<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<body text="#000000" bgcolor="#FFFFFF" link="#0000EE" vlink="#551A8B" alink="#FF0000">
<i>> Jean-Paul,</i>
<br><i>></i>
<br><i>> I get the same problem, and I have found it is a botched change
I made</i>
<br><i>> to the Convert_Geodetic_To_Geocentric() routine in September.
As far as</i>
<br><i>> I can tell this will always screw up any conversion from geodetic
to</i>
<br><i>> geocentric. I have correct the geocent.c file and
now I get values close</i>
<br><i>> to your reference position, but still off by 12 meters.
I am assuming this</i>
<br><i>> is due to the +towgs84 transform being only an approximation of
the correct</i>
<br><i>> transformation.</i>
<br><i>></i>
<br><i>> I have committed the fixed geocent.c, and also place it at:</i>
<br><i>></i>
<br><i>> <A HREF="ftp://gdal.velocet.ca/pub/outgoing/geocent.c">ftp://gdal.velocet.ca/pub/outgoing/geocent.c</A></i>
<br><i>></i>
<br><i>> I would appreciate your trying the fix and verifying things are
OK for you.</i>
<br><i>> I think I had better issue a PROJ.4.4.5 soon since this is a very
serious</i>
<br><i>> flaw in the datum shifting support.</i><i></i>
<p>Thank you Mark
<br>I've tried your fix and as you say, the results agree with reference
points by less than 12 meters -even less with other points-, which is more
than enough for me. And more, the +towgs84 specification now has
a normal behavior
<p>many thanks again - and happy new year (with version 4.4.5?)
<p>Regards
<p>Jean-Paul
<br>
<br><i></i>
<br><i></i>
<pre>--
----------------------
AIR project - INRIA - BP 105 - F78153 Le Chesnay Cedex
Tel: +33 1 3963 5703 Fax: +33 1 3963 5880
jean-paul.berroir@inria.fr</pre>
</body>
</html>