[gps-talkusers] Re: Earthmate to Map Calibration

Thanks Mike and thanks Chris for the neat idea of moving the poi's out.
What was and is still concerning me is that in each case where I have done
this test I come up with a differential of 950 to 1000 feet with the poi
always southwest of the GPS position.  That consistency in both distance and
direction  is disconcerting.  Also, a distance of approximately 0.2 miles
seems a lot.  Mikes explanation is very acceptable for the restaurant, as it
was in a strip mall, however, the parking lot did not seem to be 1000 feet
wide or long.  The explanation would be quite valid for the Amtrak station
in Emeryville, which had a much larger parking lot.  However, when it came
to the test in Fresno where I set the virtual position to an intersection of
two 2 lane streets and I was standing on the southeast corner of the
intersection and I am told that the intersection is nearly 1000 feet behind
me and to the southwest, I began questioning  calibration.

Rich Irwin
rich.irwin@xxxxxxxxxxx <mailto:rich.irwin@xxxxxxxxxxx>


-----Original Message-----
From: gps-talkusers-bounce@xxxxxxxxxxxxx
[mailto:gps-talkusers-bounce@xxxxxxxxxxxxx]On Behalf Of Kristeen Hughes
Sent: Saturday, August 28, 2004 9:07 AM
To: gps-talkusers@xxxxxxxxxxxxx
Subject: [gps-talkusers] Re: Earthmate to Map Calibration


It has amazed me to see how some places are nowhere near the mark and then
I'll get one that couldn't be better if I'd marked it myself. I have been
using the .bpi file simply for exploration of an area and once I've
correctly all of the things I want, I will often move that file out of the
poi folder so that I'm getting my own pois. I put it back if I'm doing
something new. Probably a bit of extra work, but I like using my own file
the best once it's been made.

Kris
----- Original Message -----
From: "Michael May" <MikeMay@xxxxxxxxxxxxxxxx>
To: <gps-talkusers@xxxxxxxxxxxxx>
Sent: Saturday, August 28, 2004 1:09 AM
Subject: [gps-talkusers] Re: Earthmate to Map Calibration


> Rich,
>
> This probably has little to do with the GPS accuracy and everything to do
> with the geo coding of the restaurant's address. We talk about this quite
> a
> bit in the manual.
>
> Restaurants are in the database based upon a computer projection of where
> they should be based upon averaging of address numbers on a given street.
> Nobody actually drives up to the restaurant and records a GPS position in
> front of the main doors other than you and I.
>
> So the database geo coding gives you a GPS position based on a computer's
> best guess of where the restaurant address may be. There is also the fact
> that many restaurants are not exactly on the street. They may be a ways
> back in a strip mall or sometimes the main door is on one street and the
> address is on a perpendicular street.
>
>
>
>
> At 08:02 PM 8/27/2004, you wrote:
>>I have now noticed a consistent difference, or should I say offset,
>>between
>>the GPS position and the defined POI position.  This has been observed in
>>three different locations as much as 200 miles apart in location.  Let me
>>use today's example to explain.
>>After dining at a restaurant, I was sitting outside waiting for a cab.
>>While sitting there I decided to investigate what was around me.  With the
>>Earthmate turned on and indicating that it was picking up 7 satellites I
>>pressed f and then enter to search for the poi's nearest me.  In going
>>down
>>the list, I came across the restaurant I was sitting in front of.  I set
>>this poi as my virtual position and checked the difference between the GPS
>>and virtual position.  The response was that the poi was 981 feet
>>southwest
>>of the GPS position.  As I said earlier, I have done this two other times
>>in
>>locations as far away from where I was today by as much as 200 miles
>>(specifically Fresno, Emeryville, and Novato, CA.) always with almost the
>>exact same results.  Could the Earthmate be out of calibration, or there
>>be
>>some sort of offset being interpreted from the maps by the Braille note?
>>Is
>>there any way I can close this offset?
>>Rich Irwin
>>rich.irwin@xxxxxxxxxxx
>
>

Other related posts: