[m2258734a]

Hello, I thought it would be better to create a new topic rather than bumping an ancient one that ended abruptly.

With LS' nebulae available, I take that the majority of you all are satisfied with the background features of Freespace. However, I still believe that the main stars of all systems need to be enhanced. A long time ago, I started a topic over on the SCP forum called "Background features". The topic didn't really get anywhere as far as improving the look of the stars of Freespace, but I believe it did create an initiative to apply LS' nebulae templates to FS2 (Are the multiple star configurations the same as those featured in the original templates, such as Capella being represented as the quadruple star system that it is? I don't have the capability of running Freespace SCP on my laptop, so I currently don't have anyway of knowing).

The purpose of this topic is to gather ideas on how each star should appear in the game, and hopefully encourage any of our extremely talented designers to recreate them. I understand that Freespace is not an astronomy simulator, but applying one's artistic license in a manner based on real stellar data might help liven up the Freespace atmosphere and awe the users. So far, I have edited some information on the Freespace Wiki. I have only touched stars beginning with 'A' that way you all can see what I have done. I honestly didn't do anything major: I fixed some abbreviations, changed the description "color" to "spectral class", updated numerical values, etc. If you would like to see more information, like stellar radii or temperature, let me know and I'll go ahead and add it to the Wiki. I am doing this not only to update the Wiki, but to establish a template for those who would like to redo the stars. Every star has its own characteristics and behaviors, and no two stars should look the same in Freespace.

What do you all think? I honestly would love to do this myself as astronomy is my passion, but I simply lack the tools and expertise necessary to recreate my visions. Well, that's all from me. If anyone is interested I would be glad to provide any helpful information. Please feel free to provide your ideas, information, and feedback.

EDIT: I recently tried some new programs and it looks like I got the links to work. I will take the liberty to enhance the stars' appearances myself, provide the original images of the stars on this thread, and allow anyone else to finalize the overall images.

[S-99]

Having accurate star maps would be cool. But, since 3.6.7 we do have the new much better star field as compared with the starless vanilla fs2 starfield.

[Darklord42]

never the less it would be nice if they were accurate,  Only there won't be as many wonderful nebulas in the background.

[Raven2001]

hmmm theres a big problem with that... remember that unless you calculate for each system, the star pattern your referring to is the one you can see from earth

That means that in any other system, the stars wont be seen on the same pattern

[m2258734a]

I see what you all are saying. Well, determining what the local star field should look like is no problem if you have programs like Celestia.

Ultimately, this is not what I was referring to. In most of the Freespace and Freespace 2 missions there is a main star, or multiple stars, which is much larger and brighter than the stars within the field. I was referring to stars like Antares, Betelgeuse, Vega, Deneb, Gamma Draconis, Capella, i.e. the star systems within the storyline. What you all have mentioned is another cool idea, as I also think it would be a nice addition to add a more accurate star field according to the location of each system.

IIRC, Deneb is the first star system we encounter in FS2. So I'm saying why not improve the look of Deneb, because right now it looks dull. Deneb is one of the most luminous stars in the sky, and the most luminous star of its spectral class. As a white supergiant, Deneb in Freespace just doesn't live up to its real-life counterpart. It's not even the right color. There's more to be said about Deneb, and all other stars in Freespace.

As for local star fields, I can produce some images if you all would like.

[Admiral Nelson]

I started something like this.  I have a list of all the spectral types of all the stars in all of the FS2 system that exist IRL.  I also have a list of the RGB color values for all spectral types.  The idea was to create stars that are of the correct color for every system and insert these into the Lightspeed mission pack.  This activity proved quite time consuming, and I never got very far.  If you are interested in working on such a project, I'd love to include these in the mission pack.

I do think it might be interesting to have a skybox for Sol that reflected actual star positions.  We wouldn't be able to recognize the night sky in other stellar systems and so have these correct is of less value.  After all, the nebulae background themselves aren't really realistic, anyway.

