[Kstars-devel] Bug in moon position?

[Mike Rosseel]

Detecting regressions in calculations is the kind of case for which
unit tests are invaluable (CUnit for instance).
Maybe that's a good idea for next year's Summer of code ;)

Mike

On Fri, May 30, 2008 at 2:08 AM, Jason Harris <kstars at 30doradus.org> wrote:
> Ok, I just tried to reproduce your finding, but I am not seeing the
> positional offset you described.
>
> Location: Bangalore, India (77.3 E, 13.8 N)
> Local Time: 20:37 (UT: 15:07)
>
> KStars Moon position:            08h 12m 02s  +21d 47' 09"
> JPL Horizons Moon position:   08h 11m 35s +21d 48' 30"
>
> So the positions do disagree by about 6 arcminutes, which is pretty
> large, but not like what you saw.  My guess is that the discrepancy
> comes from the conversion from geocentric coordinates to topocentric
> coordinates.  We probably use a different model for the geometry of
> the Earth than JPL does.
>
> I'm not sure why you got a different position for the Moon.  Maybe you
> set UT=20:37 instead of local time?
>
> BTW, I see the occultation in my KStars window (see attached
> screenshot).  It's pretty funny that Mars is drawn in front of the
> Moon.  We'll have to fix that :)
>
> regards,
> Jason
>
> On Thu, May 29, 2008 at 3:15 PM, Jason Harris <kstars at 30doradus.org> wrote:
>> Ouch.  I'm surprised by this because in the past, KStars has been found to
>> do a good job of predicting solar eclipses, which obviously requires very
>> accurate moon positions.
>>
>> Maybe there's been some kind of regression in recent commits, but it's a bit
>> of a mystery, because the code to compute the moon's position hasn't been
>> changed AFAIK.
>>
>> I'll see if I can confirm the problem...
>>
>> thanks,
>> Jason
>>
>> On May 29, 2008, at 11:21 AM, Akarsh Simha wrote:
>>
>>> Hi
>>>
>>> I was just fiddling with the conjunction tool, while I noticed what
>>> seems like a glaring discrepency in KStars. The occultation of Mars by
>>> the Moon on May 10th 2008 never turned up. I went to that particular
>>> date and time and tried checking up the RA and Dec of Moon / Mars.
>>>
>>> While the RA/Dec of Mars is in good agreement with that generated from
>>> JPL's HORIZONS and CalSky.com, The moon's RA seems to have about half
>>> an hour's difference, and the Dec about half a degree difference! [I
>>> don't know if I messed up with something here!]
>>>
>>> Date/Time: 10th May 2008, 20:37 +05:30
>>> Location: Bangalore, India [77d35m E, ~13d N]
>>>
>>> Moon's RA/Dec reported by KStars:
>>> RA ~ 07h 41m, Dec ~ 21d 12m
>>>
>>> Moon's RA/Dec reported by HORIZONS:
>>> RA ~ 8h 11m, Dec ~ 21d 50m
>>>
>>> Could somebody verify this? What can this be due to?
>>>
>>> Regards
>>> Akarsh
>>> _______________________________________________
>>> Kstars-devel mailing list
>>> Kstars-devel at kde.org
>>> https://mail.kde.org/mailman/listinfo/kstars-devel
>>
>>
>
> _______________________________________________
> Kstars-devel mailing list
> Kstars-devel at kde.org
> https://mail.kde.org/mailman/listinfo/kstars-devel
>
>