Celestia 2320AD

[Anders]

I'm starting to get close to finishing a Celestia add-on with the 2320 universe. You can see some samples of what I have done at
http://www.flickr.com/photos/arenamontanus/sets/72157604565519320/
(best viewed at full size or as a slideshow - the big pictures suffers a bit from being compressed)

My program now generates a working 3D starmap based on the revised Near Star Map, plots a "nebula" showing the possible stutterwarp routes, generates random orbits for multiple star systems and adds planets to a few systems.

Now the fun starts! Adding new planets is relatively easy, and I have been digging through the Colonial Atlas and Dan Schirren's system charts to get the right systems (some tweaking/caution is needed; the chart for Tirane was not compatible with the real orbit of the stars). I expect that in a few days I will have added most of the canon systems. Another fun thing is to make textures; right now I have just taken a few versions of solar system textures and thrown them onto the planets.

Some things remaining for me to do: add better control for multiple star systems with canon orbits, add a bit of random inclination, make sure everything rotates as it should, add "special features" like rings and eccentric orbits.

I'd love to make this a group effort; once I get the add-on out I would welcome additions in the form of corrected orbits, new systems, better textures, locations of colonies, space stations and beanstalks and whatnot. The nice thing about Celestia is that it is easy to stack add-ons on top of each other, so if somebody got a 3ds model of a Kennedy cruiser it is quite easy to put it in orbit around Aurore. I'm also interested in hearing what people would use something like this for.

Thoughts?

 

[Colin]

If you want planet textures, I can check with Hunter regarding making the textures I used for the maps in 2320 available. He owns them now, so it has to go through him.

 

[Anders]

If you want planet textures, I can check with Hunter regarding making the textures I used for the maps in 2320 available. He owns them now, so it has to go through him.

That would be excellent! It is so annoying having my Earth texture (for a generic garden) show up on planets I know are far more interesting and exotic.

I just spent the evening inputting system data into a spreadsheet, things are getting ready for a beta soon.

One issue I'd like to discuss is how to distinguish canon from non-canon elements. I have been planning to separate canon systems from non-canon ones into separate files, but it might not be 100% possible. Most colony systems were relatively well described in the Colonial Atlas, but usually left out minor details like orbital radius or diameter for some worlds. I have used Dan Schirren's great system charts for orbital distances and made up plausible values elsewhere. So already in established places like Neubayern there are small non-canon elements. The real trouble starts with naming; there are many worlds with no names at all, like the Beta Hydri system (beside Daikoku).
Would the proper approach be to have a canon version naming them "Star <roman numeral>" and a separate file with the non-canon version with a plausible name? (in the beta hydri case, I would expect the other planets to be named after the Seven Gods of Good Luck) Or some discreet marker that the name is made up? Or that we just all admit that the Celestia version will be a slightly off-canon synthesis?

 

[Anders]

Yesternight I started thinking about the right way of filling in missing numbers, in particular planet diameters. Given that we know that a planet is a gas giant, what is its diameter? One way of getting the numbers is just to simulate rolling the dice in the 2300AD planet generation system. I made a program to do it and generated 10,000 planets in each of the zones. I have plotted the results here:
http://www.flickr.com/photos/arenamontanus/2420915580/sizes/o/

Disregarding whether we believe the system is physically plausible (we now know there should be more gas giants in the inner zone), it seems to be a rational way of filling in numbers - I just draw sizes randomly from these categories. I might add a cut-off for absurdly large gas giants, since brown dwarf theory shows that non-fusioning bodies will not get much larger than ~150,000 km.

 

[aramis]

Aside from the color choices, nice graphs.

 

[Anders]

Yeah, I'm kicking myself for using warm colours for the cold outer system objects and cool colours for the hothouses. Never program at three in the morning!

 

[Beech]

Anders, excellent work & it seems to have only taken a couple of weeks!
A man of far too much talent, do you not have a day job?
A suggestion - When you have assembled a version you are happy with and want to make public why not upload it onto the Celestia motherlode (I think that's the name of it), a lot of people surf that site & it would work as a good ad for & conduit into the 2300/2320 universe.
By the way that's how I first heard of OA - are you by any chance the same Anders responsible for so much of that too?

 

[Anders]

Thanks! I'm happy my boss (likely) does not surf these forums...

