Hi Jukka,

thank you for your posting!

Hello Micha, thanks for your prompt response. Please see my comments below.

few additional infos on the reporting:

Which modul do you mean - JAG3D, CoordTrans, GeoTra, FormFittingToolbox or each one?

My comments below are referring to coordinate transformations.

I'm used to evaluate the success of any adjustments (networks, transformations) by checking the residuals as well as standardized residuals (Baarda). With these figures one get's a good overview on what's really going on in the adjustment.

- residuals (vx, vy, vz) and sigma naught (m0) of a transformation

You found the residuals in the result-table indirectly. CoordTrans transformed the points from the startsystem into the targetsystem. So, the difference between both systems are the residuals.

This is clear, but it would be nice to have one additional table in the report containing transformed points, residuals, standard error of the unit weight m0, and possibly also the outlier test along with the estimated gross errors (GF)in the same table.

sigma naught is called σ(a-priori) in JAG3D (and moduls) - this value is fixed (σ = 1.0).

This is a good method for testing statistically the success of the adjustment. Statistical comparison of a priori and a posteriori variances gives indication on the given weighting and it also tells whether there are gross errors present in the data.

However, in addition to the above procedure a practical or a concrete figure would be necessary for evaluation of the success of the coordinate transformation. m0 is a good figure for that and is computed from the residuals (which I like so much). As we know m0 = SQRT(weighted square sum of residuals / redundancy of the adjustment). It is given in [m] and can be understood by any "man of the street". Statistical testing is important, so is computing of m0.

- printout of parameters a,b,c and d in addition to the current ones (Helmert)

CoordTrans do not calculate this values.

I believe you are using the standard formulas for computing coordinate transformations (e.g. Helmert). Usually in the Helmert adjustment the parameters a, b, c and d are estimated . a & b are the coordinate translations and c & d are the remaining two parameters that contain the scale factor and rotation angle.

I would like to have these four parameters, too, in the report as many times they are needed for subsequent transformations. It is easiest to use the original adjusted c & d parameters instead of readily calculated scale factor and rotation angle.

- rotation given in gon seconds instead of radian

I reconsider the printout...

Would be nice as it is quite dull to use Excel for computing the rotation in gon. So my wish is to have it in gon (e.g. 0.0007 gon), not in gon seconds.

Best regards
Micha

I look forward to hearing from you soon

Best regards

Jukka