Distance/radii limits for planets labels/orbits to show

[Cham]

Andrea,

I'm thinking about the "half-Earth" model, placed on the back of elephants and/or a turtle. At Ptolemy time, they knew the Earth was round, but not as a sphere.

[Cham]

Something in this spirit :

http://en.wikipedia.org/wiki/Image:Paul ... cworld.jpg

[ANDREA]

Please, please, HELP!

What value do you have the minorbitsize set to?
--Chris

Chris, I used

Code: Select all

 set {name "MinOrbitSize"  value 3.0}

as suggested by Vincent.
I'm using 1.4.1
Bye

Andrea

Try setting it to a smaller value--1.0, or even 0.5.

--Chris

Chris, I tested up to 0.01, the max distance doesn't change going lower than 0.5.
This is the result at 0.5:



So I think that with this (0.5) setting the situation is satisfactory for our purposes, many thanks.
Just for my curiosity: why orbits, labels and globes appearing/disappearing are connected among them?
Cannot understand the reasons why, sorry.
Bye

Andrea

[ANDREA]

Something in this spirit :
http://en.wikipedia.org/wiki/Image:Paul ... cworld.jpg

Oh, I already understood, I remember your temple with Buddha sitting inside.
Sorry, I never found anything available for such purposes.
Anyhow, all the images we used and will use for the Earth textures at the various ages of geographical knowledge, have been found here:

http://www.henry-davis.com/MAPS/carto.html

Hope you'll find something you can use.
Good luck.
Bye

Andrea

[Cham]

Andrea,

I'm talking also about things like this picture here :

http://www.henry-davis.com/MAPS/Ancient ... ientL.html

[ANDREA]

Andrea, I'm talking also about things like this picture here :
http://www.henry-davis.com/MAPS/Ancient ... ientL.html

Well, ve like same things, Cham.
In my show on the Earth, starting from the vision that men had in the old times, I use this image:



that's the same of this



but heavily enhanced/modified.

Andrea, .. At Ptolemy time, they knew the Earth was round, but not as a sphere.

Cham, I don't agree.
For what I know on the matter, on 500 B.C. Pit? goras believes the Earth to be a sphere. on 250 B.C. Aristarcus demonstrates it IS a sphere, and on 200 B.C. Heratostenes obtains a sufficiently precise measure of Earth diameter, using the shadow of Siene's well.
Regarding Ptolemy, we are speaking of Claudius Ptolemaeus (Greek: ca. 90 ?€“ ca. 168 AD.
Bye

Andrea

[buggs_moran]

Andrea,

I'm thinking about the "half-Earth" model, placed on the back of elephants and/or a turtle. At Ptolemy time, they knew the Earth was round, but not as a sphere.

I disagree. As you are no doubt aware, Eratosthenes figured the circumference of the Earth within a few percent error and although Ptolemy (only about 100 years later) used a smaller globe, he still knew it was round. Also, everything was spheres to him, so I find it difficult to believe that he thought otherwise...

Turtles all the way down comes from another culture
http://en.wikipedia.org/wiki/Turtles_all_the_way_down

[buggs_moran]

Oh and BTW, I am reworking my celestial sphere but will not be done until I return from holiday next week. All the data is there, just need to compile...

[Cham]

On the sphere, let me precise what I was saying :

Yes, Erathostene and Hipparque (among some others) suspected the Earth was spherical, AT LEAST A PART OF IT. They in fact even suspected that the Sun was at the center, not the Earth ! They were able to estimate the size of the Sun and even the distances. Few times later, that knowledge was forgotten, in part because of Aristotle and his "physics".

[Cham]

Andrea,

for your model of the Ptolemy system, are you trying to be rigourous, or just to build an approximation of his system ? Did you included the excentricity of the planets (the circular orbits weren't centered) ? If I remember correctly, Earth wasn't in the exact center : Ptolemy placed it at a different location (not exactly at the center) to explain the observations. I think this was the "point ?©quant" (sorry, it's in French).