I'll do an upload to the motherlode. That would be the logical place. I'm rather proud of this package, and it is also a fun way of mapping everything for game planning. Now I know where Gaaumouhjeung is! And Beta Canum 8-A seems to be an interesting place despite the name.

Yes, I'm the OA Anders too - the one who ended up a search engine with a primality-obsessed alife civilization as descendants

 

[Anders]

TA-DA!!! It is here! Or rather, at http://www.aleph.se/2320AD/2320AD.zip
until it is officially placed at the Celestia Motherlode.


2300AD/2320AD Universe 1.0

DESCRIPTION

This addon represents the universe of the 2300AD and 2320AD
roleplaying games. The setting is the stars within ~60 ly from the sun
(originally based on the Gliese catalogue).

In the game the FTL drive (the main fudge in terms of physics) has a
distance limit of 7.7 ly before it must be discharged in a sizeable
gravity well. This makes only some routes possible, leading to "arms"
of expansion and various political and military complications. The
addon displays the arms using a green "nebula" depicting possible
paths.

There is also a yellow network corresponding to the "61 Cygni
cluster", an arm that may become available due to recent technical
developments; this is as yet (2320) unexplored.

When the original game was written in 1989 the second edition Gliese
catalogue was used. Since then better parallax measurements have been
done, new stars have been discovered and some objects have
disappeared. Since the map is sensitive to such changes the game must
retain the old coordinates rather than use modern ones. It is not just
an alternate timeline (since the Twilight War did not occur in the
1990s) but an alternate cosmography. As far as possible this addon
moves stars to the game coordinates rather than their current best
estimated (see below for some of the issues with this). The locations
are based on Andy Brick's Near Star List II.

My generation software generates orbits for multiple star
systems. This is done randomly, with orbit periods distributed
lognormally with mean of 14 years and masses estimated from the
mass-luminosity relation of main sequence stars.

Planets and systems are primarily based on descriptions in 2300AD
Colonial Atlas, the 2320AD book and Dan Schirren's non-canon list of
planets (available from http://stalexone.tripod.com/gg2/map-info.htm )
Wherever possible I have used 2320 data, then 2300 data, then Dan's
list and finally made up information that seems plausible. This has
mainly been done for planet diameters, which were generated based on
the planet type and the 2300AD random planet generation system (i.e. I
generated a large set of planets and then selected one randomly
depending on type).

Note that this addon is not suitable for planning any interstellar or
interplanetary excursions. For accurate navigational data, please
consult the current 2320 IAU Object Catalogue, the ESA Carte
Astronomique Database (Annex IV and upwards) or the full AECA
Atlas. The author is in no way responsible for navigational hazards,
loss of PC lives or equipment damages due to this addon. The addon
does not make any statements about the colonizability of different
worlds.


INSTALLATION

Place the 2320AD folder in the Celestia extras folder.

It includes a models subfolder (the pathways cmod model, the 61 cygni
cluster and an empty model), a textures folder and dsc, ssc and stc
files.

Also included are a Matlab script to generate these files from a star
database file and a planet list.



INCLUDED SYSTEMS

Augereau, Aurore, Austin's World, Avalon, Beowulf, Berthier,
Bessieres, Beta Canum Venaticorum, Botany Bay, Chengdu (buggy), Cold
Mountain, Daikoku, Doris, Dukou, Dunkelheim, Ellis, Eriksson,
Freiland, Haifeng, Heidelsheimat, Henry's Star, Hermes, Hochbaden,
Joi, Kie-Yuma, Kimanjano, King, Kingsland, Kormoran, Kwantung,
Lightfall, Montana, Neubayern, Nous Voila, Nyotekundu, Paulo, Proxima
Centauri, Ross 128, Ross 614, Sans Souci, Stark, Syuhulahm, Tirane,
Vogelheim


KNOWN ISSUES

This addon tries to override the existing star database, but is not
entirely successful. Epsilon Indi appears in two places, and Epsilon
Eridani suffers from having the real exoplanets show up among the game
ones.

Multiple star systems are all in same plane. The subsystems ought to
have different inclinations. Systems with canon distances and periods
are in most cases not looking like they should yet.

Orbits are not necessarily collision free.

The solar system browser appears to get confused by planets in
multiple star systems and only show some of the planets.

