Celestia Forums

A test version for the most luminous stars, in M33. I find it pretty nice. The HR plot is too homogeneous, though...

There are 20 000 stars in this case, and I'm using the "scaled disk" rendering of stars to see them more easily.



I'm beginning to think about galactic arms, since with very luminous stars, it could be interesting...

I'm still waiting Fridger's response, about the function used to define the galactic arms (galactic templates)...

About the arms, I'll play with the classical logarithmic spiral. However, there are lots of possibilities with the three parameters which defines the logarithmic spiral (in parametric form) and I need to know what Celestia is using to draw its spiral galaxies :



EDIT : Mathematica has the ability to superpose a picture and a mathematical curve. So I could easily find the best match, while playing with the curve parameters (in this case, M33).

When is this ready for use??

Cham wrote:I'm still waiting Fridger's response, about the function used to define the galactic arms (galactic templates)...

The templates were NOT produced by means of an analytical math formula. This would not look realistic in the end, since also between the arms some "glowing" material is required. On average, logarithmic spirals are of course the simplest mathematical approximation for the shape of the arms.

The templates are quite complicated to do, requiring a lot of experience how the code associates sprites of a large range of sizes to the pixels on the template. I did all templates by hand (Gimp) in multi-layer technique. A number of layers are the same for each template, as noted by my friend ElChristou (Christophe).

The basic shapes are obtained from an overlay with reference photos that first need to be reprojected to exact face-on view. A crucial aspect are the various brightness levels of the pixel "noise" in-between the arms. The template size is 128x128 pixels.

[ In Celestia.Sci the templates for spirals are much improved and the elliptical galaxies are entirely computer generated with dedicated shader code.]

Cham wrote:I need a list of all spiral galaxies within 10 millions LY ! Someone ?

My galaxy database (galaxies.dsc) in the Celestia distribution comprises ALL galaxies (including dwarfs) within the Local Group volume V <~ (10 Mpc)^3. That should do, right?

Fridger

t00fri wrote:On average, logarithmic spirals are of course the simplest mathematical approximation for the shape of the arms.

I intend to use the logarithmic curves (blurry version, actually) to dispose some stars over the quasi-homogeneous (radial) distribution. That should do it.

t00fri wrote:My galaxy database (galaxies.dsc) in the Celestia distribution comprises ALL galaxies (including dwarfs) within the Local Group volume V <~ (10 Mpc)^3. That should do, right?

Actually, there aren't much spirals in this volume, I think (Milky Way, M31 and M33), unfortunately (or fortunately ?).

Fridger, you didn't responded to my pm yet ?

Cham wrote:Fridger, you didn't responded to my pm yet ?

Didn't arrive here. Typo?

If you have attached an executable file (...zip,...), it will be resent to you.

Fridger

t00fri wrote:

Cham wrote:Fridger, you didn't responded to my pm yet ?

Didn't arrive here. Typo?

If you have attached an executable file (...zip,...), it will be resent to you.

Fridger

????!!!???

I'm sending it back to you... in 2 min.

EDIT : please check your pm...

Today, I built an STC file of 1 MILLION stars for M31 (Andromeda galaxy). The STC file is 138.3 MB in size and Celestia takes about 1 minute to load.

The star search is pretty laggy, but the 3D view is absolutely spectacular, especially with a sprite based nebula (arms) !
The stars are following the M31 arms structure.


With the Magnitude pushed to the max :

Here's the cross-section of an arm, in M31. The nebula visible on the left isn't the default one. It's a sprites based model superposed on the default representation (the default representation is OFF, in the view below), and truly looks like a real picture ! The frame rate is surprisingly high on my 2009 Mac mini (with 1 million of stars !?!).



EDIT : Here's a small video of the 1 million stars distribution. The video is a bit laggy because of the screen capture utility...

http://fsgregs.startlogic.com/Public_Fi ... es/M31.mov

3 years ago I tried to fill the Milky Way galaxy with stars, but the project was finally abandoned. It is possible to use the new star generator for our galaxy as well?

Spaceman wrote:It is possible to use the new star generator for our galaxy as well?

Of course. However, I don't want to build a stellar distribution for our Milky Way, to not interfere with the realistic stars there. Maybe for the unknown "quadran" behind the MW's galactic core, one of these days, but I think there's enough stars available in other galaxies (M31, M33, NGC 404, ...). See this topic :

viewtopic.php?f=23&t=16626

When is new star generator to be available?

Reiko wrote:When is new star generator to be available?

It will never be available. It's a Mathematica code. I'm releasing its output STC files only.

Cham wrote:

Spaceman wrote:It is possible to use the new star generator for our galaxy as well?

Of course. However, I don't want to build a stellar distribution for our Milky Way, to not interfere with the realistic stars there. Maybe for the unknown "quadran" behind the MW's galactic core, one of these days, but I think there's enough stars available in other galaxies (M31, M33, NGC 404, ...). See this topic :

viewtopic.php?f=23&t=16626

Having stars with fictional planetary systems in other galaxies is something always welcomed. I'll grab them too very soon
However I'm sure that not everybody cares about realism, when it's about having stars in our galaxy as well. Sure having a red giant 1 light year from the Sun is not the best , but still it can be very useful for world builders and science fiction writers . Just my 2 cents...