[Loadstone] MTK chipset, was Re: royaltek rbt 2110
rob at loadstone-gps.com
Sun Jun 24 18:55:17 BST 2007
When using programs like Wayfinder or Trekker this software wants to put
you in a street, not on top of a building. If you're put in the wrong
street, the faulty reading is as faulty as the distance between the street
you are in and the street the software thinks you are in. So, if these
streets are seperated by 40 meters, you're off by 40 meters, even though
your receiver may not be more off than 15 meters. When using Loadstone I do
not experience faults bigger than 20 meters in urban canyons and I'm afraid
that's what we have to live with using GPS as an orientational tool. In the
area I live in (a mix of low and high buildings seperated by open areas)
accuracy usually is between 10 and 15 meters. In good conditions this can
improve to between 5 and 10 meters, but that's about it. Again I would like
to warn for receivers that use averaging models and 'improvements after
reception'. So far these improvements have had no influence on the kind of
accuracy we are looking for.
At 6/24/2007, you wrote:
>; In theory, MPR is responsible for a plus minus 3 meters degradation of
>is this what you see in practice as well?
>I'd would say accuracy is degradated up to 40 meters in practice.
>I made some testing with another software which announces me in realtime
>on which street i am.
>Sometimes, on the same street, it gives me a parallel street in about 40
>meters, but as this happens not all the time, it cannot be a problem of map.
>Hence, i'd say it's a problem of reflexions.
>Loadstone mailing list
>Loadstone at loadstone-gps.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Loadstone