[OSRS-PROJ] Geocentric systems and ellipsoids

[czaht at ocag.ch]

Gerald,

I see your point. To me it was also confusing that I was supposed to provide
the
switch "proj=geocent" in order to transform from something to a 3D system. I
was
a matter of trial-and-error and looking at the sources to reach the
unterstanding
how to use it in this particular case in the first place.

Regards Thomas

-----Original Message-----
From: Gerald Evenden [mailto:gerald.evenden at verizon.net] 
Sent: Mittwoch, 14. Januar 2004 15:43
To: osrs-proj at remotesensing.org
Subject: Re: [OSRS-PROJ] Geocentric systems and ellipsoids


PROJ's derived its name from "projection," that operation relating a  
point
on a sphere or ellipsoid to a two dimensional, cartesian system.  What   
seems to
be discussed below is the transformation between three dimensional systems
and thus has no relation to the projection operation in the original and
definition of PROJ's operation.

Sorry, this mixing of apples and oranges within the OSRS distribution of
PROJ has always been a sore point with me.  It is also the prime bottleneck
in updating OSRS PROJ with the expanded libproj4 material---the later stays
within the initial boundaries.

Alas.

On Jan 13, 2004, at 9:03 AM, Frank Warmerdam wrote:

> czaht at ocag.ch wrote:
>> Hi,
>> I have a question regarding the handling of geocentric system
>> libproj. It
>> seems that I need to define an ellipsoid together with a geocentric  
>> system,
>> in my case the WGS84 Cartesian
>> system. Why? In my opinion, a geocentric system is defined by the  
>> centre of the earth and
>> a rotation relative to the WGS84 Cartesian system.
>> Could someone shed a bit of light on this matter?
>
> Thomas,
>
> The geocentric support in PROJ.4 (as opposed to Gerald's newer
> libproj4)
> is quite new and somewhat experimental.  The need for an ellipse is  
> really
> an artifact of internal assumptions of the library before geocentric  
> support
> was added.
>
> Just select the same ellipsoid/datum as the output coordinate system
> you
> will be transforming to (or the input system you are transforming  
> from).
> You are correct that the geocentric coordinates define the location  
> without
> any real need for an ellipse.
>
> Best regards,
>
> --
> --------------------------------------- 
> +--------------------------------------
> I set the clouds in motion - turn up   | Frank Warmerdam,  
> warmerdam at pobox.com
> light and sound - activate the windows | http://pobox.com/~warmerdam
> and watch the world go round - Rush    | Geospatial Programmer for Rent
>
>
> ----------------------------------------
> PROJ.4 Discussion List
> See http://www.remotesensing.org/proj for subscription, unsubscription 
> and other information.
>
>
_____________________________________
Jerry and the low riders: Daisy Mae and Joshua

----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription and
other information.
----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.