Displaying orbits

Discussion forum for Celestia developers; topics may only be started by members of the developers group, but anyone can post replies.
ajtribick
Posts: 1780
Joined: 11.08.2003
With us: 16 years 3 months
Location: Switzerland

Re: Displaying orbits

Post #21by ajtribick » 07.02.2010, 13:33

Having played around with the code, I still have to say I still don't like the effect for nearly elliptical orbits, though it would certainly be useful for things like orbits of moons displayed in a heliocentric frame. Having the fade effect ending at a different point to the planet looks wrong too: I thought it was a bug until I looked more carefully at the code and saw the parameter.

One reason I decided to release the Gliese 876 add-on was to provide an example of a rapidly-evolving system to investigate how well the fade effect represents it: unfortunately no fading is applied to the xyzv trajectory (no ability to specify an orbital period here).

Topic author
chris
Site Admin
Posts: 4211
Joined: 28.01.2002
With us: 17 years 9 months
Location: Seattle, Washington, USA

Re: Displaying orbits

Post #22by chris » 08.02.2010, 20:16

ajtribick wrote:Having played around with the code, I still have to say I still don't like the effect for nearly elliptical orbits, though it would certainly be useful for things like orbits of moons displayed in a heliocentric frame. Having the fade effect ending at a different point to the planet looks wrong too: I thought it was a bug until I looked more carefully at the code and saw the parameter.

One reason I decided to release the Gliese 876 add-on was to provide an example of a rapidly-evolving system to investigate how well the fade effect represents it: unfortunately no fading is applied to the xyzv trajectory (no ability to specify an orbital period here).

It's not a problem to enable fading for non-periodic trajectories provided that there's some way for either users or add-on creators to specify an appropriate time window. We should add a Period parameter to SampledTrajectory for cases when an xyzv file is used for an approximately periodic orbit like in your Gliese 876 add-on. For non-periodic trajectories, I'm not sure... One possibility would be to offer provide the user with two settings: a duration of visibility for all non-periodic orbits, and a boolean setting that can override the duration of visibility and show the entire orbit. A more comprehensive plotting UI would also be useful, but it seems like we should include some easy-to-use settings that cover common usages.

--Chris

abramson
Posts: 404
Joined: 22.07.2003
With us: 16 years 3 months
Location: Bariloche, Argentina

Re: Displaying orbits

Post #23by abramson » 09.02.2010, 15:48

Hi. I have noticed that the code for orbit fading is included in the latest SVNs. How can activate it for testing? Is it a switch in the compiler? (I'm using VC++ 2008 Express, Windows 7). Thanks.
Guillermo

Topic author
chris
Site Admin
Posts: 4211
Joined: 28.01.2002
With us: 17 years 9 months
Location: Seattle, Washington, USA

Re: Displaying orbits

Post #24by chris » 09.02.2010, 17:17

abramson wrote:Hi. I have noticed that the code for orbit fading is included in the latest SVNs. How can activate it for testing? Is it a switch in the compiler? (I'm using VC++ 2008 Express, Windows 7). Thanks.

Guillermo,

At the moment, you have to edit render.cpp. There are are three constants you can adjust, all pretty well documented in the code starting at line 1953. For a start, try setting OrbitWindowEnd to 0.0 and LinearFadeFraction to 0.5.

Then, if you want to visualize orbits over multiple revolutions, adjust the setting of OrbitalPeriodsShown to a value greater than 1.0. When viewing multiple orbits, I've found that it looks better when fading occurs over the whole path, so set LinearFadeFraction to 1.0.

--Chris

abramson
Posts: 404
Joined: 22.07.2003
With us: 16 years 3 months
Location: Bariloche, Argentina

Re: Displaying orbits

Post #25by abramson » 09.02.2010, 18:32

Thanks. Done. Beautiful. It also serves to see if you are watching from the north or from the south.
Guillermo

duds26
Posts: 328
Joined: 05.02.2007
Age: 29
With us: 12 years 9 months
Location: Europe

Re: Displaying orbits

Post #26by duds26 » 11.02.2010, 20:37

I would like to see a tabbed render menu.
With tabs like general, orbits.
Because the options will only increase in the future.
I also second Guckytos opinion about general rules and per-object override.

BobHegwood
Posts: 1803
Joined: 12.10.2007
With us: 12 years 1 month

Re: Displaying orbits

Post #27by BobHegwood » 11.02.2010, 23:38

duds26 wrote:I would like to see a tabbed render menu.
With tabs like general, orbits.
Because the options will only increase in the future.
I also second Guckytos opinion about general rules and per-object override.

