Re: [Proj] wrong computation of meridinal distance / pj mlfn.c

[Peter Albert]

Hi Noel,

> Don't have a solution for you, but I have duplicated your 
> MuPAD computations 
> with similar numerical integrations in Matlab.  So, that's cool.

I also duplicated the computations in IDL, but this is very similar to
Matlab, so that's no wonder.

> also dusted off a meridional arc function from an old
> TM code and duplicated 
> Proj.4's computation as well.  So, that's cool, too.

I don't know what TM is, but if I understand you correctly you did the
elliptical integral here, not the polynomial approximation as in Proj? In
that case a look at the code would probably be helpful. Could you provide me
with this?

> Rather, my intuition suggests that either MuPAD and Matlab are 
> integrating over a prolate (egg-shaped) ellipsoid (not oblate)

Hm, I am lost here; my english-german dictionary gives the same translation
for both "oblate" and "prolate"... is there a way you could illustrate the
difference between both?

> or that the 
> zero latitude is interpreted differently (vertically?). 
> Don't know why that 
> would be so, but worth investigating.  Too late tonight, however.

Well, in the MuPAD calculations I for sure interpreted latitude as being 0°
at the equator and 90° at the pole, the elliptical integral I used for the
calculations is based on the definition of the ellipse x=a*cos(t) and
y=b*sin(t), so zero latitude is on the equator with x=a and y=0.


Regards,

     Peter

-- 
5 GB Mailbox, 50 FreeSMS http://www.gmx.net/de/go/promail
+++ GMX - die erste Adresse für Mail, Message, More +++