Hi Bert,
You aren't alone in running into this problem. I heard about it
from Reinder Bouma yesterday. It affects observations made after
2016 January 1.5. To fix this, edit the file 'environ.dat' and
look for this line :
TIME_RANGE_OBS=1000,2016
You may not actually have that line, in which case you'd have
to add it, changing 2016 to (say) 2020.
This really ought to work even with older versions of Find_Orb,
but you may find that you have to go to
http://www.projectpluto.com/pluto/devel/find_orb.htm
and move to that version. (Which I would recommend doing anyway.
As you'll see, it has a variety of fixes and improvements.)
After Reinder e-mailed me about the 2016 bug, I implemented further
fixes for it (not yet posted). In the next update, the above line
will be replaced by a MAX_OBSERVATION_SPAN line, defaulting to 200
years. If your observations span more than that, you get a warning
message, and some observations are dropped.
Also, if _any_ observations are in the future, you get a warning
message about that. In both cases, the messages are pretty verbose,
and you'll know exactly what's going on and what can be done about it.
Both fixes match up to what I really should have done in the first
place. Find_Orb doesn't really have a problem with "distant" dates,
but it can take a very long time on multi-century arcs (to the point
where the user thinks the program has hung). It already warns you
about possible errors in the data, and really should warn you if
observations are provided for times that haven't happened yet.
I'm also investigating an odd issue Man-To Hui reported, in which
he got certain comet non-gravitational parameters when using MPC data
and quite different ones when using AstDyS data. I am pretty sure that
this is due to AstDyS having a very different weighting scheme, one
that is assigning excessively low sigmas to some observations, but am
still investigating. (For example, if three observations had near-zero
uncertainties, the orbit would basically be a fit to just those three
and the other data would be ignored.)
So I do want to fix the AstDyS issue, but the 2016 January 1.5
issue is probably going to be _very_ annoying; I expect to get the fix
described above posted later today, even if the AstDyS problem is
still unresolved.
-- Bill
On 01/03/2016 11:46 AM, blslcnm@... [find_orb] wrote:
> Is anyone else having problems processing 2016 observations? I have a file with NEOCP
> observations and when I try to load them into Findorb, I get a message saying that no
> observations were loaded from that file. It loads observations from 2015 just fine.