For whatever it's worth here, I agree with duds, especially concerning the tabbed render menu. :wink:

Thanks, Brain-Dead
Brain-Dead Geezer Bob is now using...
Windows Vista Home Premium, 64-bit on a
Gateway Pentium Dual-Core CPU E5200, 2.5GHz
7 GB RAM, 500 GB hard disk, Nvidia GeForce 7100
Nvidia nForce 630i, 1680x1050 screen, Latest SVN

lithium3
Posts: 8
Joined: 25.08.2009
With us: 10 years 2 months

Re: Displaying orbits

Post #28by lithium3 » 14.02.2010, 02:02

I cant find the *.CPP

and i get a run time error if i pout it in the *.cfg file it loocks so cool i just cant see it
eney help?

I'm not a Developer of celestia i only do *.SSC textures 3DS and scripts. :?:

abramson
Posts: 404
Joined: 22.07.2003
With us: 16 years 3 months
Location: Bariloche, Argentina

Re: Displaying orbits

Post #29by abramson » 14.02.2010, 14:05

lithium3 wrote:I'm not a Developer of celestia i only do *.SSC textures 3DS and scripts. :?:
Li, that explanation was to make a custom compilation of Celestia, for which you know all the source files (the cpp's, h's, etc), plus a C++ compiler. If you are not into it, you can safely wait a little and the feature will come to a public release. The fading orbits are nice, but you can definitely wait for them.

Guillermo

PS: in the wikibook there are instructions to compile celestia yourself as well as in my page (for windows).

Topic author
chris
Site Admin
Posts: 4211
Joined: 28.01.2002
With us: 17 years 9 months
Location: Seattle, Washington, USA

Re: Displaying orbits

Post #30by chris » 15.02.2010, 20:30

Another orbital evolution example... 20 orbits of the International Space Station are shown here, with fading enabled to indicate time elapsed. The precession of ISS's inclined orbit is very apparent here.

iss-orbits.jpg


Note that for this image, I did not use the Celestia's default ISS orbit, which is just a simple ellipse. I'm testing some new code that calculates object positions from NORAD two-line elements. You can see that simple elliptical elements are inadequate for determining the location of satellites in low Earth orbit--the 20 orbits shown in the image are the trajectory of ISS over 30 hours.

--Chris

Avatar
Hungry4info
Posts: 1133
Joined: 11.09.2005
With us: 14 years 2 months
Location: Indiana, United States

Re: Displaying orbits

Post #31by Hungry4info » 15.02.2010, 21:19

What causes that precession of ISS's orbit? Oblateness of Earth?
Current Setup:
Windows 7 64 bit. Celestia 1.6.0.
AMD Athlon Processor, 1.6 Ghz, 3 Gb RAM
ATI Radeon HD 3200 Graphics

Topic author
chris
Site Admin
Posts: 4211
Joined: 28.01.2002
With us: 17 years 9 months
Location: Seattle, Washington, USA

Re: Displaying orbits

Post #32by chris » 15.02.2010, 21:47

Hungry4info wrote:What causes that precession of ISS's orbit? Oblateness of Earth?

Yes, the Earth's oblateness. The rate of precession is dependent on the inclination and semi-major axis. This is to good effect in sun-synchronous orbits, where the satellite's orbit precesses at the same rate the Earth orbits the Sun.

--Chris

CAP-Team
Posts: 194
Joined: 27.12.2006
Age: 44
With us: 12 years 10 months
Location: Vriezenveen, the Netherlands
Contact:

Re: Displaying orbits

Post #33by CAP-Team » 17.02.2010, 23:29

I've been experimenting with orbits of spacecraft, for example, Cassini.

For each spice kernel, I entered the orbital period in days. Celestia now always wants to close the orbit, while displaying just the part the kernel covers would be nicer.
Any chance this can be improved (or use a switch or something)

Another idea would be to draw the orbit of a body a selectable amount of days in front and behind a body.
Windows 7 Ultimate x64, Intel Core i7 2600K 3.4 Ghz, 4 GB RAM, 120 GB SSD + 1 TB hdd, nVidia GTX460 1 GB, Celestia 1.6.0.xxxx
Download my latest SVN Build

MKruer
Posts: 501
Joined: 18.09.2002
With us: 17 years 2 months

Re: Displaying orbits

Post #34by MKruer » 05.04.2010, 22:51

Chris,

I really like idea of fading, however I was also wondering if its possible to add ticks to the line. Something to see that the body is moving in addition to the fade.

ticks depending on distance viewable could be in second, minute, hour, day, year, decade, etc...

Just an idea

scroll
Posts: 34
Joined: 30.12.2005
With us: 13 years 10 months

Re: Displaying orbits

Post #35by scroll » 07.04.2010, 22:15

How do you turn fading on/off? I have solid line orbits and can find no reference to fading and orbit paths in 1.6.0. Thanks.

abramson
Posts: 404
Joined: 22.07.2003
With us: 16 years 3 months
Location: Bariloche, Argentina

Re: Displaying orbits

Post #36by abramson » 07.04.2010, 22:34

scroll wrote:How do you turn fading on/off? I have solid line orbits and can find no reference to fading and orbit paths in 1.6.0. Thanks.
Hi, Scroll. Fading orbits are still not in the official release version of Celestia. You will have to wait until the next version.

