From: RAParker (RAParker_at_Quadzilla.NET)
Date: Mon Apr 11 2005 - 10:25:49 PDT
A couple things I agree with (and/or disagree with) on this topic. Last
month, I (very accurately) calibrated a street map of South Lake Tahoe,
California and had a heck of a time calibrating it, UNTIL I STOPPED
using a decimal format and started to use a degree format.
On Monday, April 11, 2005, at 06:17 AM, Victor Rehorst wrote:
>> Lat 34 26 17 (North)
>> Long -119 17 17 (West)
>
> Stop! You have to use decimal degrees as the input to the RECT
> programs.
>
Although the documentation states that the calibration programs take
either, decimal or degree, formats for data entry, I was ONLY able to
succeed in calibrating my maps (accurately) by using the DEGREE format
and RECT2 calibration program.
> You also need very accurate points.
Agreed...and I cannot emphasis accurate enough.
As suggested earlier in this thread, you should drive to the furthest
calibration points on your map and take a GPS reading. Write down (or
save) the values with a full 3 decimal point precision. I used TeleType
GPS because it allowed to add a note to each of the precise values I
gathered for calibration points.
Here is how my points looked like when I entered them into RECT2 (note
leading zeros and spaces):
Point #1
38 55 57.060
-120 01 04.740
Point #2
38 58 22.200
-119 54 31.380
Point #3
38 47 46.980
-120 01 08.580
In addition to recording these 3 far points, I also took 3 or 4
additional readings, at other random locations (in the middle of the
map), so I could verify the calibration program's calculations after
entering the data.
On your computer, use a graphic editor to determine the x/y locations
for each pixel that matches each position you've logged, again writing
them down.
Longitude changes with the X
Latitude changes with the Y
As noted in the MapPrep.pdf manual, your points should NOT create a
straight line and should be as far away from each other (the extreme
corners of your map) as possible. Your points can be in any order as
long as the don't create a straight line.
1. . . . . . . . .2
.
.
.
3
or
1. . . . . . . . .3
.
.
.
2
The manual also recommends that each point (in relation to each other)
should try to have a longitude or latitude point, in common with each
other. But, as long as they are close, form a near 90 degree corner,
exact matches aren't necessary.
NOTE: during my first attempts, I tried converting the degree format to
decimal using a converter on the web and entering the decimals into
RECT2. i.e.:
38 47 46.980 = 38.79638
-120 01 08.580 = -120.01905
Entering the decimals didn't properly calibrate my map and, in fact,
gave wild results with my other calibration points. Only by entering
degree w/spaces and leading zeros did I finally make it work.
Ron
--
Sent using Mail for Mac OS X. Cyberdog lives!
--
RAParker
|\/|\
|/-|/
|\ | @ Quadzilla.NET
--
From the "Who says, 'Size Doesn't Matter'" file:
Ever notice, when someone comments on your Newton PDA,
the first thing they say is, "Wow, that's a big one."
--
--
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/
This archive was generated by hypermail 2.1.5 : Mon Apr 11 2005 - 11:00:01 PDT