I don't think the problems with accuracy stem from the limited number of
decimals in a /double/ variable
* The circumference of earth: ca. 40.076 km.
* 1 degree in meters 40076 km / 360 = 111.32 km/degree = 111320 m/degree.
* A double has 52 bits precision, roughly 15 digits (in 10 exponent).
That means that the decimal degree has a precision of around
111320/1000000000000000 = 1.1132e-10 m = (ca.) 1e-7 mm.
(That should be enough for everyone...)
* However, if you store your decimal degree value as a single
precision real with 23 bits precision (aka. 7 digits precision in 10
exponent), it's another story:
This will give you around: 111320/10000000 m = 0.011132 m = 11.132 mm
So store your decimal degree as doubles, not single precision real.
Regards
Bo Victor Thomsen
Municipality of Frederikssund, Denmark
Den 10/04/17 kl. 08:39 skrev Richard Duivenvoorde:
Hi List,
Had a discussion here with some people about the use of showing the
latlon degrees in degrees/minute/seconds tuple instead of the decimal
notation:
"4.40069397,51.71060887"
vs
"4°24′02.49807059″E,51°42′38.19184916″N" (or variations on that)
Is there some general preference in this? To me the first one is just
easier to handle, but I'm not sure what the general preference is in the
GIS world.
Any comments?
Or is everybody using the national crs's anyway :-)
Regards,
Richard Duivenvoorde
_______________________________________________
Qgis-user mailing list
[email protected]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
_______________________________________________
Qgis-user mailing list
[email protected]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