Periods of planets and moons are not correct, since they are all based
on the assumption of 1 solar mass stars or 1 earth mass planets.

There should be many more moons in the systems, but I have not
generated them. Moons in the canon have been added.

Not all outposts have been added yet. Ditto for Kafer space worlds.

Gas giants generated using the 2300AD system tend to be far too
small. Occasionally they snowball the absurdly large sizes; due to
degeneracy pressure they are not likely to become much larger than
~150,000 km diameter.

Currently garden worlds have just the same earth placeholder texture
rather than canon maps. Ideally the textures should be improved
radically.

Since Octave to my knowledge does not have textread (and may have
trouble with my use of cell arrays) the Matlab script would need a
rewrite to function under that program. In which case a porting to,
say, Python would make much more sense anyway.


CREDITS

This package makes use of:

Grant's empty object for Celestia,
http://www.lns.cornell.edu/~seb/celestia/hutchison/empty.html

The kimanjano map is based on an original by Laurent Esmiol.

Thanks to Beech on the Citizens of the Imperium forum for suggesting
the idea in the first place.

 

[Beech]

This is absolutely bloody brilliant Anders, well done.
I am having immense fun playing with this, its just great.
I (& I am sure others too) will happily contribute by modelling up a few ships/objects for it too if you would like.
(& thanks for the mention too!)

 

[kmsoice]

We are not worthy! Thankyou!

 

[Anders]

Thanks everyone! I'm eager to hear your comments, suggestions and additions.

Over the next few days I plan to add remaining outposts and a few other well-described systems, as well as start to deal with the issues on my list in a leisurely manner.

 

[Rick Baumhauer]

Anders - thanks for putting the time into this.

I'm having a problem with the 2320ad.ssc file crashing Celestia on my Mac Pro. I'm running OS X 10.5.2 and Celestia 1.5 with an ATI 2600HD 256mb graphics card. Is anybody else having a similar problem? Celestia starts up fine if I remove the 2320AD folder.

Rick

 

[Beech]

Rick,
I have a MacBook Pro with OS X 10.5.2 it worked fine with Celestia 1.4.1
but I just upgraded to 1.5 & it now crashes just as you say.
I will have a look on the Celestia forums & try and figure out what may be wrong & in the meantime I will see if I can source an older version.
I love my Mac but sometimes...

 

[Anders]

It could be a texture size issue. Try temporarily removing the textures in my addon, and see if that stabilizes things.

 

[Colin]

Run planetary orbits and names past me, and I'll talk to Hunter. We can canon-ize them.

 

[Anders]

Run planetary orbits and names past me, and I'll talk to Hunter. We can canon-ize them.

That would be great! If you have the addon, it is already in the file planets.csv in the matlab directory, easy to open in Excel or Openoffice.

One system I know there is something up with is Avalon. Clearly the names are supposed to be Arthurian, but the names in the 2320AD pdf seemed to be garbled. Who is Coalman?!

In the case of Kie-Yuma, there seems to be a conflict between CA and 2320 names; I also invented names for the two moons (using a random American last name generator).

Other systems where I know names are non-canon: For Beta Hydri I made up names following the Japanese pantheon of good luck/prosperity gods (given the name Daikoku for the planet; the moons were named for his attributes but should of course be in Japanese).

DM-56 328 got two stone names (but Gerollblock, should that really be a gas giant?)

Epsilon Eridani names were clearly Sichuan regional capitals (the things you learn from roleplaying! I actually ran a campaign where I needed to know this!) so I added a few to unnamed planets.

On Zeta Herculis B I continued the theme of Australian nature preserves.

In the 82 Eridani system I invented names for the unnamed two rocks, and the two unnamed moons (as soon as the Ebers heard about them from the humans they ought to have given them names).

At DM+17 2611 (Lightfall) I used other random personal names, referring to the casualties of the crashed Trilon expedition.


Freiland ought to have some names for neighboring planets; given the wonderful "Kleiner Bruder" for the moon I doubt the Freihafen people lack creativity.

And you probably know my opinions on the need for names for the BC numbered planets

 

[Rick Baumhauer]

Beech - thanks for the pointer on 1.4.1. I was able to find it on SourceForge, and it does work with Anders' data. Haven't experimented with removing the textures on 1.5 yet.

Rick