[MetaCRS] CS-Map Conversion Errors
Hugues Wisniewski
hugues.wisniewski at autodesk.com
Thu Oct 27 04:28:35 EDT 2011
Hello Owen,
I'm sorry for the delay answering this.
The test points listed in your email are located in British Columbia, North of Juneau.So, the failing tests are all from the Canadian NTv2 test suite.
These tests should normally be commented out in the TEST.DAT version posted on osgeo. They are not so we could fix that with a simple ticket.
The reason is the Canadian .gsb file isn't part of the osgeo distribution.
There's the following comment in the test.dat file that helps understanding the situation:
# The following is the test suite required by Geomatics Canada. We must
# pass this test to maintain our license to use Version 2 of the
# Canadian National Transformation. Of course, all of these tests will
# fail if you don't have the Canandian Version 2 data file; so these are
# commented out in the CS-MAP distribution version.
So, you could either comment out all these test points, or, if you're interested in the area and its test points do the following in order to pass these test points:
- get the Canadian grid file and put it in the Canadian grid file folder
http://svn.osgeo.org/metacrs/csmap/trunk/CsMapDev/Dictionaries/Canada/
- in that folder you'll find a readme.txt file that provides some instructions
http://svn.osgeo.org/metacrs/csmap/trunk/CsMapDev/Dictionaries/Canada/ReadMe.txt
- update the geodetic transformation definition NAD27_to_NAD83 in the dictionary GeodeticTransformation.asc and add the grid file. The dictionary file will then have to be rebuilt into GeodeticTransformation.csd
In the old version of CS-Map, the one from roughly 2 years ago, this was done in the text file Nad27ToNad83.gdc. You'll also find some comments in there. This file is still posted with CS-Map for legacy reasons and will eventually be removed.
If the Geodetic Transformation Dictionary is properly updated to point to the .gsb file, these conversions will work fine.
You have control over the order of the grid files listed in the geodetic transformation NAD27_to_NAD83
By listing the Canadian .gsb file first, it will take precedence over the other files in the regions of overlap, which are substantial in this case (especially in and around Detroit Michigan).
I hope that helps
Hugues
-----Original Message-----
From: metacrs-bounces at lists.osgeo.org [mailto:metacrs-bounces at lists.osgeo.org] On Behalf Of Owen Wengerd
Sent: Friday, October 14, 2011 7:40 AM
To: metacrs at lists.osgeo.org
Subject: [MetaCRS] CS-Map Conversion Errors
I'm a first time user of CS-Map. In testing the CS_cnvrt function, I
noticed small errors (on an order of several hundred feet) in converting
from LL to UTM coordinates. When I ran the CS_Test utility, it also
reported conversion errors. To try narrowing down the cause, I downloaded a
fresh set of files via SVN, built on both 64-bit and 32-bit versions of
Windows 7, using both Visual Studio 2010 and Visual Studio 2008, and both
Release and Debug configurations. In every case the solution built
apparently without errors, but in every case I received errors when
subsequently running the unit tests. Here is some sample output:
cs_test /d..\..\Dictionaries ..\..\Test\Test.dat
====================================================
Verification failure (line 992):
LL27 -135.000000000 60.666666667
LL83 -135.001868469 60.666373600
CALC -135.001917855 60.666382663
Verification failure (line 993):
LL27 -135.083333333 60.666666667
LL83 -135.085204461 60.666372736
CALC -135.085249270 60.666379468
Verification failure (line 994):
LL27 -135.000000000 60.750000000
LL83 -135.001874553 60.749708753
CALC -135.001919581 60.749716895
Verification failure (line 995):
LL27 -135.083333333 60.750000000
LL83 -135.085210281 60.749707872
CALC -135.085251603 60.749713956
Verification failure (line 996):
LL27 -135.041666667 60.708333333
LL83 -135.043539478 60.708040700
CALC -135.043584578 60.708048245
Test 4 failed (973 failures detected)!
Measuring performance: 300000 coordinates; UTM27-13 ==> CO83-C
Converted 300000 coordinates from UTM27-13 to CO83C in 0.16 seconds.
Effective conversion rate = 1935483 conversions per second.
====================================================
Am I doing something wrong, or are these tests no longer valid?
--
Owen Wengerd
ManuSoft
_______________________________________________
MetaCRS mailing list
MetaCRS at lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/metacrs
More information about the MetaCRS
mailing list