It is currently Tue, 17-10-17, 20:39 GMT

All times are UTC




Post new topic Reply to topic  [ 47 posts ]  Go to page Previous  1, 2, 3, 4  Next
Author Message
 Post subject:
PostPosted: Fri, 07-12-12, 9:21 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4496
Location: Hamburg, Germany
Max,

unfortunately, in your add-on there is another essential problem (besides incorrect zonal flow speeds and directions): the position of the Great Red Spot is incorrect (unsyncronized), as can be demonstrated by displaying in addition the original Jupiter texture underneath the various cloud layers. When toggling the 'I' key (clouds ON|OFF) the problem becomes obvious!

Moreover, I recommend taking the latest 4k Jupiter texture from Cassini as a basis for constructing the various cloud layers. That texture is not part of the Celestia distribution (which includes only an older 1k texture)

Meanwhile, I have also made a cloud add-on myself, based on that 4k texture along with the published, measured zonal flow speeds and directions. I'll present the add-on here, when I am content with it.

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 08-12-12, 3:43 GMT 
Offline
User avatar

Joined: Fri, 29-02-08, 17:58 GMT
Posts: 86
Goofy wrote:
BTW, doubling all cloud layers heights solves the problem, no more defects at any distance (dunno why).

Great!


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 02-03-13, 1:48 GMT 
Offline

Joined: Thu, 28-02-13, 5:14 GMT
Posts: 4
I've been wondering about swapping out the default planet meshes in Celestia as I get more familiar with it.

The sphere in a sphere method is one I've used for, I think, improved effect in standard 3d programs. Useful for cloud layers on my version of Earth, as an example, and obviously with Jupiter too.

I've also found that using hi-rez sphere meshes deformed via a displacement map gets worthwhile results on bodies like Earth, moon, Mars. Is this something anyone has tried within Celestia? If so, what have you found the performance tradeoff to be?


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 02-03-13, 16:14 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4496
Location: Hamburg, Germany
UFO Partisan wrote:
I've been wondering about swapping out the default planet meshes in Celestia as I get more familiar with it.

The sphere in a sphere method is one I've used for, I think, improved effect in standard 3d programs. Useful for cloud layers on my version of Earth, as an example, and obviously with Jupiter too.

I've also found that using hi-rez sphere meshes deformed via a displacement map gets worthwhile results on bodies like Earth, moon, Mars. Is this something anyone has tried within Celestia? If so, what have you found the performance tradeoff to be?


Welcome at Celestial Matters!

Celestia renders planets by wrapping simple cylindrical 2d textures around spheroidal (ellispoidal) shapes in general. Irregular 3d meshes are not supported for planets (and not much motivated either). Celestia is to render the surfaces of planets and other bodies as seen from space, i.e. from some distance. For such non ground-based planetary perspectives, 3d effects via normalmaps on a slightly spheroidal shape are sufficient and much faster than other approaches.

When you talk about "worthwhile" results from displacement maps, what do you expect the advantage over normalmap approaches could be??
What do you call "worthwhile"? What texture sizes are you talking about?

Since unlike bump and normalmaps the displacement map deforms the underlying geometry, the generated 3d effects can correspondingly look good also at close distance..., yet at the expense of cpu time.

Here is the resolution level I have in mind:
Image
This earthbound image is normalmap-based.

For the VERY hi-rez renderings that some of us consider, like virtual textures (VTs) from 256k x128k 'monster textures', the main issues are fighting noise, lack of RAM & HD memory etc. Since the official Celestia approach is based on real elevation data, the smooth implementation of fine 3d surface details from 16bit integer scientific data via smooth, high-precision normalmaps can be done very well and fast with my Nm-Tools, for example.

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 02-03-13, 19:56 GMT 
Offline

Joined: Thu, 28-02-13, 5:14 GMT
Posts: 4
Haha, I'm not talking about a displacement map except as it deforms the sphere mesh in Blender. Then the mesh is deformed, exported as .3ds then converted to .cmod and popped into Celestia. It does work as I've done it, but haven't gotten the results I'm looking for. Will keep trying as I go, but I'm working on a lot with the video editing. It might be just as well to use a regular sphere with the normals mapping, but I like hacking around in text files as I used to render in POV-Ray and some of it's extensions, compiled my own Renderman shaders and so on.


Top
 Profile  
 
 Post subject: re
PostPosted: Sat, 02-03-13, 21:31 GMT 
Offline
User avatar

Joined: Tue, 04-09-07, 21:55 GMT
Posts: 766
Location: N 42.38846 W 83.45456
for blender there is a tool to import PDS DEM's
for example the moon LOLA 4ppd map ( th 16 ppd can be used but the triangle count will slow down anything but a cluster with 128 Gig ram
http://blenderartists.org/forum/showthr ... 1-06-25%29

i know for a fact it runs on blender 2.64
2.66 ? is unknown

you can see my last post on the thread
http://blenderartists.org/forum/showthr ... ost2246212

BUT this thread is getting Off Topic with
"Dynamic model of the atmosphere of the Jupiter"

so if you have not done so
Please start a new thread

_________________
"I don't pitch Linux to my friends, I let Microsoft do that for me."
Using OpenSUSE 42.1 & Scientific Linux 6.7


Top
 Profile  
 
 Post subject:
PostPosted: Sun, 03-03-13, 4:57 GMT 
Offline

Joined: Thu, 28-02-13, 5:14 GMT
Posts: 4
It sounds like normals mapping is very similar to displacement mapping. Still, I think your sphrere-in-a-sphere solution for Jupiter is an excellent one and can be a solution for other objects in Celestia or 3D apps.


