New addon : WMAP layer

[Cham]

Here's another layer for Celestia SVN : the Cosmological Microwaves Background Radiation (WMAP for short). The texture is properly aligned with our galaxy (but see the note below) and comes from this nice web page :

http://lambda.gsfc.nasa.gov/product/map/current/sos/

I'm giving here a very low resolution texture in jpg format since I can't upload anything larger than 256 KB in the forum. If you want a larger (and MUCH nicer) texture, go to the previous web page and download the 4k PNG version in the upper-left corner. Do a 180° horizontal flip in your image editor, to have a properly oriented texture in Celestia (this operation needs a confirmation, though). Resize to 2048X1024 if your card can't handle the 4k version.

There's a celx script included to toggle ON/OFF the layer in Celestia, using "Q" (shift-q). Of course, a layer defined in your start.celx file is MUCH better, though :

WMAP.zip

With the high resolution version (2k or 4k), you shouldn't notice any polar pinch effect !

NOTE : I would much appreciate any comparison with other sources, just to confirm the texture is properly shown in Celestia.

[ElChristou]

After running the script, Q will crash Celestia...

[Cham]

Hmmm, maybe I forgot to save it as UTF-8 ?

Just open the file with your text editor, edit the French comments, and save it as UTF-8. Still crashing ?

Here's a view of the layer :

wmap.jpg

The green line is the galactic plane.

[ElChristou]

Hmmm, maybe I forgot to save it as UTF-8 ?

Just open the file with your text editor, edit the French comments, and save it as UTF-8. Still crashing ?

Ok, now. Do you have a shot showing the alignment of the texture with some of the bodies we have in database? (it's to compare and see if all is alright)

[Cham]

Ok, now. Do you have a shot showing the alignment of the texture with some of the bodies we have in database? (it's to compare and see if all is alright)

I know the texture is properly aligned with the galactic plane. There's no doubt about this (there are other textures with a view of our galaxy, on the web page I gave above). The only uncertainty is the 180° horizontal flip I made. I used the old addon from Selden as a reference. But I want a confirmation of that flip. And maybe there's an offset to do. I strongly suspect there isn't any offset to do, though, since there are some "bulges" properly aligned with the galactic center (using other textures from the web site).

[Cham]

I've found a clue that my texture is correctly flipped. On this web site (textures for Celestia), there's a CMB texture which is almost identical to mine (last texture, at the bottom of the page) :

http://laps.noaa.gov/albers/sos/sos.html.050613

Can somebody find a map of the CMB with some galactic coordinates on it ?

[Cham]

Got it !

I had to make an offset of 180° and a vertical flip. Here's a comparison with what is shown in GoogleEarth (from this site : http://lambda.gsfc.nasa.gov/product/map ... oogle_sky/) and the corrected texture in Celestia :

GE :

wmapgoogle.jpg

Celestia :

wmapcelestia.jpg

As you can see with the black handmarks, the match is perfect ! Even the Rosetta nebula is at the right spot !

And I find quite curious that there's a "hot" spot in the wmap which coincides exactly with the Orion Nebula :

orion.jpg

[ElChristou]

Got it !

I had to make an offset of 180°...

Right or left?

[Cham]

Got it !

I had to make an offset of 180°...

Right or left?

Well, it's the same, no ? It was simply an offset of half the texture (so 180° in coordinates).

[ElChristou]

Got it !

I had to make an offset of 180°...

Right or left?

Well, it's the same, no ? It was simply an offset of half the texture (so 180° in coordinates).

Yes but I don't get the same as you screen...
Will try again a bit later...

[Cham]

Ok, here's the procedure :

Get the texture from the site (any texture on this page) :
http://lambda.gsfc.nasa.gov/product/map/current/sos/

0. Resize the texture to 2048 X 1024 (4096 X 2048 on a mesh is very hard for the video card).
1. Do an horizontal flip.
2. Then, do an horizontal offset of half the size of the texture (offset 1024).
3. Do a vertical flip.
4. Save to JPG. I'll publish a version for PNG, which is MUCH better looking in Celestia.