[Raven2001]

In games like FS, theres certain things of realism that can and should be forgotten, in order to promote the coolness factor

[Herra Tohtori]

After all, the nebulae background themselves aren't really realistic, anyway.

Meh, FS2 != realistic game. Old argument and almost as old response; you just have to presume that the helmet visor or cockpit canopy multiplies background light from weak targets in order to make it easier for pilots to determine their state of motion better; almost uniform starfield would do a poor job at that, but when you multiply weak intensities, you get this in Sol system:



...I dunno, I think that (but perhaps slightly downtoned in brightness/contrast) would be a rather cool environment to fly in...

[Admiral Nelson]

Actually, it is a different argument this time.  The fellow seems to want to produce an accurate starfield for each system that exists IRL.  I presume an application such as Celestia can do so.  However, the nebulae selected to appear in mission are not those that would appear if enhanced.  Thus the gain from the accurate starfield seems minimal, except in the case of Sol, where we can recognize constellations, etc.  I do think it would be nice to correct the colors of the stars themselves.

[m2258734a]

I apologize if I wasn't clear.

I agree with you; I do believe that color is one of the physical properties of the main stars of Freespace that need to be corrected and enhanced. I also have a list of RGB color values that correspond to each stars' spectral classification and luminosity type, so it looks like we're on the same page. When I say stellar enhancements, I don't mean those applied to the background as a whole, but only to the main stars of a particular system, which is either one, two, or more depending on the type of binary system the star is a part of. You all brought up some interesting ideas, and it would be possible to create specific star field backgrounds for each system. I could provide pictures of what these different star fields should look like, but it was never my intention to go that deep into making the Freespace environment that realistic. Since LS' nebulae are fantasy, they obviously do not pertain to a specific Messier Object or any other recognizable deep space object. The purpose of trying to make these accurate star fields would go unnoticed with so many of these fictitious nebulae obscuring a large portion of the sky. I am in absolutely no way saying that the nebulae shouldn't be there, they look gorgeous and really liven up the scenery. But if someone would like to take the time to make these accurate star fields regardless, I would be glad to provide images for them.

Ultimately, the purpose of this topic was to come up with ways to enhance the main stars of a particular system. I mentioned the supergiant Deneb as one example that really needs a makeover. IMO, the stars do not fit very well or add any awe to the Freespace environment. In all honesty, they look dull and lack any interesting feature which sets each one apart.  If anyone is interested, I would be glad to share my ideas on how each star can be physically improved. I don't own Photoshop, or any other application similar to that, so I cannot express my ideas as an image. I provided links to Photoshop plugins in the first post to see if anyone might want to give it a shot. I hope I cleared things up 

EDIT: That shot of the galactic plane is beautiful. I couldn't agree with you more.

[Admiral Nelson]

I think we are exactly on the same page.  I would recommend downloading GIMP, which is free, and should be good enough for the work, if you are interested.  I have every system in FS2 already mapped to the spectral classes of all the stars in them.

[m2258734a]

Awesome, I'll look into GIMP asap. This looks very interesting....

[Alan Bolte]

Agreed on the main stars concept, that never really occurred to me but it would liven things up a bit. One thing that I'd really like to see though is a pic like above of the milky way. If you can see it from the ground, you should damn well be able to see it from space. It would be a really good way to stay oriented, and would give a real sense that each system is in the same part of the galaxy.

[Herra Tohtori]

Actually, it is a different argument this time.  The fellow seems to want to produce an accurate starfield for each system that exists IRL.  I presume an application such as Celestia can do so.  However, the nebulae selected to appear in mission are not those that would appear if enhanced.  Thus the gain from the accurate starfield seems minimal, except in the case of Sol, where we can recognize constellations, etc.  I do think it would be nice to correct the colors of the stars themselves.

Ah. Yes, it is a bit different, I agree.

