Binary star system screen shots

Post #41by **granthutchison** » 23.10.2004, 21:48

Evil Dr Ganymede wrote:Why do you need to define the RA and Dec for the companion? Isn't that a bit redundant, if Celestia already knows it should be orbiting the A star (or the barycentre, if applicable)? i.e. why not just define the RA/Dec for the barycentre of the system and leave it out of the definition of the stars?

You'll need to ask Chris that if you want a detailed reply. But there are trade-offs between the most parsimonious grammar for the star definitions and the maximum speed of rendering, I understand. Having each star explicitly assigned to a particular location (without Celestia having to chase down a list of connections) makes for more rapid decisions about whether a given star needs to be rendered or not, I believe. Repeated many times over for each frame refresh, such small rendering efficiencies can become important.

Grant

Post #42by **selden** » 23.10.2004, 22:00

Oh Evil One,

Yes, they're logically redundant, and they aren't needed in the most recent build from CVS. However, I think the test version of Celestia that Grant's using is a slightly older build which didn't include that optimization.

For example, the current CVS build accepts the following

Code: Select all

Barycenter "AlfCen"
{
   RA 219.916998
   Dec -60.83748372
   Distance 4.395 
}
 71683 # ALF Cen A
{
   OrbitBarycenter "AlfCen"
   SpectralType "G2V"
   AppMag 0.01

   EllipticalOrbit {
      Period          79.914
      SemiMajorAxis   10.8928   # mass ratio 1.09:0.92
      Eccentricity    0.5179
      Inclination   82.98
      AscendingNode   67.71
      ArgOfPericenter 3.77
      MeanAnomaly     200.12
   }
} 
71681 # ALF cen B
{
   OrbitBarycenter "AlfCen"
   SpectralType "K0V"
   AppMag 1.34

   EllipticalOrbit {
      Period          79.914
      SemiMajorAxis   12.7872   # mass ratio 1.09:0.92
      Eccentricity    0.5179
      Inclination   82.98
      AscendingNode   67.71
      ArgOfPericenter 183.77
      MeanAnomaly     200.12
   }
}

Post #43by **granthutchison** » 23.10.2004, 22:09

selden wrote:Yes, they're logically redundant, and they aren't needed in the most recent build from CVS. However, I think the test version of Celestia that Grant's using is a slightly older build which didn't include that optimization.

For example, the current CVS build accepts the following ...

Ah, this is good news. While I was content to cut-and-paste if it meant Celestia had a clearer run at rendering, it did make for an untidy page.

Grant

symaski62 · Post #44by **symaski62** » 23.10.2004, 22:15

cel://Follow/Rigel%20Kentaurus%20A/2004-10-23T22:00:00.26227?x=AADA2r9tM5zu6eb//////w&y=AAAAPiCSQ5MvtNL//////w&z=AAAAawg//9O7nio&ow=-0.199117&ox=-0.177769&oy=0.365268&oz=0.891813&select=Rigel%20Kentaurus%20A&fov=0.012185&ts=1.000000<d=0&rf=38659&lm=0

Rigel Kentaurus A et Rigel Kentaurus b

Evil Dr Ganymede · Post #45by **Evil Dr Ganymede** » 23.10.2004, 22:31

selden wrote:Oh Evil One,

Yes, they're logically redundant, and they aren't needed in the most recent build from CVS. However, I think the test version of Celestia that Grant's using is a slightly older build which didn't include that optimization.

*snip*

Aha. That looks much better now

.

BTW, are these test builds only available to Celestia developers?

Post #46by **selden** » 23.10.2004, 22:55

Oh Evil One,

Once Chris is satisfied that the new code is sufficiently stable, I'm sure he'll post an announcement of the availability of a PreRelease kit for Windows.

Until that happens, you have to have a cvs client and a compatible C++ compiler in order to build the most recent version of Celestia for yourself. I use Cygwin's cvs plus Microsoft's Visual Studio 2003 Academic.

Celestia's source code is freely available from SourceForge's CVS server. The server that provides anonymous access for non-developers like me usually is updated about 5 hours after the server that's used by developers. That's worked out well for me, though. I usually do a build at about 7AM Eastern = 4AM Pacific. Recently Chris dosn't seem to have been uploading new code after 11PM Pacific.

Visit Celestia's summary page on SourceForge for more information. http://sourceforge.net/projects/celestia
Information about SourceForge's anonymous CVS access is at http://sourceforge.net/docman/display_doc.php?docid=14033&group_id=1

Does this help?

Evil Dr Ganymede · Post #47by **Evil Dr Ganymede** » 23.10.2004, 23:00

OK, I neither have a compiler nor do I know C++ anyway

. It just sounded like there were executables floating around somewhere that had the new stuff incorporated, kinda like "playtest versions" before the prereleases were made public. I can wait til that happens though

.