Should be correctly oriented, with the mesh I made (basic operations in Cheetah).

[Cham]

I've added another layer, to show the CMB with and without the effects of the Milky Way (it's fun to be able to switch the layers on the fly, using the keyboard only ! ) :

CMB without the Milky Way :

layer1.jpg

Same view, with the effects of the Milky Way. The green line is the galactic plane :

layer2.jpg

It is unfortunate that the polar pinch effect is still a bit visible, in the second case (not shown here).

Where is Fridger ? He would surely appreciate this kind of "Cosmo-Celestia" stuff !

[ElChristou]

Where is Fridger ? He would surely appreciate this kind of "Cosmo-Celestia" stuff !

Still on holidays!

[Cham]

Using another layer, I'm noticing some interesting correlations between the WMAP and the H2 regions of the Milky Way :

WMAP alone :

wmap.jpg

H2 region alone (same view) :

H2.jpg

Both layers at the same time :

wmapH2.jpg

[Cham]

Despite what I've said in another topic, I'm adding the velocity of our solar system relative to the CMB. The arrow is pointing to the famous Virgo super-cluster :

velocity0.jpg

velocity1.jpg

Sadly, I'm experiencing a precision problem at the base :

velocity3.jpg

I'm not sure how I'll solve this problem. Maybe I'll give the base some fading effect, so we don't see the offset. Maybe I'll have to use a purely CMOD arrow.

[ElChristou]

Despite what I've said in another topic, I'm adding...

I knew you would not be able to stay a long time without falling again!

[Cham]

Despite what I've said in another topic, I'm adding...

I knew you would not be able to stay a long time without falling again!

DOH !

[t00fri]

Where is Fridger ? He would surely appreciate this kind of "Cosmo-Celestia" stuff !

Still on holidays!

I'am back (as you know) and have yesterday exchanged exactly 19 emails with Cham about some subtleties of WMAP data. I certainly find it worthwhile to further enhance Cosmo-Celestia aspects.

Fridger

[Cham]

I've found a pretty nice way to correct the base offset :

base.jpg

The trick was a bit risky : I exported the offending piece to 3ds, and converted it to a TEXT CMOD file. I then edited the points on the base so it's now exactly on Sol !

[Cham]

I finally got the orientation right !

So here's an updated version of the WMAP (no script included), with solar system velocity vector oriented toward the CMB dipole (about the same direction as the Virgo cluster). For a larger hires PNG texture, follow the instruction I gave on the first page. The texture given here is tiny !

WMAP.zip

To toggle ON the CMB WMAP and the velocity vector, you'll have to use a script like this (French information in the example below). I also suggest you toggle ON the "cosmic layer grid" I published elsewhere :

Code: Select all

wmap_t = {
      "WMAP",
      "Vitesse Sol"
   }

for k, object in pairs(wmap_t) do
  wmapObj = celestia:find(object)
  wmapObj:setvisible(false)
end

function toggleWMAP()
  for k, object in pairs(wmap_t) do
    wmap = celestia:find(object)
    wmap:setvisible(not wmap:visible())

   if wmap:visible() then
      celestia:print("Fond diffus de rayonnement cosmologique affich?.\nTemp?rature moyenne du rayonnement : 2,725 K (COBE).\nFluctuations angulaires de la temp?rature : ± 0,0002 K (WMAP).", 12)
   else
      celestia:print("WMAP d?sactiv?", 2)
   end
  end
end
---------------------------------------------------------------
---------------------------------------------------------------
keyhandlers =
{
    Q = toggleWMAP,
    S = toggleCosmicGrid
}

function handlekey(k)
    handler = keyhandlers[k.char]
    if (handler ~= nil) then
        handler()
        return true
    else
        return false
    end
end

celestia:registereventhandler("key", handlekey)

The data used for the orientation of the velocity vector comes from this paper :
http://articles.adsabs.harvard.edu/cgi- ... .470...38L

If there is some better data, please let me know.