[Fenerit]

Andrea, I see only now the continues of this topic, sorry. A question: is the sphere of the fixed stars syncronized to turn with the other celestial bodies? This it's not important for educational purpose whereas the word subst the images, but mechanically, as I've clarifyed to Buggs, that sphere must rotate with all planets carried altogether by her in 24h. And then the planets must moves of seasonal motion. This fact lead Copernicus to say the it was a "chimera", a "monstrum" and Galileo that an unique motion (of Earth) would be explained all. Another one: the motion of Mercury is very complex; it must be on an epicycle the center of which move itself from west to east about an eccentric deferent which centre moves in opposite sense around a small circle on which circumference is contained the equant at midway from the centre of small circle and the centre of Earth. Your project do this really?
Nevertheless GO AHEAD in this project, it's REALLY important for the whole encompass of the astronomy.

What you think about a sphere of fixed star mapped with the Uranographia addon : http://www.celestiamotherlode.net/catal ... _stars.php as figurative rapresentation of stellar vault?

[Cham]

Chris,

I confirm Andrea's bugs. I'm unable to setup a mini-solar system. Most orbits and labels aren't drawn. As a specific example, I defined a simple small star as follow (a neutron star) :

STC code :

Code: Select all

"Center1"
{
   RA  0
   Dec 90
   Distance  1000
   SpectralType "Q"
   AbsMag 27
   RotationPeriod 26
}


Then, I defined some small planets moving around, like this (notice the size of the bodies and their orbits) :

SSC code :

Code: Select all

"Earth" "Center1"
{
   Class "spacecraft"
   Texture "earth.*"
   Radius 6378

   EllipticalOrbit {   
      Period      9E12
      SemiMajorAxis   0.001
      MeanAnomaly   0
   }
}

"Center2" "Center1"
{
   Class "planet"
   Radius 0.001

   EllipticalOrbit {   
      Period      9E12
      SemiMajorAxis   0.001
      MeanAnomaly   180
   }
}

"Center3" "Center1/Center2"
{
   Class "moon"
   Radius 0.001

   EllipticalOrbit {   
      Period      1
      SemiMajorAxis   100000
   }
}

"Test Planet" "Center1/Center2/Center3"
{
   Class "spacecraft"
   Radius 4000

   EllipticalOrbit {   
      Period      0.1
      SemiMajorAxis   10000
   }
}


As you can see here, "Center2" is a fixed planet around the star "Center1". The moon "Center3" is orbiting the "Center2" planet, while "Test Planet" is orbiting the moon "Center3".

I then noticed several bugs :

1- The central star "Center1" has a depth sorting problem in front of the "Earth" planet. At some scale, the star can't be seen IN FRONT of "Earth". It appears BEHIND it while it should be the reverse.

2- the "Center3" and "Test Planet" orbits are invisible, and their labels as well.

3- I can select the "Test Planet" and watch it moving around, using the markers. Its motion is fine (cycloidal motion). But it can't be seen otherwise. That body is invisible. It should be a white ball (no texture is defined here).

What is going on ? Did I made some stupid mistake here ?

As you know, I'm using Celestia 1.5.0pre3.

[selden]

1. It isn't the star that's being depth sorted incorrectly, it's the glow around it that's being drawn wrong. If you GoTo Center1, you'll see that the star itself is drawn with the correct depth sorting.

2. & 3. Whether or not a body, its orbit and its own satellites are drawn depends on the radius of the object it orbits around and your distance from it. If you set Center2's Radius to be greater than about 10 or 100, you'll see them when you get closer. My understanding is that this is "as designed" -- an octree culling decision which one can disagree with.

To design the orrery "correctly" one should use the new object type ReferencePoint. But that has its own bugs: the orbits of ReferencePoint's aren't drawn.

The orbit of a Class "invisible" isn't drawn either.

I think both of these types of SSC objects need to have orbit and label rendering options. And colors.

So at the moment, one needs to specify a "reasonable" Radius and use a Mesh for the Centers which draws an invisible object: one with a transparent material. Or one can specify a Mesh and not provide it. Although Celestia complains, it'll produce the desired visual effect. It won't draw the Centers, but it will draw their orbits.

