<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Thanks Mikael. Looks like I made an error with the rotation signs. Your
PROJ parameters in (C) are the right ones. My faulty version (B) was
taken over from the EPSG file in the latest (4.7) PROJ, which should be
corrected too. I don't know how the parameters in the epsg-file are
derived form the EPSG database, but in this case, the derivation isn't
correct.<br>
<br>
The whole problem of datum conversions is a permanent pain in the head,
to express it mildly. Contributions like yours should be stored in some
sort of central place, e.g. a Wiki, instead of lying around all over
the Internet, together with all kind of faulty comments (like mine).
Perhaps you should have a look at <a class="moz-txt-link-freetext" href="http://trac.osgeo.org/metacrs/">http://trac.osgeo.org/metacrs/</a>, and
think about setting up such a wiki about datum conversions, not only
from and to WGS84, but also for other historical datums.<br>
<br>
Thanks for the trouble you have taken,<br>
<br>
Jan <br>
<br>
<br>
<br>
On 25-1-2010 10:20, Mikael Rittri wrote:
<blockquote
cite="mid:FAF6E56B26D2044696772EBE89B1535101BDE862@posty.carmenta.se"
type="cite">
<title></title>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<meta name="GENERATOR" content="MSHTML 8.00.6001.18854">
<div dir="ltr" align="left">
<div dir="ltr" align="left"><span class="973163508-25012010"><span
class="973163508-25012010"><span class="084220609-25012010"><font
face="Arial">Jan, you wrote:<br>
</font></span></span></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010">> </span>As
I said in a previous email:</div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">> </font></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">><span class="084220609-25012010"> </span></font></span><a
moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://www.ngi.be/FR/FR4-4.shtm">http://www.ngi.be/FR/FR4-4.shtm</a></div>
<div dir="ltr" align="left"> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">Thanks. </font></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">This page does not specify the rotation sign convention,
as far as I could see, </font></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">but I downloaded the converter. Then I tried to convert
the sample point 2°30' E, <span class="084220609-25012010">51</span>°N </font></span></div>
<div dir="ltr" align="left"><font face="Arial"><span
class="973163508-25012010">(west Belgium) from </span><span
class="973163508-25012010">BD72 to WGS84<span
class="084220609-25012010">, and compared the results with PROJ.4.</span></span></font></div>
<div dir="ltr" align="left"><span class="973163508-25012010"></span> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">(A) Older transform, according to Jan.<br>
<span class="084220609-25012010"> </span>FWTools2.2.8> cs2cs
+proj=longlat +ellps=intl
+towgs84=-99.059,53.322,-112.486,0.419,-0.830,1.885,-1 +to
+proj=longlat +datum=WGS84<br>
<span class="084220609-25012010"> </span>2.5 51<br>
<span class="084220609-25012010"> </span>2d30'4.369"E
50d59'58.092"N 41.378</font></span></div>
<div> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">(B) Newer transform according to Jan, assumes </font><a
moz-do-not-send="true" href="http://www.ngi.be/FR/FR4-4.shtm"><font
face="Arial">www.ngi.be/FR/FR4-4.shtm</font></a><font face="Arial">
uses Position Vector Transform:<br>
<span class="084220609-25012010"> </span>FWTools2.2.8> cs2cs
+proj=longlat +ellps=intl
+towgs84=-106.8686,52.2978,-103.7329,-0.3366,0.457,-1.8422,-1.2747 +to
+proj=longlat +datum=WGS84<br>
<span class="084220609-25012010"> </span>2.5 51<br>
<span class="084220609-25012010"> </span>2d30'1.464"E
50d59'57.151"N 41.359</font></span></div>
<div> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">(C) Newer transform, but assuming that </font><a
moz-do-not-send="true" href="http://www.ngi.be/FR/FR4-4.shtm"><font
face="Arial">www.ngi.be/FR/FR4-4.shtm</font></a><font face="Arial">
uses Coordinate Frame Rotation:<br>
<span class="084220609-25012010"> </span>FWTools2.2.8> cs2cs
+proj=longlat +ellps=intl
+towgs84=-106.8686,52.2978,-103.7329,0.3366,-0.457,1.8422,-1.2747 +to
+proj=longlat +datum=WGS84<br>
<span class="084220609-25012010"> </span>2.5 51<br>
<span class="084220609-25012010"> </span>2d30'4.373"E
50d59'58.093"N 41.455</font></span></div>
<div> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">(D) Result from Dutch <span class="084220609-25012010">converter</span>,
downloaded via </font><a moz-do-not-send="true"
href="http://www.ngi.be/FR/FR4-4.shtm"><font face="Arial">www.ngi.be/FR/FR4-4.shtm</font></a><br>
<font face="Arial"><span class="084220609-25012010"> </span>2°30'04.37266"
E, 50°59'58.09273" N </font></span></div>
<div> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">Distance (A) to (D):<span class="084220609-25012010"> </span>0.08
