Star textures in celestia.cfg

[Cham]

EDIT: The changes are now checked into CVS.

I recompiled and tested that new code. It's working great ! And our stars have now their personality back.

I just have to find a good set of stars textures. My old set is now looking a bit "oldish".

From now on, the stars textures generation from the users may become as much active than for the planets.

[ElChristou]

...I just have to find a good set of stars textures. My old set is now looking a bit "oldish".

What about a set using your nill momentum trick?

[Cham]

What about a set using your nill momentum trick?

Unfortunately, that trick is only working dynamically. I don't think I could generate a good looking static texture from this, and without any polar pinch effect. The only way I know is to edit all the old stars textures to generate new textures. We're in the same boat as for the planets here.

At least, there's now a kind of "textures democracy" between planets and stars, from now on.

[Vincent]

1.5.0 is going to be a nice release.
I think there will be no doubt anymore on what is the "Best Version of Celestia to Use"...

[Hungry4info]

Cham, do you have some screenshots?

Good question. Can somebody please post some screen shots showing the new stars?

[Cham]

Good question. Can somebody please post some screen shots showing the new stars?

Hungry4info, the new stars are almost exactly like the ones in 1.4.1, we just have a choice for their textures, now. And since the default textures are the same as before, the changes aren't obvious, unless you add your own textures.

[Hungry4info]

Ah, alright. Thank-you. Will this be implimented in the next 1.5.0pre/final?

Post Scriptum: Thank-you very much for modifying the stars.

[chris]

Ah, alright. Thank-you. Will this be implimented in the next 1.5.0pre/final?

The changes will be in the next prerelease, which I'd like to have ready within 2-3 days (depending on the feelings of the other developers.)

--Chris

[Cham]

he changes will be in the next prerelease, which I'd like to have ready within 2-3 days (depending on the feelings of the other developers.)

What are the other changes to be done, before that prerelease ?

[chris]

he changes will be in the next prerelease, which I'd like to have ready within 2-3 days (depending on the feelings of the other developers.)

What are the other changes to be done, before that prerelease ?

There are some fixes to star rendering in scaled disc mode, and one bug fix related to reference frames:

https://sourceforge.net/tracker/index.p ... tid=121302

--Chris

[Cham]

and one bug fix related to reference frames:

https://sourceforge.net/tracker/index.p ... tid=121302

Will this also fix the bug I described to you, about the rotation behavior of a model after a long period of time (4000 years), if a reference frame is used ?

[chris]

and one bug fix related to reference frames:

https://sourceforge.net/tracker/index.p ... tid=121302

Will this also fix the bug I described to you, about the rotation behavior of a model after a long period of time (4000 years), if a reference frame is used ?

No, I still don't understand the source of that bug, and I'm not yet able to replicate the problem myself.

--Chris

[Cham]

Vincent has replicated the rotation problem on his machine.

Did you tried the SSC (with reference frame) that I gave to you, with Jack's data for the Spitzer space telescope ?

[chris]

Vincent has replicated the rotation problem on his machine.

Did you tried the SSC (with reference frame) that I gave to you, with Jack's data for the Spitzer space telescope ?

I was able to replicate the problem by letting the year reach 6000 (not 4000 as in the original report.) That's the year at which the VSOP87 series expressing the motion of the Sun around the solar system barycenter is no longer used. Instead, the Sun remains fixed with respect to the SSB. With the Sun not moving, the reference frame described in spitzer.ssc is no longer valid. One of the axes of that frame is the velocity of Spitzer with respect to M51, a fixed object as far as Celestia is concerned. Spitzer isn't moving either--the SSB is considered fixed, the Sun is considered fixed with respect to the SSB after the year 6000, and the motion of Spitzer relative to the Sun is described by an xyz file valid over a limited time range some time around the present. Thus, the velocity vector of Spitzer is zero, and can't be used to define a reference frame. In short: Don't Do That. I could figure out something better to do with the Sun's motion outside the range for which VSOP87 is valid, but I don't know of any longer range predictions for this motion. If you just want to point Spitzer at M51, you don't need a velocity vector anyway; use a constant vector instead and the frame will be valid at any time.

--Chris

[Cham]

Ah ! Thanks for the explanation Chris. Makes sense.

EDIT : Ok, I changed the frame used and it solved the problem. So your explanation was the right one. Thanks.

[chris]

Ah ! Thanks for the explanation Chris. Makes sense.

I'll have some time here soon to give an example of a reference frame using that you can use to point at any object in the sky and that doesn't require a velocity vector.

--Chris

[Cham]

Ah ! Thanks for the explanation Chris. Makes sense.

I'll have some time here soon to give an example of a reference frame using that you can use to point at any object in the sky and that doesn't require a velocity vector.

I'm using this one now (no bug so far) :

Code: Select all

"Spitzer Space Telescope" "Sol"
{
   Class "spacecraft"
   Mesh "spitzer.3ds"
   Radius 0.0025
   Beginning 2452876.772222222
   #Ending 2455195.795793577
   Albedo 0.7
   SampledOrbit "spitzer.xyz"
   Orientation [90 1 0 0]
}

Modify "Spitzer Space Telescope" "Sol"
{
   BodyFrame
   {
      TwoVector
      {
         Center "Sol/Spitzer Space Telescope"
         Primary
         {
            Axis "z"
            RelativePosition
            {
               Observer "Sol/Spitzer Space Telescope"
               Target "m 51"
            }
         }

         Secondary
         {
            Axis "y"
            ConstantVector
            {
               Vector [0 0 1]
               Frame { BodyFixed { Center "Sol" } }
            }
         }
      }
   }
}

Can this code be simplified while doing the same ?

[chris]

Can this code be simplified while doing the same ?

The only change that I would make is to change the frame of the constant vector to EquatorJ2000. In this frame, the vector [ 0 0 1] points at the north celestial pole. Using the BodyFixed frame of the Sun works means vectors other than [0 0 1] and [0 0 -1] will be moving--potentially quite confusing if you naively modify those vectors at some point in the future.

--Chris

[Cham]

Oh by the way Chris, before doing the new prerelease, how do you activate the new rendered vectors in Celestia ? These may be pretty usefull in some situations.

[ElChristou]

I know, I know, this is a trivial issue but...

Could the Stars textures being renamed something like:

star_a.jpg
star_b.jpg
star_g.jpg
star_m.jpg
star_browndwarf.jpg

?

This would help to keep the textures folder in order...