3d nebulas?

[david1701e]

Hi, I'm new to Celestia and I have been trying to find 3d "immersion" nebulas. All I have found is 2d nebulas that are floating in space. I have found nebulas you can fly through in various video games, so I know the tech exists I just need help finding them! thanks.

[fsgregs]

David:

Most of the 3D nebulas we have available for Celestia can be found on the Celestia Motherlode add-on page for nebula. Its address is: http://www.celestiamotherlode.net/

Look in the Extrasolar catagory under both Messier objects and other nebula, and read the descriptions. It will say whether the add-on is 3D of not, allowing you to fly through it.

Frank

[t00fri]

Hi, I'm new to Celestia and I have been trying to find 3d "immersion" nebulas. All I have found is 2d nebulas that are floating in space. I have found nebulas you can fly through in various video games, so I know the tech exists I just need help finding them! thanks.

If you know of scientific publications that tell us reliably, how the 3rd dimension looks like, 3d nebulas are no problem .

Galaxies are different since they are more regular objects and we can see them from all sorts of inclinations.

F.

[Kreso]

If you know of scientific publications that tell us reliably, how the 3rd dimension looks like, 3d nebulas are no problem .

Galaxies are different since they are more regular objects and we can see them from all sorts of inclinations.

F.

Well, I wouldn't mind seeing nebulas in Celestia that aren't entirely accurate in the 3rd dimension, if the following two guidelines were followed:

First, this objects should look as close as possible to the actual nebulas, at least as seen from Earth, and second, it should be made clear in the readme file that the appearance of nebula as seen from another point in space is purely hypothetical and artistic.

That way, such addons would not be very different from content already in Celestia, like textures of asteroids and extrasolar planets or surface details outside the limit of knowledge... There are no "scientific publications that tell us reliably" how this things look, nevertheless, we can accept them as hypothetical and artistic visual improvements. 2D billboard representations of nebulas and deep sky objects are plain ugly, not realistic at all, and, because of their flatness, are definitely not accurate representations of actual nebulas.

[BobHegwood]

2D billboard representations of nebulas and deep sky objects are plain ugly, not realistic at all, and, because of their flatness, are definitely not accurate representations of actual nebulas.

To this, I heartily agree...
However, you may wish to take a closer look at some of the nebulae listed on the Motherlode HERE.
The He2-104 Southern Crab and the M2-9 Planetary Nebula are both fully 3D representations of these nebulae if I remember correctly.

Just FYI...

EDIT: One other note here... I have listed all of the "Billboard" nebulae in the primary description fields, and those that are NOT billboards are somewhat more pleasing to the realist's point of view. Keep in mind, however, that these too are also "fictional" representations of real phenomena.

[rthorvald]

Galaxies are different since they are more regular objects and we can see them from all sorts of inclinations.

Without better information, i should think assuming any given one looks the same from all directions is better than pretending all nebulas to be flat...
Perhaps not scientifically correct, but less uncorrect than a 2D billboard?

- rthorvald

[t00fri]

add-on creators are free of course to do and upload whatever they please. I am only concerned with renderings within the official Celestia distribution. These are to be based on scientific input only.

Fridger

[rthorvald]

add-on creators are free of course to do and upload whatever they please. I am only concerned with renderings within the official Celestia distribution. These are to be based on scientific input only.

Of course, and i know the challenges of mass rendering nebulas - you have described it before. My post yesterday was a question, not an opinion.

Let me put it a different way: is it not more correct to add the nebulas as generic 3d "blobs" than not having them at all? Then detail can be added if/when better understanding of the objects arrive (or as individual add-ons, as you say).

- rthorvald

[Kreso]

add-on creators are free of course to do and upload whatever they please. I am only concerned with renderings within the official Celestia distribution. These are to be based on scientific input only.

I would really like to read scientific publications that tell us reliably about the surface and cartographic features of extrasolar planets implemented in the official Celestia distributions. Because in the official Celestia distribution extrasolar planets have textures which are, of course, purely artistic and hypothetical, and nobody seems to complain about that.

