Page 1 of 3
Nebulae in 1.2.6
Posted: 27.01.2003, 01:50
by chris
I've made significant progress in supporting nebulae . . . They'll be defined in a manner similar to galaxies, except that they're rendered with .3ds meshes instead of fuzzy blobs . . . Here is an image I captured of Rassilon's NGC1999 model:
And links to a couple more:
http://ennui.shatters.net/gallery/view_photo.php?set_albumName=celestia&id=ngc1999_3
http://ennui.shatters.net/gallery/view_photo.php?set_albumName=celestia&id=ngc1999_1
These use the new deep sky object code that I've been working on, so there's no longer the problem of the nebula disappearing when you're more than one light year away. The mesh can be any .3ds file, so you could create a planetary nebula by applying a texture to a slightly distored sphere. The bulk of the work in adding nebula to Celestia is going to be actually creating the meshes and textures . . .
Many thanks to Rassilon for creating the NGC 1999 model; I don't think I'd have gotten around to writing the deep sky object code so soon if he hadn't provided the content first.
--Chris
Posted: 27.01.2003, 02:50
by Calculus
Great work. I'm eager to try it.
Posted: 27.01.2003, 02:53
by billybob884
this is gonna be nice, can we get a few shots from an angle?
Posted: 27.01.2003, 02:54
by billybob884
Posted: 27.01.2003, 02:57
by selden
Unfortunately, the Celestia Album on Shatters.Net isn't working for me: I get the backdrop with titles but no images.
Posted: 27.01.2003, 03:17
by billybob884
when is 1.2.6 coming out anyway?
Posted: 27.01.2003, 03:57
by Rassilon
Posted: 27.01.2003, 04:12
by selden
Thanks, Ras'!
Your numeric links work fine.
The named links on the Celestia Gallery page don't
Apparently Shatters' nameserver's down. Response time through Shatters' gateway is quick, so it doesn't seem to be due to the usual filesharing server traffic.
Posted: 27.01.2003, 09:53
by boo
Talking about deep sky objects..
Is anybody interested in creating a little galaxy update through the data available from the
Sloan Digital Sky Survey?
Here you can get a tool that shows the impressive structure of about 131,000 galaxies. Imagine Celestia would include them
Nebulae
Posted: 27.01.2003, 11:00
by ogg
This is excellent news!
I guess an initial 'quick and dirty' way to get a lot of nebulae into Celestia would be to have a 'flat plane' 3ds to paste photos onto, or perhaps a bubble or shape like Ras's new nebula? how would the models be oriented? if the default were were sol-facing and oriented according to geocentric RA-DEC (ie with sides parallel to the celestial grid) then it might be a relatively easy job to assign textures derived from FITS files...
3D models and nicer textures can then replace these as they become available.
not that I have any clue how to actually go about this of course...
As some hard data to start with... HNSky freeware planetarium software includes an optional download of 255 FITS images extracted from Realsky, with positioning data:
http://www.hnsky.org/software.htm
they're not especially pretty but they're accurate and potentially more useful (at least initially) than hubble beauty shots. Can at least serve as a guide for development?
screenshot:
http://www.hnsky.org/hns_m43.gif
problem is, I don't think they include distance measurements? I know of some distance tables on webpages...
http://www.anzwers.org/free/universe/nebclust.html
http://anzwers.org/free/universe/5000lys.html
http://www.rca-omsi.org/howfar.htm
... but I assume there must be a catalog or something to work off?
Congrats Chris and Rassilon
Posted: 10.02.2003, 01:52
by ClimbingRog
I saw Rassilons additions of V380 Orion and had to download it and the Eta Carinae system as well (Awesome Paul! The planets are incredible. Got Oh's & Ah's from my Kids the first time I displayed them)! Thats what I've been waiting for. I've loved looking to the sky's for those beautiful nebulae since I was a kid. Chris I'm glad you will have the 1 Light year problem sorted out for the 1.2.6 version. That is Fantastic! I'm just trying to learn how to do 3D imaging myself using the Anim8or program, but I'm a long way from being able to do those nebulae yet.
On another note Chris do you have any plans of creating a texture of Venus's surface without the cloud layer. I haven't seen it yet. Any way I found a really good flat surface texture image in an old Astronomy book I've got, and have been attempting to convert it into a usable texture for my Celestia program.
Posted: 10.02.2003, 13:32
by selden
Rog,
Grant Hutchison has generated a Venus surface texture, revised cloudmap and an updated ssc entry. (Celestia's current Venus texture has an orientation problem.) They're available as part of his package of planetary and satellite orientation and texture corrections. See
http://www.lns.cornell.edu/~seb/celestia/hutchison/
Deep Space Explorer's version of nebula
Posted: 15.02.2003, 17:52
by fsgregs
HI folks. I own another software program put out by the same folks who publish Starry Night. Its called Deep Space Explorer. Its a galaxy program that draws over 70,000 spiral, elliptical and irregular galaxies in proper position. In general, the program appears to use flat plane textures that it wraps around a severely oblate mesh, to give a strongly flattened spiral shape. Here is one example:
DSE appears to draw elliptical and irregular galaxies differently. It creates a distinct 3-D haze/galactic nebula that has good 3D texture from all sides and angles, and populates it with stars. Zooming in or out does not change its detail. Here is an image of it.
