Celestia Forums

Hello.
I was going to post this in Bugs, but I don't think it's a bug... it's likely my computer. I don't recall this ever being a topic, but you folks will know better than I.

I noticed something odd today in the display of the moon's orbit (enabled), a quirky little bend. The link below leads to the image (in my Picasa web album).
https://picasaweb.google.com/108051224509230368846/CelestiaImages?authuser=0&authkey=Gv1sRgCMW3x_mJqfXkyAE&feat=directlink

Celestia v1.6.1
Toshiba Satellite P105 (old) with Intel Centrino Duo @ 1600mhz, 1gig mem (2gig virtual), with Mobile Intel 945 graphics chipset (Intel GMA 950 adapter) with a whopping 128mb mem
WinXP Professional, Svc pack 3, with more patches than months of life I've lived.

Any ideas?
Ferret (Alan)

--- edit ---

john Van Vliet wrote:is this a default install of 1.6.1
or some custom windows build

what were the last few windows installs ?
on Black Tuesday - for this month it was July 9

windows updated a bunch of things
including a few VERY VERY VERY buggy updates .They have been causing people all kinds of problems .

have you tried uninstalling this months updates and rolling bask xp to the pre update state

as i recall xp normally did not show the hot-fixes in the "install / remove" GUI
but i think sp3 added an option to show them , and removed the need for a third party tool to see the updates .

Yes, it is a default install. I have addons, but none are active. This laptop has only had WinXP Pro on it, pre-installed (Thursday, March 02, 2006, 9:14:08 AM); The only changes were 'normal' autoupdates (last 7/10/2013, security update and net framework svcpks 2&3). I've not tried any uninstalls or rollbacks... just reported what I ran across.

--- edit----

Because the moons position is affected by gravitational perturbations of Sun and planets, it's position relative to Earth will vary between the start of one orbit to the start of the next (all replicated in the VSOP-87 definition) ie. it won't end up at the same point it started, hence the kink .. it's just a kludge to make the ends meet.
Realistically they don't actually meet, the orbit-line would in fact be a continuous line..... so, not a bug, but a compromise.

I think there was some discussion about this in the past related to a different approach towards drawing orbit lines (ie. as a trailing fading line). That post may explain in more detail, but a quick search by me failed to find it, but you may have better luck.

Here's one of the previous posts about this:
http://216.231.48.101/forum/viewtopic.p ... 96&start=0

There has been more discussion but I haven't been able to find it yet.

As I recall, Chris or Selden once explained that the Moon's orbit is just calculated and drawn at startup time to unburden the CPU. But since the true lunar orbit is constantly changing, a simple ellipse (with joined ends) can't accurately represent it. To make the actual jog or discontinuity less "glaring", the transition was slightly smoothed out in version 1.60, though it is still noticeable.

As CC mentions above, there are constant perturbations due primarily to the gravitional pulls of Earth, the Sun and Earth's equatorial bulge. So the Moon's resultant path relative to Earth is highly irregular. If accurately traced even for just a few orbits, it looks more like an "oscilloscope track gone wild". Moreover, the perigee, the line of nodes and other orbital factors precess at different rates. So, because the Moon's actual position in Celestia is recalculated fairly often, the Moon fairly quickly strays away from the path that Celestia initially draws. Try the following and you'll see:

Start Celestia and go to the Moon. Turn on "Orbits", and you will see that the Moon is right on its displayed orbital path. Make sure you're "Following" the Moon and then reset Celestia's time to about six years in the future. Now the Moon will appear quite some distance from its displayed orbital path. It's not a bug or a problem with your computer. It's just a simplification to keep Celestia's performance from lagging.