[Proj] cs2cs's +towgs84 usage

[Noel Zinn]

Gotcha.  Frank answered your question yesterday:

 

The parameters are always expressed in terms of what it takes to transform
to WGS84.  If PROJ is transforming from WGS84 it will internally reverse the
sense of the transformation. The parameters are always expressed in terms of
what it takes to transform to WGS84.  If PROJ is transforming from WGS84 it
will internally reverse the sense of the transformation.

 

And you can test this as you propose.

 

But this test tells you nothing about the correct rotation sense to use.
For that you need an external test (truth) point with coordinates in both
datums.

 

Noel

 

  _____  

From: proj-bounces at lists.maptools.org
[mailto:proj-bounces at lists.maptools.org] On Behalf Of Ilumas
Sent: Saturday, April 24, 2010 7:52 AM
To: PROJ.4 and general Projections Discussions
Subject: Re: [Proj] cs2cs's +towgs84 usage

 

Sure,

but that's not the point. The original question was if the values need to be
negated when going FROM wgs84 TO something else. If that was correct, one
should do the round trip as follows:

1.  cs2cs +proj=utm +zone=a +datum=wgs84  +to +proj=utm +zone=a
+datum=datum2 +towgs84 dx,dy,dz
input: x1,y2
output: x2,y2 
2.  +proj=utm +zone=a +datum=datum2 +towgs84 (-1)dx,(-1)dy,(-1)dz +to
+proj=utm +zone=a +datum=wgs84  
input: x2,y2
output:x1,y1

Or am I completely lost here?

Thanks for your patience...

I



On Sat, Apr 24, 2010 at 3:06 PM, Noel Zinn <ndzinn at comcast.net> wrote:

Not true.  A round trip will always close.  You’ll have learned nothing
about the correct rotation sense.  You can test this yourself.  Do one round
trip with the rotations unchanged and do another round trip with the
rotations negated.  They’ll both close on the original point.  Correctness
is determined by the intermediate coordinates (right or wrong), not by the
round-trip closure.  –Noel Zinn

 

  _____  

From: proj-bounces at lists.maptools.org
[mailto:proj-bounces at lists.maptools.org] On Behalf Of Ilumas
Sent: Saturday, April 24, 2010 3:02 AM


To: PROJ.4 and general Projections Discussions
Subject: Re: [Proj] cs2cs's +towgs84 usage

 

Well, I agree that for assessing the accuracy of the re-projection results
one needs a reference, but that should not be necessary for understanding
the behavior of the cs2cs or the towgs84 string. Therefore I still argue
that the test I suggested below is a valid way to test whether the parameter
values need to be negated or not. Anyone?

 

ilumas

On Fri, Apr 23, 2010 at 3:39 PM, Jean-Claude REPETTO <jrepetto at free.fr>
wrote:

Le 23/04/2010 14:24, Ilumas a écrit :

>
> Anyway, I believe that the behaviour of a certain towgs84 string is easy

> to test with cs2cs itself: just do the transformation to one direction

> (e.g. from WGS84) and convert the resulting coordinates back to the
> original datum using exactly the same parameter string. If the resulting
> coordinates equal to the original ones, no negation is needed. Right?
>
> ilumas

Hi,nding the behavious of the cs2cs

I think you are wrong. To check the parameters, you need to know the
coordinates of at least one point in both datums, and compare them with
the results of cs2cs.

Jean-Claude

_______________________________________________
Proj mailing list
Proj at lists.maptools.org
http://lists.maptools.org/mailman/listinfo/proj

 


_______________________________________________
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.maptools.org/pipermail/proj/attachments/20100424/3ab9d5a1/attachment.htm