Flying mode for Celestia

[Seb]

I'm sure this has been mentioned before. I would like to see some flying 'mode'. Using the AZ & cursor controls, not orbiting; but as well the current speed of direction it would need to include its own internal relative speed of movement as with any bodies within space.

Also restrict the speed to more 'real' speeds & acceleration, and add in something that highlights the nearest stars to you with some info like its distance and type. (example below)

What would be interesting if after this, to the including the effects of gravity of the nearest planets on the vehicle. We can then simulate the forces of getting the vehicle off the surface and more.

[Enio]

Very interesting. The Doctor Who Universe have the flying mode witch is vey cool, and I think this feature could be implanted in Celestia. It's a very good idea, because you feel that you are really in space.

[andyrock]

Please don't limit the speed

I don't want to wait 10 years in front of the monitor to travel in fly mode to some star out there..

[Seb]

I knew someone would say that. I am talking about a flying 'mode', ie. a mode you can switch on and off.

[Bob Hegwood]

Excuse me for being Brain-Dead, but I thought we were flying through
space already?

Can you explain what you mean a little better? When I'm viewing a planet, star or moon, I'm flying around it, over it, under it and even through
it. What else is there?

Thanks, Bob

[t00fri]

...

What would be interesting if after this, to the including the effects of gravity of the nearest planets on the vehicle. We can then simulate the forces of getting the vehicle off the surface and more.

Celestia is not an arcade game, but supposed to be a realistic space simulation. Taking account of gravity properly is much more difficult than you can presumably imagine.

Otherwise we would have incorporated gravity since a long time. This subject has indeed been brought up about a hundred times already;-)

Bye Fridger

[Seb]

Bob - Yes we can, but when switching to the flying mode, it adds a couple more tweaks as to make movement similar of that of a space craft.

All it needs is to changes some controls, have its own relative movement and to restrict acceleration and speed down a bit. This would give a more realistic movement as to give a more realistic observation on space.

t00fri - I'm not suggesting an arcade game, just to simulate movement a little better - in tune with Celestia being a space simulator.

If then to take this a stage further by adding in gravity I'm sure I can provide you with the formula. Just look at say the nearest 10 objects, look at each of them in turn, calculate or estimate they local g force, scale it down (probably logarithmic) by distance and apply the force to our space craft.

I've done some physics and a programmer myself, but maybe I'm being naive in thinking this is simple.

[t00fri]

...
t00fri - I'm not suggesting an arcade game, just to simulate movement a little better - in tune with Celestia being a space simulator.

If then to take this a stage further by adding in gravity I'm sure I can provide you with the formula. Just look at say the nearest 10 objects, look at each of them in turn, calculate or estimate they local g force, scale it down (probably logarithmic) by distance and apply the force to our space craft.

I've done some physics and a programmer myself, but maybe I'm being naive in thinking this is simple.

Cool, ...

I am sure you can provide me with a formula. But I want a /correct/ one;-)

Look:

it is clearly "Kindergarden level" to add up a certain number of idealized point sources for the gravitational field. But this can only be be a decent approximation to the true force field in a very limited extremal case. If we were to incorporate gravity effects on the observer's "spaceship" in Celestia, we also have to deal with the most complicated case of navigating in between the satellites of Jupiter, say. What do you guess how much even the ellipsoidal shape of Jupi distorts the field around one of the moons? Etc.

Now let me see your formula;-)

Bye Fridger

[Seb]

Gravitational force is directly proportional to mass.

Basically find the mass of the objects, divide this by the square of the distance. (from the centre of each object)

eg.

G = 6•67 ? 10^-11 (a constant)
M = star/plant mass (earth = 5.97?10^24 kg)
m = our space craft mass
r = distance between the two
s = Craft current speed

force = G x M x m / (r^2)

s = s +f (this last one is not really this simple, need to apply force at the correct angle)

[t00fri]

Gravitational is directly proportional to mass.

Basically find the mass of the objects, divide this by the square of the distance. (from the centre of each object)

eg.

G = 6•67 ? 10^-11 (a constant)
M = star/plant mass (earth = 5.97?10^24 kg)
m = our space craft mass
r = distance between the two
s = Craft current speed

force = G x M x m / (r^2)

s = s +f (this last one is not really this simple, need to apply force at the correct angle)

You may have written this before reading my post above...

This is "Kindergarden gravity" and at best OK for arcade games! I explained above that adding up a number of point sources of mass M to form the total field is highly incorrect in many cases of interest.

As you can see from your formula above, it implies that the gravitational force only depends on the distance from the source mass. Clearly this is wrong in general. Take a big body with nonspherical shape like Jupiter. Move around close to it....

In case of planetary systems the total force arises from the highly overlapping individual contributions due to the (small) moons and the big planet. A situation that can at best be simulated with supercomputer power.

Again, let me see your formula;-)

Bye Fridger

[granthutchison]

Fridger:
You're being offensively patronizing. Why don't you just explain clearly what the problem is, rather than taunting Seb repeatedly? Then we can all move on.

Grant

[t00fri]

Fridger:
You're being offensively patronizing. Why don't you just explain clearly what the problem is, rather than taunting Seb repeatedly? Then we can all move on.

Grant

Grant.

I think I did explain the problem clearly without becoming technical.