But that is not my point here. To me, it would be completely acceptable that these nebula objects are published as a single add-on and are not included in the official Celestia distribution.

As I have already mentioned above, these objects should be "scientifically accurate" when viewed from Earth, but I wouldn't mind if the 3rd dimension is artistic and hypothetical, because, at present, there is no way to scientifically verify that it is incorrect.

The He2-104 Southern Crab and the M2-9 Planetary Nebula are both fully 3D representations of these nebulae if I remember correctly.

These 3D objects are, of course, better than 2D pictures, but they could be made even better, instead of just sticking a semi-transparent photo of the nebula to a sphere or any other geometrical body. Nebulae should be fluffy, cloudy and foggy - of couse, on their scale. They should be lit by stars within and emit their own light, and scatter light realistically...

Has anybody tried to implement something like that?

I am more interested in the technical solutions than the actual modeling of every single nebula in the sky (I guess that would be an enormous job).

[the bluemarble guy]

something interesting to try would be some sort of layering method, in which we take the image of the nebula, and go over the little details with a layer marker to define certain parts of the image as being in front of other parts. Also, it should be possible to define the shape of certain parts, such as branching off nodes and such. It could then be placed through some sort of cloud renderer to get the shape and save to a model (or multiple models if needed).

One example of a nebula where this reasoning could be used to aproximate the shape would be the carinae nebula, which when you look at it in high res hubble images, you can realy see the ridges and nodes that make up its shape.

Also, we could do this with something like the dark cloud part of the horse head nebula, which would b placed in front of an emmision nebula to make it more realistic than a single billboard (cloud type rendering within a certain shape could be used for the more shapeless nebulae.).

[t00fri]

add-on creators are free of course to do and upload whatever they please. I am only concerned with renderings within the official Celestia distribution. These are to be based on scientific input only.

I would really like to read scientific publications that tell us reliably about the surface and cartographic features of extrasolar planets implemented in the official Celestia distributions. Because in the official Celestia distribution extrasolar planets have textures which are, of course, purely artistic and hypothetical, and nobody seems to complain about that.
...

Not really true, we apply a quite consistent rendering philosophy and documentation throughout!

A good prototype example are Celestia's 10000+ galaxies. They are computer rendered from generic templates for the Hubble morphology classes, any other property of the Celestia galaxies (including size, orientation, color, ...) being rendered directly from reading out the original catalalog parameters. Nothing else has been added!

In the same sense, the rendering of exo planets is based on just a few generic template textures, exo-class1 ... exo-class5, that express (besides size information if existing), some generic (temperature/spectral,...) knowledge about these planets. No further details were assumed, really. The class assignment is based on published scientific work!

More precisely as documented in great detail in the respective datafile (extrasolar.ssc):

• Generic textures are assigned to giant planets based on their estimated temperatures, according to the five categories set out by D Sudarsky, A Burrows & P Pinto in Albedo and Reflection Spectra of Extrasolar Giant Planets, Astrophysical Journal 2000; 538: 885-903 and http://www.journals.uchicago.edu/doi/pdf/10.1086/309160
• Color values and albedos are based on the theoretical reflectance spectra given in that paper.
• Planets with an estimated temperature greater than 1000K have been assigned a glowing nightside texture.

The assignment algorithm of exo-planet radii is also precisely documented there.

So it's always good to have a look in the data files (and respective PERL scripts) first, where all scientific input is usually quoted ...

Fridger

[BobHegwood]

So it's always good to have a look in the data files (and respective PERL scripts) first, where all scientific input is usually quoted ...
Fridger

And very much appreciated too.

[Kreso]

A good prototype example are Celestia's 10000+ galaxies. They are computer rendered from generic templates for the Hubble morphology classes, any other property of the Celestia galaxies (including size, orientation, color, ...) being rendered directly from reading out the original catalalog parameters. Nothing else has been added!

