS will show your speed as though you were driving along a
>> flat road

>
>Good point. Let's look at the error. On a 1:3 hill (probably the steepest to
>be encountered), the angle will be arctan(1/3) = 18 degrees. For a
>hypotenuse of 1 unit (the true distance that the car travels), the
>horizontal distance that the GPS senses will be cos(18 degrees) = 0.95
>
>So the GPS will read 95%% of the true speed and the error will be (1-0.95)/1
>= 5%%.
>
>I suppose a really clever GPS could calculate the gradient from the rate of
>change of altitude and apply a suitable correction factor.
>
>