Hello Thomas! Thank you for your interest! There is nothing wrong with using VSOP87 for planetarium software for the present time, except if you want to steer a spacecraft to Uranus, then you should rather use DE405. ;-)
VSOP87 is nice because it almost immediately provides planet positions with relatively few data, while DE408 has several (tens ?) of MB of data. However, VSOP87 is limited to (I think) 4000BC..6000AD, and I don't know how well it behaves outside this time. This is where DE408, an (unpublished) extension of DE406, would come in handy for a few more millennia (I think down to 10000BC), which would fulfill almost the whole timespan needs for prehistoric astronomical research, except for ice-age cave art times. Of course, DeltaT for those times is completely unknown, models diverge by a few hours already in the 3rd mill. BC, so this should be handled in a flexible way. Data files for DE408 is available on some JPL site (Google!), but I did not have time yet to do something with them. The current time range for Stellarium going -100.000..100.000 is "a bit" optimistic, planetary positions will hardly be useful. I was playing with an early idea of an optional extension which, if data files are installed, would switch planet positions to DE408 for 10.000BC...2.000BC or so. It will not be something appreciated by many, I am afraid. But if you are aware of a tool that claims to accurately provide planetary positions and terrestrial orientation for longer times before VSOP, this may be also interesting to connect to! I an still not sure about the diverse reference frames and if all is implemented correctly in Stellarium, this would have to be verified. At least ecliptic obliquity is wrong in antiquity as of now. Kind regards, Georg On Di, 27.03.2012, 14:32, Thomas Morris wrote: > Hi Georg, > > Thank you for raising my awareness of new ephermides from JPL which is > very useful to know. I'd like to read around them and find out if any > are superior to our current methods (bearing in mind that Stellarium's > current VSOP87 algorithm I think is optimised for speed). But the > DE408 you mention is described as "a very long time span ephemeris" so > it may not be accurate enough for day-to-day usage. DE405 is the one I > have used before. > > Out of interest, did you have a timespan in mind? I know of a > numerical integration tool but that would be overkill, particularly as > the stellar positions are only accurate to ~10^4 years or so. > > Many thanks, > Thomas > > On 26 March 2012 19:02, Georg Zotti <georg.zo...@univie.ac.at> wrote: >> Hmm, if you are good in astronomy, maybe you could help fixing some real >> astronomical issues before going into games? It takes quite some time to >> find your way through the code until you understand how things work >> together. Not every enthusiastic programmer is knowledgable about >> astronomy, and tracking down errors in the astronomical model takes some >> time and knowledge. Things I immediately recall are the DeltaT, fixed >> ecliptic, lunar rotation, inner-planet phases and the bug about not >> being >> able to observe from the north pole. One extension useful for prehistory >> could be to replace VSOP by DE408 or maybe still something else. On >> usability, it would be great to have a tool that goes e.g. from sunrise >> to >> sunrise or moonrise to moonrise (keep moon on same altitude in east or >> west, resp., or even better, take the current landscape image into >> account), or a tool to find and display heliacal rising/setting of stars >> or/and planets. >> >> On the graphical side, extinction for meshes is still open, or better >> methods to match constellation figure textures. >> >> Just a few ideas. >> >> Kind regards, >> Georg >> -- DI Dr Georg Zotti Archaeoastronomy / ASTROSIM VIAS-Vienna Institute for Archaeological Science University of Vienna ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure _______________________________________________ Stellarium-pubdevel mailing list Stellarium-pubdevel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/stellarium-pubdevel