Top
 Profile  
 
 Post subject:
PostPosted: Mon, 04-03-13, 15:34 GMT 
Offline
User avatar

Joined: Sun, 02-09-07, 23:16 GMT
Posts: 119
Location: Upstate NY, USA
One of Celestia's limitations is that it does not draw shadows which are cast on the surface of an object by that object's own protrusions. It only draws an area on a surface as dark (shadowed) when its surface normals point away from a light source. An area which has its surface normals pointing toward a light source is drawn as if it were illuminated even if there's a mountain between it and the light source.

_________________
Selden


Top
 Profile  
 
 Post subject:
PostPosted: Wed, 06-03-13, 5:04 GMT 
Offline

Joined: Thu, 28-02-13, 5:14 GMT
Posts: 4
Selden wrote:
One of Celestia's limitations is that it does not draw shadows which are cast on the surface of an object by that object's own protrusions. It only draws an area on a surface as dark (shadowed) when its surface normals point away from a light source. An area which has its surface normals pointing toward a light source is drawn as if it were illuminated even if there's a mountain between it and the light source.


That's a pretty small limitation when you look at the big picture of what Celestia can do. If you really felt the need for that, you might as well render out a scene in a 3D app. Even though the path setup is really simple, the time to render it out takes a lot longer than making a movie in Celestia and frankly most people wouldn't even notice the difference.


Top
 Profile  
 
 Post subject:
PostPosted: Wed, 06-03-13, 18:44 GMT 
Offline
User avatar

Joined: Sun, 02-09-07, 23:16 GMT
Posts: 119
Location: Upstate NY, USA
FWIW, Cosmographia does support that kind of "self shadowing".
http://www.cosmographia.info/
http://code.google.com/p/cosmographia/

Cosmographia is the 3D program Chris Laurel has spent the past few years developing. Most of the source is freely downloadable and isn't hard to build for Windows (maybe Linux; I haven't tried). The commercial Mac app adds some descriptive text.

_________________
Selden


Top
 Profile  
 
 Post subject: re
PostPosted: Wed, 06-03-13, 19:29 GMT 
Offline
User avatar

Joined: Tue, 04-09-07, 21:55 GMT
Posts: 766
Location: N 42.38846 W 83.45456
Cosmographia builds just fine on Opensuse 12.1 and 12.2 64 bit
the svn pull has no issues
( there was an old one but it has been fixed )

Code:
svn checkout http://cosmographia.googlecode.com/svn/trunk/ cosmographia

_________________
"I don't pitch Linux to my friends, I let Microsoft do that for me."
Using OpenSUSE 42.1 & Scientific Linux 6.7


Top
 Profile  
 
 Post subject:
PostPosted: Thu, 02-05-13, 11:59 GMT 
Offline
User avatar

Joined: Sun, 25-11-12, 11:08 GMT
Posts: 9
Hi all :) I've corrected some layers, speeds (real) and directions. Now it's as close to reality as i can make :D
Image
Download yadi.sk/d/p8m68juC4W2pQ


Top
 Profile  
 
 Post subject:
PostPosted: Thu, 02-05-13, 13:28 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2726
Location: France, South, not far from Montpellier
Tx arctodus, it's a too big add-on for my machine (10x 4096 maps for a single planet!) but apart a few artifacts here and there it looks nice.


Top
 Profile  
 
 Post subject:
PostPosted: Thu, 30-05-13, 19:44 GMT 
Offline
User avatar

Joined: Tue, 04-09-07, 8:49 GMT
Posts: 121
Location: Oxford UK
Okay... I'd like to try... how do I install?


Top
 Profile  
 
 Post subject: Re:
PostPosted: Fri, 07-06-13, 21:59 GMT 
Offline
User avatar

Joined: Thu, 25-10-07, 15:20 GMT
Posts: 991
Location: NE PA, USA
t00fri wrote:
For the VERY hi-rez renderings that some of us consider, like virtual textures (VTs) from 256k x128k 'monster textures', the main issues are fighting noise, lack of RAM & HD memory etc. Since the official Celestia approach is based on real elevation data, the smooth implementation of fine 3d surface details from 16bit integer scientific data via smooth, high-precision normalmaps can be done very well and fast with my Nm-Tools, for example.

Fridger

He He,
Nice looking forum. Kinda looks like it used to when logged in. Kinda. Thought I'd test it out. I was doing some calculations on what size a height map would be if I used the SRTM-3 at full resolution. I think it comes to 432000 x 216000. Not 43200x21600. I don't know what to call it as far as a K number. 432K? This is about a 100 GB file! A Monster indeed. Maybe I made a mistake? The font colors are cool. Good job on the forum.
cartrite


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 47 posts ]  Go to page Previous  1, 2, 3, 4  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group