Celestia Forums

At some point in the future, Andrew Tribick (ajtribick)'s updated stars.dat will be added to Celestia. This update happens to fix nearly all instances where the star was missing due to a missing or low-quality parallax. This makes the resulting revised.stc much smaller. But I think it would be a shame if we were to not use this as an opportunity to fix or add in many of the features missing in Celestia. For example:

Some of the brighest stars like Castor (α Gem) and Algol (β Per) are rendered as single stars. For the brightest stars (above some limiting magnitude so we don't go overboard) we could have extra binary star definitions.

A few stars have been resolved through interferometry, so we know their oblateness and orientation of their rotational axis. Examples include Achernar (α Eri) and Regulus (α Leo). A while back I wrote bits of STC code in my revised.stc in that fix this.

But those are just two of my suggestions. I'd love to hear everybody's suggestions on this.

Luke

Yes, this is a great idea. I don't think binary stars should be in revised.stc though, instead the entire binary star catalog should be updated. I suggest you go ahead and make these pull requests (but not yet, after the CelestiaContent issue is resolved).

SevenSpheres wrote:I don't think binary stars should be in revised.stc though, instead the entire binary star catalog should be updated.

I think the binary/multiple star catalog will probably look like the existing version in that it will be generated from a single source. As such, not every star system will be included, because no source even comes close to being comprehensive. If we want to add them into Celestia, some star systems will have to be done "by hand", and I don't want to mix them into the script-generated STC file.

Also, automating the addition of binary stars into Celestia is insanely difficult.

LukeCEL wrote:I think the binary/multiple star catalog will probably look like the existing version in that it will be generated from a single source.

I was thinking it would be generated from multiple sources, in the same way as your existing Binary Star Catalog addon.

What do you mean by "multiple sources"? Indeed it's possible to merge multiple sources into one and create a sort of "super-catalog". But my Binary Star Catalog has literally hundreds of sources, a few for each system. I did each system "by hand", but that's totally different from visualbins.stc and spectbins.stc.

Oh, I thought it was generated from the "catalog.txt" file included with the addon. Another example would be the exoplanet catalog (which I expect will eventually be added to the code base) - Sirius_Alpha has a "super-catalog" created from multiple sources, from which the Celestia files are generated.

SevenSpheres wrote:Oh, I thought it was generated from the "catalog.txt" file included with the addon.

I wish. I created the catalog.txt files in the anticipation that eventually I would figure out some way to automate it. But even then, the catalog.txt was made by hand.

Also, I think Sirius_Alpha's super-catalog is the same, I think he copies values directly from papers instead of looking at databases. But you'll have to ask him.

LukeCEL wrote:Also, I think Sirius_Alpha's super-catalog is the same, I think he copies values directly from papers instead of looking at databases.

Exactly. Point is that if there isn't a uniform catalog of <thing>, we can make one, and it can be added to Celestia's code base.

SevenSpheres wrote:Point is that if there isn't a uniform catalog of <thing>, we can make one,

I don't think that's ideal because any time you do that, you're guaranteed to add in typos and mistakes.

Added after 10 minutes 5 seconds:
Maybe confusion might arise because I have two ways of making uniform catalogs of "things". One is to cobble together existing catalogs. You can find them on VizieR and it's relatively straightforward because it's just a case of aligning the columns. I did this for my Andromeda globulars add-on.

But what I did for the Binary Star Catalog was create a whole new table from scratch, including all the columns, and fill in each of the entries by hand. I don't feel like doing this for the Celestia data files because maintaining a database would be out of the scope of Celestia. Also, it would be a massive pain in the rear end. It took me several months to fill in the "catalog.txt" for the Binary Star Catalog.

Anyway, I feel like we're quickly getting off topic, since this is what revised.stc is supposed to be about, not visualbins.stc or spectbins.stc.

Added after 29 minutes 32 seconds:
pedro_jg reminds us that we have to remove the two spurious HIP stars, HIP 114110 and HIP 114176.

LukeCEL wrote:Also, I think Sirius_Alpha's super-catalog is the same, I think he copies values directly from papers instead of looking at databases. But you'll have to ask him.

You are precisely correct. There's trade-offs. Sometimes databases are not assembled very well, or are missing information that are contained in the literature. Nothing on the NASA Exoplanet Archive tells you about spin-orbit (mis-)alignment, inclination of the stellar spin axis, binary orbits, etc, so I have my own generalized data architecture from which the files are compiled. On the downside, as LukeCEL mentioned, there will be typos and mistakes (though I have several various other programmes that compile information from the ExoFOP, and from downloaded .tex files to extract typo-free information).

In my opinion, the general philosophy for the default Celestia base would be database-derived files. However I'm not aware of any uniform source of information that covers both nearby dim stars in the solar neighborhood that evaded Hipparcos, and nearby/bright stars that are included in Gaia. So we cannot rely on a future Gaia DR to give us, for example, Alpha Centauri's orbit to include in a hypothetical "BinaryStars.stc" file. Perhaps a merger of the HIP binary catalogue with the Gaia binary catalogue could be feasible in the near-future. I would like to see nearstars.stc should be removed as soon as possible (as well as visualbins.stc and spectbins.stc removed as well).

Sirius_Alpha wrote:I would like to see nearstars.stc should be removed as soon as possible

Why? It includes many nearby stars and brown dwarfs that wouldn't otherwise be included, since they aren't in the HIP/TYC catalogs, multiple stars, or exoplanet hosts.

Because it isn't assembled from a uniform data source, and hopefully many of its contents would be in a newly assembled Gaia+Hipparcos stars.dat file at some point in the future. Clearly we are not currently at the point where dumping nearstars.stc is feasible (lest we lose α Cen as a binary system).

I think that a lot of the other content (brown dwarfs that would be outside of a Gaia+Hipparcos-derived file) should be made available as parts of separate add-ons.

But that's just my opinion -- namely that the default Celestia download should have only uniform data in it. It's not like that's anything more than a personal preference, and it's probably tainted by difficulty I have had trying to create an exoplanet catalogue that can deal with the scattered, haphazard contents of the various "extra" .stc files in the data folder.

Well said sirius_alpha !

Your methodology seems serious...