[ANDREA]

Andrea, for your model of the Ptolemy system, are you trying to be rigourous, or just to build an approximation of his system ? Did you included the excentricity of the planets (the circular orbits weren't centred) ? If I remember correctly, Earth wasn't in the exact cente: Ptolemy placed it at a different location (not exactly at the center) to explain the observations. I think this was the "point ?©quant" (sorry, it's in French).

Cham, we wished to be as rigorous as possible.
For this reason yes, the Earth in not at the center, but it rotates symmetrically around the system barycentre, that anyhow, in order to show a bigger Earth, lies within the Earth globe (like the real Earth-Moon barycentre lies within the Earth, deep under its surface).
All the planets rotate aroung epycicles, and all planets epycicles and Sun rotate around the ?€?equant?€

[chris]

Chris,

I confirm Andrea's bugs. I'm unable to setup a mini-solar system. Most orbits and labels aren't drawn. As a specific example, I defined a simple small star as follow (a neutron star) :


As you can see here, "Center2" is a fixed planet around the star "Center1". The moon "Center3" is orbiting the "Center2" planet, while "Test Planet" is orbiting the moon "Center3".

I then noticed several bugs :

1- The central star "Center1" has a depth sorting problem in front of the "Earth" planet. At some scale, the star can't be seen IN FRONT of "Earth". It appears BEHIND it while it should be the reverse.

I have a simple fix for this that I will check in later today. It turns out that any stars with a screen size of less than one pixel weren't being depth sorted. This wasn't a problem in normal solar systems because the star is larger than any of the planets orbiting it. But in your sample system, the star is tiny relative to the planets, so it's easy to have a view where the subpixel star is in front of a > 1 pixel sized planet.

Code: Select all

2- the "Center3" and "Test Planet" orbits are invisible, and their labels as well.

3- I can select the "Test Planet" and watch it moving around, using the markers.  Its motion is fine (cycloidal motion).  But it can't be seen otherwise.  That body is invisible.  It should be a white ball (no texture is defined here).

What is going on ?  Did I made some stupid mistake here ?


Selden is correct: what's occurring here is that your tiny bodies are being culled along with the larger objects orbiting them. Celestia doesn't render the satellites of a planet when the planet itself isn't visible. Normally, this works just fine, but you can construct cases where objects are culled improperly. In your sample solar system, the radius of the object test planet is four million times larger than the planet it orbits! I will revisit some of these culling decisions in 1.5.1. For now ReferencePoints (or invisible objects) are the right solution, not tiny planets.

--Chris

[Cham]

So if I use large invisible objects, it should solve the problem ?

[chris]

So if I use large invisible objects, it should solve the problem ?

They don't even have to be large . . . If you're not worried about Celestia version prior to 1.5.0, you can use ReferencePoints instead of invisible objects. Both will work, but a ReferencePoint is more concise and more clearly expresses its intended function.

--Chris

[Fenerit]

[quote]Moreover remember that, due to the final educational destination of the full addon, even if we have some numerical data given by Ptolemy, we have been obliged to ?€?compress?€

[ANDREA]

Andrea, for what concern the distances amongst planets, remember that Ptolemaic system maintained the "orror vacui" doctrine, so that the planets mights be succesively placed one after one; and that the theory of planets showed in the Almagest make sense when they are taken one by one and not altogether.
Bye.

Fenerit, we have the Ptolemy distance ratios, that were 1, 6, 12, 50, 115, 170 for Mercury, Venus, Sol, Mars, Jupiter and Saturn respectively, but they were unpractical because the System looked too "sparse" this way.
So we decided to vary such distances compressing the System, maintaining anyhow the relevant movements, or at least, as I told, the most important ones.
Bye

Andrea

[Fenerit]

Well Andrea, I apologize if I insist, but I'm very curious about how you and Buggs did have realized the rotation of the whole planet's system and of the fixed stars in 24h.