[Proj] NTv2 info

[Clifford J Mugnier]

This phenomenon might be termed "hemispherical ethnocentric geodetics" complicated by ellipsoids ...
 
Once had to write a translator to get the extended character set to compile properly between Intel compilers and Unix compilers.
 
C. Mugnier
LOUISIANA STATE UNIVERSITY

________________________________

From: proj-bounces at lists.maptools.org on behalf of Gerald I. Evenden
Sent: Tue 02-Jun-09 14:41
To: Cuinet Jérôme
Cc: PROJ.4 and general Projections Discussions
Subject: Re: [Proj] NTv2 info



On Tuesday 02 June 2009 3:14:37 pm Cuinet Jérôme wrote:
> We can see the existence of a NTV2 Developer's Guide Junkins and Farley,
> 1995, but any link is broken. The closest document is a NTV2 User's Guide
> that you have pointed.
>
> This developper guide must be on some computers...
>
>
> How silly we are, using a canadian built-in format. Fortunately it wasn't
> an Autralian format ! By talking australians, they noticed that NTv2 format
> had seemed to be compiler dependant, although it is finally not compiler
> dependant : http://www.geom.unimelb.edu.au/gda94/NTv2_Binary.pdf
>

It is what seems to be most commonly call "endian" dependent.  That is; if a
character string of 8 bytes is overlayed on a type double word, is the first
byte of the string at the begining of the float word or at the end.  That is,
the byte string might be in reverse order.

In the case of reading the NTv2 format one must reverse order the bytes of the
float and double words in the file before they make sense to an Intel or
clone CPU (read "Microsoft").

In the case of the NGS NADCON files, an Intel machine can process them without
byte flipping.

Data transfer among different CPUs can be a *REAL* headache!

If it was an Australian format, latitude would be positive south of the
equator.  ;-)

--
The whole religious complexion of the modern world is due
to the absence from Jerusalem of a lunatic asylum.
-- Havelock Ellis (1859-1939) British psychologist
_______________________________________________
Proj mailing list
Proj at lists.maptools.org
http://lists.maptools.org/mailman/listinfo/proj


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20090602/497b5af7/attachment.html>