Bye Fridger

[Seb]

If we are working from the centre of the mass, though not perfectly accurate, it does work quite well. As with moons, would need to add in all the forces from them all the nearest ones. This I should think should be at least is 95% accurate.

Arcade games use much simpler formula than I have suggested here, as they use a fixed force which is not even applied relatively to the object.

Think we are getting of the main point now.

[t00fri]

If we are working from the centre of the mass, though not perfectly accurate, it does work quite well. As with moons, would need to add in all the forces from them all the nearest ones. This I should think should be at least is 95% accurate.

Arcade games use much simpler formula than I have suggested here, as they use a fixed force which is not even applied relatively to the object.

Think we are getting of the main point now.

What precisely do you mean with the statement
"it works quite well"?? Against what measurements did you test the formula?

Bye Fridger

[Seb]

"it works quite well" - ie. force increases/decreases with surface attitude.

I thought this formula was well known. Newton / Kepler's Laws.

Yes - if we did want to measure it perfectly we would need a super computer. But then do we really need to do this for the purposes of Celestia?

[Bob Hegwood]

Bob - Yes we can, but when switching to the flying mode, it adds a couple more tweaks as to make movement similar of that of a space craft.

All it needs is to changes some controls, have its own relative movement and to restrict acceleration and speed down a bit. This would give a more realistic movement as to give a more realistic observation on space.

Well, perhaps I still don't understand correctly, but wouldn't a change of
this type further degrade the movement options we have now?

In other words, right now I can use the wheel on my mouse to move
slowly toward or away from a given object. I *like* this feature. I also
like being able to follow an object, and being able to synchronize my
viewpoint above an object at any position I desire.

If I understand what your suggesting correctly, the simple controls we
have now would become unwieldy, would they not? I don't want to have
to worry about gravity, and solar wind and the effects of any of a million
different other problems while I'm browsing my universe.
If I have positioned myself to view a planet's surface at - say 100 km -
I wish to stay there.

Sorry... Not trying to be hostile. I just really like the way we can move
around in Celestia now. Is it that you would just like to see the effects
of gravity during your wanderings?

Thanks, Bob

[Seb]

Bob,

This is a flying 'mode', meaning this mode can be switched off, so that you can still continue to browse things as you can currently.

When switch on, yes - would degrade, as real-world movement isnt really this easy.

[t00fri]

"it works quite well" - ie. force increases/decreases with surface attitude.

I thought this formula was well known. Newton / Kepler's Laws.

Yes - if we did want to measure it perfectly we would need a super computer. But then do we really need to do this for the purposes of Celestia?

Seb,

sure Newton's law is very well known, but inapplicable in many /practical/ situations, when e.g. close-by extended (non-spherical) bodies are building up a complex gravitational field like in case of planetary systems.

Just to give you an idea: Celestia uses the very sophisticated VSOP87 theory, with hundreds of Tschebychev polynomial terms accounting for the cumulative and most nontrivial gravitational effects of other bodies onto the orbit under consideration.

[In case you remember some atomic physics: conceptionally, it's the analog of the Hatree-Fock approach.]

If you want to describe an observer's "spaceship" navigating in such a complex gravitational environment, you have to use very complicated numerical approaches to get a fair approximation to reality!

A familiar example for the inherent complexity is a precision description of the lunar motion that requires either the even more sophisticated, semi-analytical VSOP2000 theory or a most demanding numerical integration approach, the JPL ephemerides code, DE405, DE406.

The disagreement between the latest VSOP2000 solution and DE406 is now down at less than 3 arcsec in longitude and 2km in distance over the whole time interval of [-3000, +3000] years!!

Celestia for many VSOP87 orbits achieves sub arcsec precision and precise timings of mutual eclipsing events of numerous moons of Jupiter etc. All these facts have been extensively tested against a host of precision data and extensive independent calculations in the past.

It would be a shame to spoil this impressive precision of Celestia by adding in inaccurate gravity effects at this point...


Bye Fridger

[Seb]

I dont claim to be an expert in this field, thought Celestia didn’t display Earth's shape perfectly, or calculate its orbit perfectly. Actually, just thinking about it, wouldn’t earth also have a mini orbit of its own with the moon - is this included? (Quick calculation of about 5000 mile orbit.)

ref: http://shatters.net/forum/viewtopic.php?t=5785

ok, gravity may not be calculated well - I don’t know how accurate the formula performs in real life, just seems in my mind, accurate enough for this flying mode.

[t00fri]

I dont claim to be an expert in this field, thought Celestia didn’t display Earth's shape perfectly, or calculate its orbit perfectly. Actually, just thinking about it, wouldn’t earth also have a mini orbit of its own with the moon - is this included? (Quick calculation of about 5000 mile orbit.)

ref: http://shatters.net/forum/viewtopic.php?t=5785

ok, gravity may not be calculated well - I don’t know how accurate the formula performs in real life, just seems in my mind, accurate enough for this flying mode.

In order to get "spot on" predictions for satellite motion overhead, the orbit data must be updated very frequently via ftp (once every few days! It should be obvious why this is necessary...). We are thinking about a more or less automatic update facility within Celestia.

Incorporating oblateness into planets is not quite a trivial affair, since it interferes with virtual textures etc. This is still under development.


Bye Fridger