It is true that practically only the Sol system's starfield would get any advantage from having realistic stars; in other systems the relation of work required and the adcantages gained would be ridiculously off-balanced, so it doesn't really make any sense. But I would really love to see the galactic plane as a Sol skybox...

As to the star colours. I couldn't agree more either. But how are you going to define what RGB colour the stars look like? Because a star with spectral classification G2V (like our Sun) will appear white due to the fact that human eye recognizes daylight as white. On the other hand, human eye can recognize also different colours as white, depending on their current white balance mode. You can do interensting experiments yourself, by the way - keep another eye closed for perhaps five minutes when reading a book for example. Then open both eyes and compare how the paper looks like - other eye sees it a bit orange, the other sees it bluish... that's because the white balance of the eyes is different. The effect will pass in few minutes. But I digress.

What this means to the star colours is that absolute correctness is not required, because to some limits, human eye adapts to dominant light as being white light. Also note that with most stars, the actual colour of the star is usually very close to white with only slight tint of colour notable - you only need to take a look at the clear night sky to notice that most stars definitely don't have much recognizable colour into them. Most oar yellow (ie. white) stars like our sun, only the most blue or red stars are have definite colour into them. There's a table in Wikipedia about Morgan-Keenan spectral classification, which uses RGB values to indicate what colour is associated to different spectral classes. They go like this:

Code: [Select]

Spectral RGB
Class Colour

O #9bb0ff
B #aabfff
A #cad7ff
F #f8f7ff
G #fff4ea
K #ffd2a1
M #ffcc6f
For our purposes, those values should be sufficient for the colour of the stars. Of course, those values are a bit boring because they don't differ from each other very much. If artistic freedom and prettyness requires more varying colours, you can always use the other list, in Main sequence stars:

Code: [Select]

Spectral RGB
Class Colour

O2 #3D6BFF
O5 #3E6CFF
B0 #4472FF
B5 #5785FF
A0 #7CA5FF
A5 #9CBDFF
F0 #B1CCFF
F5 #D4E4FF
G0 #EDF4FF
G2 #FDFEFF
G5 #FFF6E9
K0 #FFE9CB
K5 #FFCB91
M0 #FFAE62
M5 #FF8A38
M8 #F00000
M9.5 #D00000

That will result in more variable, if not overly realistic, stars. In this case I would probably go for the second list, even if I'm usually for realism. In either case, those values should be of sufficient accuract considering the human eye adaptivity to different colours being recognized as "white".

[Darklord42]

 On the other hand, human eye can recognize also different colours as white, depending on their current white balance mode. You can do interensting experiments yourself, by the way - keep another eye closed for perhaps five minutes when reading a book for example. Then open both eyes and compare how the paper looks like - other eye sees it a bit orange, the other sees it bluish... that's because the white balance of the eyes is different. The effect will pass in few minutes. But I digress.

I noticed that along timeago but thought that it was just me going crazy

[m2258734a]

I think the galactic plane should be a background feature for all star fields, but the only problem with this is that it might become very repetitive to see the same band of light in every system. While it would be realistic, the galactic plane's features won't noticeably change as we go from system to system. Leaving the galactic plane for only the solar system would be a good idea, but at the same time I know that the plane should be visible everywhere else in our stellar neighborhood. I don't know, what do you all think about having the galactic plane visible for all systems in the Freespace Universe?

As for the defining RGB values, I have them already defined in a database that I discovered when I was working on a visual astronomy simulation with a friend of mine in Spain (still a WIP). The database I collected originated from SIMBAD, which is a very good source of astronomical data. The RGB colors chosen are dependent on the temperature of the star, which of course vary within a spectral class. So basically I have a list of all of the RGB values these stars should have according to their spectral classification from O0-M9, and based on their luminosity type from type I supergiants, to VI subdwarfs (very rare luminosity type).

