stars.txt

[selden]

Have you considered using the revised edition of the Hipparcos catalog?
Unfortunately, a free version is not yet available.
http://www.springer.com/astronomy/book/ ... 020-6341-1

[chris]

I'm also trying to get the spectral type guesser on a more solid scientific footing. I'm considering using the B-V values in this paper, but whether these are appropriate, or if there are better tables linking B-V to spectral type, please let me know!

My copy of Lang's Astrophysical Data: Planets and Stars has two tables that might be helpful:
Intrinsic Colors Zero-Age Main Sequence
Intrinsic Colors in the MK System, Population I

The second table contains a lot more spectral types than the one in the paper that you linked to. Whether that makes it 'better', it's hard to say. I'll have to read the paper more closely; I note that it the table title mentions 'computed' values.

Both of the tables in Lang list just U-B and B-V values, while the Bessel paper V-R, R-I, and V-I as well. I can scan the tables for you if you want.

EDIT: I just Googled "Intrinsic Colors in the MK System' and found this:

http://articles.adsabs.harvard.edu/cgi- ... etype=.pdf

--Chris

[ajtribick]

Have you considered using the revised edition of the Hipparcos catalog?

Yes. I rejected it because

Unfortunately, a free version is not yet available.

which rather restricts my ability to get my hands on the data.

My copy of Lang's Astrophysical Data: Planets and Stars has two tables that might be helpful:
Intrinsic Colors Zero-Age Main Sequence
Intrinsic Colors in the MK System, Population I

The second table contains a lot more spectral types than the one in the paper that you linked to. Whether that makes it 'better', it's hard to say. I'll have to read the paper more closely; I note that it the table title mentions 'computed' values.

Both of the tables in Lang list just U-B and B-V values, while the Bessel paper V-R, R-I, and V-I as well. I can scan the tables for you if you want.

Well I only really need the B-V values, so the lack of V-R, R-I and V-I is not going to be a showstopper. Scanned versions of the tables would be very useful.

EDIT: I just Googled "Intrinsic Colors in the MK System' and found this:

http://articles.adsabs.harvard.edu/cgi- ... etype=.pdf

It's not entirely clear how "B1" and "B2" relate to "B" though, so I'm not entirely sure whether conversion from "B2-V1" to "B-V" is possible. The values in that paper also don't extend later than K5V or M0III, or earlier than B2.

[ajtribick]

Ok, here's the new generator script. Run with command line switch -? to see help screen. This one ONLY uses hip_main.dat (by default it will search for this file in the current directory), since using data from other files in I/239, e.g. Tycho or the companions dataset seems a bit controversial (judging by getting jumped on for mentioning that I was planning implementing the companions dataset...), besides this way it makes it clearer just what's in the file. I might release an extended database as an add-on at some later date.

The script generates the stars.dat file from the Hipparcos catalog. The script does not generate a stars.txt file. The spectral type guessing code now uses data from Lang (2001). Thanks to Chris Laurel for sending me scans of the relevant tables.

(Note that the zip file contains the Perl script only, it does NOT contain a new version of stars.dat since this can be generated with the script. You will need to download hip_main.dat from http://cdsarc.u-strasbg.fr/viz-bin/Cat?I/239)

Any suggestions, particularly with regard to the default parameters and the rejection limits (TestDubious and CheckStar subroutines)?

(removed because of a couple of nasty bugs)

[ajtribick]

Following discussion on the developer's list, I have altered the position error cutoffs. The displayed tallies now include the number of bright stars (Vmag<=6) dropped.

(removed because of a couple of nasty bugs)

[ajtribick]

New version which uses data from new reduction of HIP catalogue (http://cdsarc.u-strasbg.fr/viz-bin/Cat?I/311). Also fixed a couple of nasty bugs which crept in.

New data is only used if it doesn't replace a positive parallax with a negative one. (This may change depending on discussion on developers' list or here)

[ajtribick]

A new version. This one also generates a stars.txt file.

This uses the new reduction of the Hipparcos data as the primary input. The old version is used solely for the Vmag and SpType fields (plus BTmag and VTmag, which may be used if the Vmag field is missing).

The Sun is not included anymore, since it can't be output into stars.txt without displacing it from the origin of the coordinate system (stars.txt uses Apparent Magnitude). This does not affect the running of Celestia since the Sun is defined in nearstars.stc anyway.