Hi Frank,
found a copy+paste bug in the copy constructor for CodingType:
...
HunitValue = ct.HunitValue;
HunitExponent = ct.HunitValue;
Seems to have no effect?
Gerd
________________________________________
Von: mkgmap-dev <[email protected]> im Auftrag von Frank
Stinner <[email protected]>
Gesendet: Freitag, 10. November 2017 13:20:21
An: [email protected]
Betreff: Re: [mkgmap-dev] DEM format Questions
Hi Gerd,
we have ever 2 problems: how to calculate the next value and how we have to
encoding this value.
#1: If we have not a plateau, then we have a "Standardwert". For the
calculation of a "Standardwert" we have to 3 differend cases depend on the
value hdiff(n, m-1) = h(n, m) - h(n-1, m) (horizontal difference, see at the
end of "Datenberechnung"). max is highest possible value in this 64x64-subtile.
#2: We have to decide, is it "Längencodierung", "Hybride Codierung" or if they
are not possible, "Binärcodierung für große Zahlen". But we have no choice, we
have to use that kind, that the algorithms say (see "Umschaltung der
Codierart"). That do in TileEncoder.cs the class CodingType and the derived
classes.
I'm sorry, but TileEncoder.cs is not written for speed or good understanding.
It is only for exploring DEM and, in Input2, for showing, what's going on.
There is much place for optimizing. ;)
Frank
_______________________________________________
mkgmap-dev mailing list
[email protected]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
