Many FlightGear constellations have the wrong shapes; in particular Orion's belt has the shape of a "V" rather than a straight line! It seems that whoever built your star database (the data/Astro/stars file in CVS) made the error of applying a negative sign only to the units, and not the fractions, of the degrees field. So when trying to turn the declination of, say, Alnilam - the middle belt star - from degrees into radians, they read
-1:12:07
as though it meant
"minus one degree, *plus* twelve arcminutes, *plus* seven arcseconds"
rather than reading it properly as either:
"minus one degree, minus twelve arcminutes, minus seven arcseconds" or "minus (one degree plus twelve arcminutes plus seven arcseconds)".
I recommend rebuilding the star database from scratch rather than trying to repair these errors, since the stars file uses very low precision and trying to correct these positions will compound that error.
I will be happy to contribute a rebuilt star database (my PyEphem package can read the xephem star database natively and output the results in radians); just let me know where to send it. :-)
I was hoping some one with some actual knowledge about this subject would step in, but anyhow. If you are positive the shapes are wrong it would be great to have an updated star database. You have to make sure the modified data could be licensed under the GPL though.
But it might (just might) have something to do with the way they are drawn. Are you sure that is not the problem?
Erik
_______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