In the same sense, the rendering of exo planets is based on just a few generic template textures, exo-class1 ... exo-class5, that express (besides size information if existing), some generic (temperature/spectral,...) knowledge about these planets. No further details were assumed, really. The class assignment is based on published scientific work!

I don't want to get into pointless debates... We should all be friends here. But, a lot of generic material in Celestia has a lot of non-scientific artistic details. For example, nobody has ever observed sunspots on HIP 23518, but, in Celestia, since this star is M class, it's texture is that of a M class star, with randomly placed sunspots. And, if we look at the planet b orbiting HIP 21850, we can see it uses the generic texture exo-class1, and has two red cloudy stripes near it's equator (which, I am sure, cannot be directly observed - I have read the Sudarsky - Burrows - Pinto paper and there is, of course, nothing about this specific two red stripes on HIP 21850 b). Also, Uranus's moon Cordelia uses the generic asteroid texture which has craters that also nobody observed. This details are artistic, and, of course, not scientifically accurate. Otherwise this would imply that generic textures of asteroids are better charted than the surface of Pluto, which is blurry, because this is our limit of knowledge. If Celestia was entirely 100% scientifically accurate, most of the asteroids would not have textures at all, or at least extremely blurry textures (much, much more blurry than Pluto - in fact, so blurry that there are no distinguishable details at all, except for those measured), and be completely spherical (because we don't know anything about their shape and texture). Actually, it would be best that this asteroids were just billboard signs with text "Details unknown". Yet, they are not such, because they would look ugly - therefore we have artistic generic textures, which, nobody denies, are based upon actual scientific results, but also on a lot of imagination. And I think that this is good and that we should tolerate imagination, as long as 1) it is obvious that this is imagination, 2) there is no conflict between science and art, and if there is a conflict between artistic imagination and science, science should always have the last word.

Kreso

[ajtribick]

If it bugs you so much, why not just switch on Limit-of-Knowledge textures using the + key and be done with it?

[Kreso]

If it bugs you so much, why not just switch on Limit-of-Knowledge textures using the + key and be done with it?

Exactly! This is a good point - if nothing in Celestia was artistic or fictional as Mr. Fridger claims, there wouldn't be any need for a Limit-of-Knowledge switch (which is there to turn of all the art and leave just the science). I, personally, enjoy all the artistic detail in Celestia as long as I'm aware it's art, and as long as it doesn't contradict real science.

To get back on topic: There is content in Celestia which is not based on scientific facts alone, but also on scientific artistic imagination, and this is a good thing, because it can be turned off if not desirable. Why then ignore all beautiful nebulae, only because their third dimension is (and probably always will be) uncertain?

[t00fri]

A good prototype example are Celestia's 10000+ galaxies. They are computer rendered from generic templates for the Hubble morphology classes, any other property of the Celestia galaxies (including size, orientation, color, ...) being rendered directly from reading out the original catalalog parameters. Nothing else has been added!

In the same sense, the rendering of exo planets is based on just a few generic template textures, exo-class1 ... exo-class5, that express (besides size information if existing), some generic (temperature/spectral,...) knowledge about these planets. No further details were assumed, really. The class assignment is based on published scientific work!

I don't want to get into pointless debates... We should all be friends here. But, a lot of generic material in Celestia has a lot of non-scientific artistic details.

A number of such remaining non-scientific rendering details are left-overs from early times of Celestia. They will certainly be eliminated in the official distribution ASAP. People who want to make artistic contributions are welcome to submit add-ons.

Certainly, since a few years, we are cleaning up the Celestia data base and improve the scientific documentation. Also details don't matter, it's the general consistent rendering philosophy that is important. Certainly in this process, we shall not release 3d nebulae that have no observational basis. Nebulae are the DSO's that are hardest for automatic rendering, since there is least symmetry that may be exploited by computer rendering.

Fridger