[Proj] Accuracy Cassini Solder -> lat lon 200m?!

[Clifford J Mugnier]

Sebastian,
 
Whether you use EPSG parameters or any other set of parameters, you will never get exact conversion to the WGS84 datum from an old classical datum.  For all intents and purposes, WGS84 is "perfect" and the old systems developed with classical triangulation and widely-spaced baselines physically measured on the ground with invar tapes were never "perfect."  All of the datum shift algorithms are compromises ... none offer an exact conversion.  Adding additional parameters such as changing from a three-parameter transformation to a seven-parameter transformation will give a better conversion, but will still fall short of perfection.
 
As a general rule, for a given number of parameters for a transformation from a classical datum to an inertial datum such as WGS84, the smaller the area from which the parameters have been derived, then the higher the accuracy of the transformation for that given area.
 
National governments commonly offer the best compromise for such datum transformations ... if they so choose to do so.  That is because they have access to the entire database of all of their geodetic control, and rarely does anyone else.  

C. Mugnier
LOUISIANA STATE UNIVERSITY
 
P.S. Consider reading some of my past columns - they may help understanding some of this material.
www.ASPRS.org/resources/grids 

________________________________

From: proj-bounces at lists.maptools.org on behalf of Sebastian Schubert
Sent: Mon 06-Jul-09 05:27
To: proj at lists.maptools.org
Subject: Re: [Proj] Accuracy Cassini Solder -> lat lon 200m?!



Dear all,

* On Friday 03 July 2009, 11:39:31, Jean-Claude Repetto wrote:
> Sebastian Schubert a écrit :
> > I want to convert from Solder system used in the city of Berlin
> > (EPSG:3068) to the standard lat lon system.
>
> What do you mean "standard lat lon system" ? If you mean WGS84, you
> should try :
> cs2cs -v +init=epsg:3068 +to +init=epsg:4326

I would like to understand this a bit more. I know this is probably not
the perfect place to ask but I do not know where to ask. I would be
also very happy to be given a hint where to find the solution myself.

As far as I understand, the standard settings, I applied before:

schubert at linux-8ubn:~> cs2cs -v +init=epsg:3068
...
# ---- To Coordinate System ----
#Lat/long (Geodetic alias)
#
# +proj=latlong +datum=potsdam +ellps=bessel +towgs84=606.0,23.0,413.0

refer to the EPSG::6314 Deutsches Hauptdreiecksnetz. The not vanishing
values of towgs84 show that this is not wgs84. This gives the same
results as the formulas found in the epsg guide
http://www.epsg.org/guides/G7-2.html
when using the parameters for the ellipsoid given in the epsg database
for EPSG::6314 Deutsches Hauptdreiecksnetz.

However, I thought that when I replace these parameters with the ones of
wgs84, I should actually get wgs84 coordinates. This does not seem to
be true. Does this mean that eg the ellipsiod has another centre? So
the formulas given in the guide are only applicable for exactly one
Base Geodetic CRS (in this case DHDN)? So proj does a two step
conversion when I want to convert to wgs84: convert to Base Geodatic
system, then convert to wgs84? Are the formulas for the latter given
somewhere?

Thanks a lot!
Sebastian


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.maptools.org/pipermail/proj/attachments/20090706/a1305683/attachment.htm