Post #48by **selden** » 23.10.2004, 23:06

Well, you don't actually need to understand C++.
The CVS archive includes a script (makerelease.bat) that Chris wrote. It usually builds the Windows version of Celestia with no significant problems.

Building under Linux seems to be another matter entirely, from what I've gathered from the various postings by people who do that.

Evil Dr Ganymede · Post #49by **Evil Dr Ganymede** » 23.10.2004, 23:30

selden wrote:Well, you don't actually need to understand C++.
The CVS archive includes a script (makerelease.bat) that Chris wrote. It usually builds the Windows version of Celestia with no significant problems.

But you still need Visual C++ or something like it to run the script don't you?

Post #50by **selden** » 23.10.2004, 23:51

Yes, unfortunately.

MS does provide a free version of their C++ compiler, but it isn't quite good enough to build Celestia without a lot of tweaking. Maybe someone with more patience than I have might be able to get it to work. I seem to recall that it was missing some include files, although maybe I just didn't have the search paths set up right.

wcomer · Post #51by **wcomer** » 24.10.2004, 00:36

Selden,

Today, I have suceeded in compiling Celestia with entirely free Microsoft tools. It took, as you surmised, some tweeking and patience. I am currently fine tuning the minimum requirements for any windows user, who lacks professional tools, to build their own Celestia executables. I hope to have a walkthrough up tonight or tomorrow.

BTW, is such a walkthrough better for the developer's or user's forum?

-Walton

Bold Italic · Post #52by **Bold Italic** » 24.10.2004, 00:43

Is it possible to compile Celestia with MINGW? :?:

I successfully use it to compile other open source programs...

It would be good to know before I download the source and attempt to compile... :wink:

BTW Hi all!

Evil Dr Ganymede · Post #53by **Evil Dr Ganymede** » 24.10.2004, 02:54

wcomer wrote:Selden,

Today, I have suceeded in compiling Celestia with entirely free Microsoft tools. It took, as you surmised, some tweeking and patience. I am currently fine tuning the minimum requirements for any windows user, who lacks professional tools, to build their own Celestia executables. I hope to have a walkthrough up tonight or tomorrow.

BTW, is such a walkthrough better for the developer's or user's forum?

Well, I'd certainly be interested to see such a walkthrough somewhere on these boards...

Evil Dr Ganymede · Post #54by **Evil Dr Ganymede** » 24.10.2004, 03:00

Waitaminute - with this barycentric system, and multiple star illumination this means it's actually possible to render a Ptolemaic solar system in Celestia doesn't it?!

I suspect you probably can't remove the Sun itself, but you could copy/paste all the bodies from the solarsystem.ssc file to a new system containing a G2 V star. Then you define a barycentre, place the barycentre in the middle of the Earth, then put the Moon in an orbit around it, then Mercury, Venus, the Sun (defined as a star here, thus providing illumination), Mars, Jupiter, and then Saturn.

I think it might be interesting to see how the solar system would look if this is possible. We could see just how many loops (like epicycles) that people had to jump through to try to get something that matched the observations...

Could this work in the new way of describing things?

wcomer · Post #55by **wcomer** » 24.10.2004, 07:54

Evil Dr,

There are some minor changes that need to be made to a few pow calls in the cvs before I can post the walkthrough. As soon as chris approves the changes, I'll post the walkthrough.

Post #56by **selden** » 24.10.2004, 12:53

Walton,

While I think that there are valid arguments for posting it to either of them, I'd suggest putting the walkthrough in the Developers Forum.

Post #57by **selden** » 24.10.2004, 13:04

Oh Evil One,

Indeed, what you suggest will work just fine, but you'd define the Barycenter first. (You have to define the orbited body before its satellites; order matters.) As you say, the star would be just one of many objects orbiting around it.

wcomer · Post #58by **wcomer** » 26.10.2004, 22:43

Selden,

I'll post it in the developers forum.

Evil Dr,

Chris fixed the relevent sections of code so I can put the walkthrough up. Unfortunately I will not likely have the free time to do so until thursday. Fortunately, Chris just made an executabel available for the 1.4 pre-release so you should have what you want... for now.

Evil Dr Ganymede · Post #59by **Evil Dr Ganymede** » 27.10.2004, 00:19

wcomer wrote:Selden,

I'll post it in the developers forum.

Evil Dr,

Chris fixed the relevent sections of code so I can put the walkthrough up. Unfortunately I will not likely have the free time to do so until thursday. Fortunately, Chris just made an executabel available for the 1.4 pre-release so you should have what you want... for now.

OK... still wouldn't mind looking at your walkthrough though. Like I said, I'm in no tearing rush here, just post whenever you're ready