Barnard's Star Add-On Problem, 1.6 Related?

[ares2101]

I maintain an online sci-fi setting and so far I've published 3 of its star systems as Celestia Addons. Since 1.6 came out I've successfully tested and tweaked my Alpha Centauri A and B addons (not uploaded yet), but I'm having a problem with my Barnard's Star one and I think it somehow is related to the new version of Celestia, but I'm stumped on the exact cause. I'll explain the problem here and post the link so you can download it and check it out if you need to.

I pasted the addon into the new Celestia and went to the system where I noticed something odd. If I zoom in just right, I see that the Dyson Swarm model has its texture. If I go to any of the Bishop Ring habitats, I see they and their independent central light sources also have their textures. What I can't figure out is why none of the over 200 asteroid habitats have theirs. The models show up, I can see the shapes, but none of them have any of their textures. All of the textures for the system are in the appropriate folder, "Celestia\extras\Garden Universe\Barnard's Star\Textures\Medres" and as I said, it's only the asteroid habitat textures that don't show up, the Dyson Swarm and Bishop Rings do show up right, as do the ones for the one planet. All of the models, the working and flawed are the same format, ".3ds" files because I use Anim8or and for whatever reason it won't open ".cmod" files.

What's going on here? This addon worked perfectly well for me in Celestia 1.5.1, I'm sure of it, I tested it extensively as it was so complex (it took tweaking the Asteroid Maker utility to generate so many habitats). It seems awfully selective in its failure. Also note that I haven't yet changed a thing, I've tweaked things to try and fix it but I've kept an original copy to come back to when stuff doesn't work to make sure cumulative mistakes don't happen. I don't get it.

http://johnsonm.com/garden_universe/Celestia_Addons.htm

Thanks in advance.

[selden]

Ares,

There's something very strange with the file Barnard Asteroid Habitats.ssc: although it's 149KB, it contains only the definitions for "Luminaire 1" and "Bishop Ring 1". I don't see any other objects in it. I looked at it with both emacs and Notepad. Celestia may be having the same problem with it. My guess is that the original file contains some Ctrl-Z character codes. They indicate an end-of-file and might cause strange things like this to happen.

I'd suggest putting one of the failing habitat definitions in an ssc by itself to see if that one works.

Anim8or will be able to export CMOD model files if you install the CMOD export plugin which is available at
http://www.lepp.cornell.edu/~seb/celest ... g.html#4.0
Be sure to read its documentation on that web page.

Unfortunately, Anim8or does not yet provide any way to write import plugins.

[ares2101]

Ares,

There's something very strange with the file Barnard Asteroid Habitats.ssc: although it's 149KB, it contains only the definitions for "Luminaire 1" and "Bishop Ring 1". I don't see any other objects in it. I looked at it with both emacs and Notepad. Celestia may be having the same problem with it. My guess is that the original file contains some Ctrl-Z character codes. They indicate an end-of-file and might cause strange things like this to happen.

Um, I think your first point may be as obvious as not having scrolled down. I generated the majority of the Asteroid Habitats document by tweaking the Asteroid Maker Utility available at The Motherlode. My alteration, like the original, creates an output which can be pasted into a .ssc file, the only problem is a large amount of blank lines. Scrolling down about 1 to 1.5 times the length of the first two objects descriptions gets you to "Habitat 2", one of the asteroid habitats. Incidentally, the Bishop Rings described in that document I put there manually, replacing the randomly generated Trojan asteroids.

I'd suggest putting one of the failing habitat definitions in an ssc by itself to see if that one works.

I tried this with the first failed object, Habitat 2 and got nothing, but perhaps this is because of a conflict with the original in the main document, I'll try a stand alone in the morning, but I seriously doubt this is the issue. All of the Bishop Rings in Asteroid Habitats.ssc work fine and they are scattered about the document. My experience with Celestia tells me that if even 1 placeholder character were out of place, everything at or below that error would not show up properly (actually at all). My past experience tells me that the .ssc file can't be the issue, it's inconsistent with errors I've had before.

Anim8or will be able to export CMOD model files if you install the CMOD export plugin which is available at
http://www.lepp.cornell.edu/~seb/celest ... g.html#4.0
Be sure to read its documentation on that web page.

Unfortunately, Anim8or does not yet provide any way to write import plugins.

I'm pretty tired, so I'll try this in the morning. Again though, I find myself skeptical this will be the issue. The working Bishop Rings are .3ds files, same as the errant asteroid habs. This is precisely what puzzles me so, why roughly 6 asteroid models ALL won't show their textures, just the Bishop Ring and Luminaire models.

[selden]

Ares,

Sorry: you're right. My bad.

However, inspection immediately reveals a problem.
Note that an empty Texture is specified.
An SSC Texture declaration overrides all diffuse textures specified within the associated model. Since the Texture explicitly specifies no image file, no surface texture is drawn.
You need to delete all Textures which specify no image files.

Code: Select all

##############################################
#VALUE!
# Orbit family: Main Belt. Inc: 01.774777
"Habitat 4" "Barnard's Star" {
    Class                "asteroid"
    Texture              ""     
    Color                [ 0.300 0.300 0.300 ]
    Albedo               0.113
    Mesh                 "bubbleasteroid.3ds"
    Radius               50.62

    EllipticalOrbit {
        Period           000.219670347424
        SemiMajorAxis    0.203127596611
        Eccentricity     0.086422159457
        Inclination      001.7748
        AscendingNode    242.2500
        MeanLongitude    147.4089
        LongOfPericenter 265.1374

    }
    Obliquity            127.1713
    EquatorAscendingNode 239.7979
    RotationOffset       319.9615
    RotationPeriod       0091.930503413883

}


With regard to CMODs, I wasn't trying to say that using one would fix the problem, rather that Anim8or can be persuaded to export that format if you want to use it.

[ares2101]

Ares,

Sorry: you're right. My bad.

However, inspection immediately reveals a problem.
Note that an empty Texture is specified.
An SSC Texture declaration overrides all diffuse textures specified within the associated model. Since the Texture explicitly specifies no image file, no surface texture is drawn.
You need to delete all Textures which specify no image files.

Code: Select all

##############################################
#VALUE!
# Orbit family: Main Belt. Inc: 01.774777
"Habitat 4" "Barnard's Star" {
    Class                "asteroid"
    Texture              ""     
    Color                [ 0.300 0.300 0.300 ]
    Albedo               0.113
    Mesh                 "bubbleasteroid.3ds"
    Radius               50.62
[Snip]

Huh, funny, it worked just fine in the earlier version. The blank texture line is an artifact from the data output, as I said, the file is largely the result of an altered Asteroid Maker worksheet, I altered the texture section give have just one blank one because, as you said, a specified texture overrides any internal ones. Celestia 1.5.1 didn't seem to regard a blank texture declaration as "Don't Draw Any Textures", the new one seems to. In any case, a mass delete of texture line combined with a program I found to delete blank lines (which has seriously reduced the sizes of my various asteroid belt .ssc files) has solved the problem in full.

Thanks a ton, Barnard's Star was my only addon giving me trouble in the new version.