I don't know how it does it and I don't know if this helps at all in developing nebula in Celestia, but .....
Regards,
Frank
Posted: 16.02.2003, 02:41
by Rassilon
That galaxy model would look sweet with a nice cluster of stars in the center...Not to mention a bit of volumatic fog...
Deep Space Explorer's version of nebula
Posted: 16.02.2003, 06:14
by jgrillo2002
fsgregs wrote:HI folks. I own another software program put out by the same folks who publish Starry Night. Its called Deep Space Explorer. Its a galaxy program that draws over 70,000 spiral, elliptical and irregular galaxies in proper position. In general, the program appears to use flat plane textures that it wraps around a severely oblate mesh, to give a strongly flattened spiral shape. Here is one example:
DSE appears to draw elliptical and irregular galaxies differently. It creates a distinct 3-D haze/galactic nebula that has good 3D texture from all sides and angles, and populates it with stars. Zooming in or out does not change its detail. Here is an image of it.
I don't know how it does it and I don't know if this helps at all in developing nebula in Celestia, but .....
Regards,
Frank
Hey, I have that. I got it as a christmas present
Posted: 21.02.2003, 15:34
by Mad Boris
This all looks too good to be true, so I'd better ask some skeptical questions...
1) Are these 3DS meshes volumetric, or just pictures on hemishperes like Rassillion's (which I shamefully have not downloaded yet
) ? Nothing wrong with that of course, just would like to know.
2) How big will they make Celestia ? 10 mb is about 30 mins download for us 56k users, and I know Rassillion's nebulae were about 40 mb.
3) Just to be rather cynical if I may, is this going to be one of those very nice features (like specular and bump mapping and oversampling), which requires a good graphics card to show them ? (I know this is probably unlikey but as I said, it looks to good to be true, so I'd better check !)
And billybob884..... of all the picture-heavy posts you've done, that was the prettiest !
On a less-nebulae related note, now that Celstia releases are arriving rather quickly again, is there any chance of just being able to download an update rather than the whole file ? Or would this not be possible for programming reasons ?
Posted: 21.02.2003, 18:21
by chris
Mad Boris wrote:This all looks too good to be true, so I'd better ask some skeptical questions...
1) Are these 3DS meshes volumetric, or just pictures on hemishperes like Rassillion's (which I shamefully have not downloaded yet
) ? Nothing wrong with that of course, just would like to know.
This is Rassilon's nebula mesh. But in Celestia 1.2.6, it works much better. 1.2.5 had no support for placing 3DS meshes in interstellar space; Rassilon had to make huge a planet billions of kilometers diameter! Celestia wasn't designed to handle such huge objects in solar systems, and the nebula would suddenly disappear when you got more than a light year away from it. In Celestia 1.2.6, you can define nebulae, galaxies, and open clusters in .dsc files, and they just work.
2) How big will they make Celestia ? 10 mb is about 30 mins download for us 56k users, and I know Rassillion's nebulae were about 40 mb.
There won't be any nebulae included in the base Celestia release. Most of the size of Rassilon's add-on was due to planet textures, and not the nebula.
3) Just to be rather cynical if I may, is this going to be one of those very nice features (like specular and bump mapping and oversampling), which requires a good graphics card to show them ? (I know this is probably unlikey but as I said, it looks to good to be true, so I'd better check !)
Nothing special is required . . . The nebula isn't rendered volumetrically or anything, so even a TNT or Rage card will do.
On a less-nebulae related note, now that Celstia releases are arriving rather quickly again, is there any chance of just being able to download an update rather than the whole file ? Or would this not be possible for programming reasons ?
The only thing preventing me from making a small update is my own laziness . . .
--Chris
Posted: 21.02.2003, 23:14
by Rassilon
I will be releasing nebulas as stand alone and with planets from now on as a choice...This will allow those with smaller connections to grab them in less than a whole day
I will also be updating the NGC1999 addon to reflect the changes chris has made to celestia...
Posted: 21.02.2003, 23:44
by selden
Ras',
Would it be possible for you to publish a .3DS model of a transparent square surface (x=y=1, z=.0001) that can accept any texture image? Your model for NGC 1999 requires that the .PNG file have a specific name.
Presumably such a model could be tested by using it for a planet and then it could be used as a nebula billboard as soon as that becomes possible.
Thanks!
Posted: 22.02.2003, 00:01
by Rassilon
Ill see what I can do on that selden...That might not be possible considering whats involved with transparencies currently...Chris may have to implement a TransparencyMap script for meshes in order for this to work universal...