You can try this feature by compiling Celestia yourself from the development sources. If you use Windows and you don't know how to compile Celestia, you can use one of my pre-compiled ones. Download from here the most recent one and unpack inside your Celestia folder (take the precaution of renaming the executable, so you do not overwrite the celestia.exe you already have there; any name will do).

Guillermo

scroll
Posts: 34
Joined: 30.12.2005
With us: 13 years 10 months

Re: Displaying orbits

Post #37by scroll » 08.04.2010, 01:17

abramson wrote:
scroll wrote:How do you turn fading on/off? I have solid line orbits and can find no reference to fading and orbit paths in 1.6.0. Thanks.
Hi, Scroll. Fading orbits are still not in the official release version of Celestia. You will have to wait until the next version.

You can try this feature by compiling Celestia yourself from the development sources. If you use Windows and you don't know how to compile Celestia, you can use one of my pre-compiled ones. Download from here the most recent one and unpack inside your Celestia folder (take the precaution of renaming the executable, so you do not overwrite the celestia.exe you already have there; any name will do).

Guillermo

Thanks Guillermo. It makes sense now.
You wouldn't happen to know how to sync a real-time orbit with that awesome ISS model that's available, woud you? (viewtopic.php?f=6&t=13360&start=120#p122203)
It's .ssc file is not accepting the usual edit...

abramson
Posts: 404
Joined: 22.07.2003
With us: 16 years 3 months
Location: Bariloche, Argentina

Re: Displaying orbits

Post #38by abramson » 08.04.2010, 02:16

scroll wrote:how to sync a real-time orbit with that awesome ISS model that's available, woud you? (viewtopic.php?f=6&t=13360&start=120#p122203)
It's .ssc file is not accepting the usual edit...
If you mean making the simulated movement of the model ISS reflect the actual position of the station, you can't do it just within Celestia. The orbit of the ISS is very variable and you can't have a fixed set of parameters represent it reliably. The programs that do it (Cartes du Ciel, for example) need to update the data for the orbit frequently to do it succesfully.

Of course, you could do it in principle: you need to donwload these data (in a form usually called "two-lines elements" (easily available) and convert them to a keplerian orbit for use in Celestia. That's perfectly possible, and I'm sure someone must have done it: a lookup of "tle to ssc celestia" came up with this old thread where there seems to be a Perl script to do it. You could even automatize it so every time you launch Celestia, you could have the right orbit already set up.

I do not understand the "It's .ssc file is not accepting the usual edit" part of your post. Every ssc file can be edited with a text editor.

Cheers,

Guillermo

scroll
Posts: 34
Joined: 30.12.2005
With us: 13 years 10 months

Re: Displaying orbits

Post #39by scroll » 08.04.2010, 12:16

abramson wrote:
scroll wrote:how to sync a real-time orbit with that awesome ISS model that's available, woud you? (viewtopic.php?f=6&t=13360&start=120#p122203)
It's .ssc file is not accepting the usual edit...
If you mean making the simulated movement of the model ISS reflect the actual position of the station, you can't do it just within Celestia. The orbit of the ISS is very variable and you can't have a fixed set of parameters represent it reliably. The programs that do it (Cartes du Ciel, for example) need to update the data for the orbit frequently to do it succesfully.

Of course, you could do it in principle: you need to donwload these data (in a form usually called "two-lines elements" (easily available) and convert them to a keplerian orbit for use in Celestia. That's perfectly possible, and I'm sure someone must have done it: a lookup of "tle to ssc celestia" came up with this old thread where there seems to be a Perl script to do it. You could even automatize it so every time you launch Celestia, you could have the right orbit already set up.

I do not understand the "It's .ssc file is not accepting the usual edit" part of your post. Every ssc file can be edited with a text editor.

Cheers,

Guillermo

Sorry Guillermo I didn't explain it well at all. I can update the ISS's position using the procedure from this post (viewtopic.php?f=2&t=12582&hilit=iss+position#p105240). In comparing with the ISS tracking site Heavens Above it appears to be an accurate procedure, although I appreciate updating fairly regularly is required as you mention. The ISS.ssc file that I modify (edit) is the one that comes with Celestia 1.6.0.

However if I try the same procedure with the relevant ISS.ssc file for the new ISS model avaiable from Linuxman then it doesn't work, Celestia doesn't even display the ISS let alone have it in the correct orbit. So I have to revert back to using the Celestia default ISS.

As you know your way around the Program I thought I'd take the opportunity to ask. A bit off topic for this thread, hopefully Linuxman will reply over on his thread.

scroll
Posts: 34
Joined: 30.12.2005
With us: 13 years 10 months

Re: Displaying orbits

Post #40by scroll » 10.04.2010, 02:54

deleted by poster


Return to “Ideas & News”

Who is online