m<br>
Distance (B) to (D):<span class="084220609-25012010"> </span>63.75 m <br>
Distance (C) to (D):<span class="084220609-25012010"> </span>0.01 m </font></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010"><span
class="084220609-25012010"><font face="Arial">(all measured by <a
moz-do-not-send="true" href="http://williams.best.vwh.net/gccalc.htm">http://williams.best.vwh.net/gccalc.htm</a>)</font></span></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010"><span
class="084220609-25012010"></span></span> </div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial">I conclude that </font><a moz-do-not-send="true"
href="http://www.ngi.be/FR/FR4-4.shtm"><font face="Arial">www.ngi.be/FR/FR4-4.shtm</font></a><font
face="Arial"> uses the Coordinate Frame Rotation, just like <br>
EPSG says in EPSG:15929, "BD72 to WGS 84(3)", and therefore, the three
rotation<br>
angles should be reversed when constructing a +towgs84 clause for
PROJ.4, as in (C) above.<br>
(<span class="084220609-25012010">EPSG has in fact reversed all
parameter signs given in FR4-4, but that's because FR4-4 gives </span></font></span></div>
<div dir="ltr" align="left"><span class="973163508-25012010"><font
face="Arial"><span class="084220609-25012010"> the</span></font></span><span
class="973163508-25012010"><font face="Arial"> <span
class="084220609-25012010">parameters _from_ ETRS89 (or WGS84) _to_
BD72.) </span><br>
<br>
Best regards,<br>
</font></span></div>
</div>
<div><font size="2">--<br>
Mikael Rittri<br>
Carmenta AB<br>
SWEDEN<br>
<a class="moz-txt-link-abbreviated" href="http://www.carmenta.com">www.carmenta.com</a></font> <br>
</div>
<div dir="ltr" class="OutlookMessageHeader" align="left" lang="en-us">
<hr tabindex="-1"><font face="Tahoma" size="2"><b>From:</b> Jan
Hartmann [<a class="moz-txt-link-freetext" href="mailto:j.l.h.hartmann@uva.nl">mailto:j.l.h.hartmann@uva.nl</a>] <br>
<b>Sent:</b> Friday, January 22, 2010 4:04 PM<br>
<b>To:</b> Mikael Rittri<br>
<b>Cc:</b> PROJ.4 and general Projections Discussions; Thibaut Gheysen<br>
<b>Subject:</b> Re: [Proj] Belge 1972 / Belgian Lambert 72 (31370) -
towgs84parameters<br>
</font><br>
</div>
<br>
<br>
On 22-1-2010 15:40, Mikael Rittri wrote:
<blockquote
cite="mid:FAF6E56B26D2044696772EBE89B1535101BDE7F4@posty.carmenta.se"
type="cite">
<meta name="GENERATOR" content="MSHTML 8.00.6001.18854">
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial">Jan, you wrote </font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"></span> </div>
<div dir="ltr" align="left"><span class="131363214-22012010">> </span>All
my observations about the towgs parameters have been tested with real
Belgian coordinates overlaid over Google.</div>
<div dir="ltr" align="left"> </div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial">Fine. But if you did make an error when transcribing the
newer transforms into Proj.4 syntax, and </font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial">the error was just in the rotation angles, then the </font></span><span
class="131363214-22012010"><font face="Arial">effect may be too small
to notice. </font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"></span> </div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial">I mean, the angles describe a rotation, so there is
presumably a fixed point in Belgium where </font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial">the rotation has no effect. If your test points happened
to be near the fixed point, the </font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial">effect of the reversed angle signs could be very small. </font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"></span> </div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial"><font face="Times New Roman">> As I said before, the
EPSG database is not very systematical in the way it represents its
formulas, </font></font></span></div>
<div dir="ltr" align="left"><span class="131363214-22012010"><font
face="Arial"><font face="Times New Roman">> it just takes them from
whatever source is available, untested AFAIK. Never use them without
testing them with real world coordinates.</font></font></span></div>
<div><span class="131363214-22012010"></span> </div>
<div><span class="131363214-22012010"><font face="Arial">I agree
completely. But I am not yet convinced that you are right and EPSG
wrong for the newer transforms,</font></span></div>
<div><span class="131363214-22012010"><font face="Arial">although
that is certainly possible. Do you have some link to the offician
Belgian documents? </font></span><span class="131363214-22012010"></span></div>
<div dir="ltr" align="left"><font face="Arial"><br>
</font></div>
</blockquote>
<br>
As I said in a previous email:<br>
<br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://www.ngi.be/FR/FR4-4.shtm">http://www.ngi.be/FR/FR4-4.shtm</a><br>
</blockquote>
</body>
</html>