In the case of the actual colors of the stars, I was thinking of adding realism combined with an artistic touch. Stars are extremely close representatives of black bodies, and by assuming that the Sun is one, we can calculate the peak radiation the Sun emits based on its temperature. With a surface temperature of ~5800 K, we get ~500 nm. By knowing that the intensity of radiation is a function of the wavelength emitted, we can produce functions for typical black bodies of various temperatures.



The intensity of radiation through the visual spectrum determines what colors we see, and this makes sense. Cooler stars have peak wavelengths in the infrared region, and so the intensity of red light is higher than that of all the other colors in the visual spectrum. The opposite is true with hotter stars, as they peak in the ultraviolet region. The Sun lies in between these two extremes, and the intensity of light through the visual spectrum is roughly constant. Combining the intensities of visible light together, we end up with an emission of white light.

From time to time, I have heard stars of different spectral classes having the colors described as follows: O-blue, B-bluish-white, A-white, F-yellowish-white, G-yellow, K-orange, M-red. This is an extreme simplification and is not really accurate; however, it hints the idea that blue stars are hot and red stars are cool. This description does not take into consideration the fact that the combination of wavelengths of different intensities will not produce a pure color. Only until we get to the extremes such as a very hot star, or a very cool star, will the dominant color be blue or red, respectively.

Anyways, I thought that was just an interesting subject to talk about. What Herra Tohtori mentioned was absolutely correct (it is a very interesting experiment BTW), but I believe the reasons which were described also have to do with the idea that the Sun actually is white and not yellow as we commonly perceive. So while the true color of the stars might not be a pure orange or pure red, I think it would be fine to increase the vivacity of the color, as long as this change in vivacity is constant throughout the stellar spectrum.

In the meantime, I'm trying to learn how to use GIMP. I have never used any software like this before let alone try to create my own images in this fashion, it's interesting nevertheless. Thanks for the input.

[JGZinv]

If you are more experienced with Photoshop, or want to use it as a prep to get
into Photoshop - I can suggest the overhauled version of GIMP called GIMPshop.

http://www.gimpshop.com

[Agent_Koopa]

Well, seeing the rest of the galaxy would be pretty realistic, but I don't know if it would be worth it. You're adding a lot to the background and seeing a background that rich in detail would certainly get irritating after a while. Sometimes you need space (no pun intended). On the other hand, if the Milky Way is only visible from Sol, won't that look quite suspect, though atmospheric, especially for the last missions of a campaign? If we leave the galactic plane out, then we'll get posts from people who say it's missing.

[m2258734a]

Well, I have completed my first star. I am content with the results, but if someone would like to take it over and provide finishing touches I would be happy to give it away. First up (going completely out of alphabetical order) is Deneb:



As I said before, Deneb is the most luminous star of its spectral class that we know of so far. I made sure that the star has the correct color based on the estimated RGB color value for a spectral class A2 supergiant. It worked out pretty well, because as you see it does have a slight bluish hint to it, especially in the corona. My main focus was on the luminosity of the star, and I tried my hardest to give the supergiant an extremely vivid appearance. To give you an idea of just how luminous Deneb is, take into consideration that Deneb is about 3000 ly away from the Sun, yet it is the 19th brightest star in our sky. The brightest star in the sky, Sirius, is only about 8 ly away. Compared to the Sun, Deneb is about 160,000 times more luminous, so hopefully I portrayed that luminosity in this image. One thing that I would like to improve is giving Deneb a "supergiant" appearance. What do you all suggest?

Well, I'd be glad to receive your feedback, and feel free to edit the star if you believe something needs to be changed.

EDIT: I went ahead and updated the picture to add a little more clarity to the surface of Deneb, as well as some other minor improvements such as enlarging the corona to correspond with the large radius of the star.

[Admiral Nelson]

Looks good so far.  Can you post a link to the original file and provide the RGB values you used for the color?  Stars.tbl will need an entry for Deneb with these RGB values so that the in game light coming from the star is accurate.