[Proj] Using Proj.4 to compress a 10-parameter Molodensky-Badekas datum shift into 7 parameters

[support.mn at elisanet.fi]

Hello,

there is also an additional source off errors when several transformations
are done one after another ... and that is the order of evaluation. You might
try to swap the order. The order of evaluation may generate differences.
It is not the same to rotate something that is already shifted than something
that is not and there after shift the rotated value. Usually the difference is very
small as now seems to be the case.

Regards: Janne.

----------------------------------------

Mikael Rittri [Mikael.Rittri at carmenta.com] kirjoitti: 
> so we have achieved a lossless compression - apart from the 
> unexplained difference of about 1.1 cm in the ellipoidal height.  
>