Board index FlightGear Development Scenery

Would you want more stars in the night sky?

Questions and discussion about enhancing and populating the FlightGear world.

Would you want more stars in the night sky?

Postby eatdirt » Sat Jan 26, 2019 2:05 pm

Hello there,
I've been playing around with FG since a few months with the SpaceShuttle, and discovering (how to crash) other planes and helis as well!

Maybe the subject has been put forward long ago, but, as an amateur astronomer, I can clearly tell that the number of visible stars by night is not enough, and by quite a lot. Our current sky is looking like a quite polluted sub-urban sky, and that is not what the sky is at 10000ft or elevation, for sure!

I've explored a bit the code, and found simgear/scene/sky/stars.cxx, which confirms that we cut the visual magnitude at 4, a comment mentions that this is to keep the number of stars below 500. I played with the code, and I have done a few things:

1) Calculate the limiting magnitude by night using this paper:

https://arxiv.org/abs/1405.4209

which gives pessimistic/realistic estimates, and I get something around 6.2. Other estimation could be made as with the Bortle scale

https://en.wikipedia.org/wiki/Bortle_scale

which looks very optimistic, limiting magnitude could go up to 7.5. In fact, the main difference is due to the atmosphere which induces star scintillation, and may transiently boost the maximal magnitude visible by naked eye. Since we don't have star scintillation, I guess that is more realistic to use the result of the first paper, that took away this effect.

2) I have changed also the limiting magnitudes during astronomical twilight, based some ESO estimates of the sky brightness, and also during nautical twilight, down to -9 degrees for the sun position. The magnitudes I am finding there are also quite different from the ones we are currently using in stars.cxx. For brighter skies, the paper mentioned above is not applicable, and I've let the current settings as they were. Precise estimation there are not very relevant, only very bright stars are visible and they are not so numerous.

3) Our current star catalog is only a subset of the BSC, the BSC counts all stars up to magnitude 7.5. So I have changed our catalog by the original ones, containing all potentially visible stars, there are more than 9000 stars in total.

The result is closer to what I am used to in terms of number of objects, and the sky indeed contains thousand of stars. If I am not the only one interested in this, I'll be happy to submit a patch to the devel branch, but I have a few questions before:

a) Are we really limited in 500 stars for some reasons? There just dots in the sky, so I would say no and I have not observed any slow-down, but I may be missing something?

b) Do we really need the latin name of the star in the first column of catalog Astro/stars.gz? Most of the real stars do not have name, but are labeled by a number. I've no idea right now how the recover latin star names from astronomical professional catalogs that use HR numbers for instance, and that sounds like a pain... :)

I might dream in the future of a stellarium-like dome for the stars, dunno if someone has ever started such a project before?

Cheers,
Chris.
eatdirt
 
Posts: 381
Joined: Wed Aug 15, 2018 2:06 pm

Re: Would you want more stars in the night sky?

Postby bugman » Sat Jan 26, 2019 2:19 pm

Hi Chris,

I suggest you discuss this on the mailing list. I think that the star/moon/nighttime lighting situation is far from ideal as well. I worked on improving the moonlight in FG based on the phase of the moon a while back, and I still have a lot of ideas I'd like to get back to on this. Stars was not on that list though. I think that not only is the star number low, but we might be able significantly improve the night sky by using a high resolution bitmap image to show the milky way, the Large Magellanic Cloud, etc. Scintillation is an interesting idea to simulate :)

Regards,
Edward
bugman
Moderator
 
Posts: 1719
Joined: Thu Mar 19, 2015 9:01 am
Version: next

Re: Would you want more stars in the night sky?

Postby wlbragg » Sat Jan 26, 2019 5:47 pm

Yes Chris, I would be in favor of it.

The 500 limit may have been from early stages of development and isn't an issue anymore. If there is a performance hit, it could be made optional.
Kansas(2-27-15)/Ohio/Midwest scenery development.
KEQA (2-27-15), 3AU, KRCP Airport Layout
Intel i5 3570K AMDRX480
User avatar
wlbragg
 
Posts: 4991
Joined: Sat Aug 25, 2012 11:31 pm
Location: Kansas (Tornado Alley), USA
Callsign: WC2020
Version: next
OS: Win10/Linux/AMDRX480

Re: Would you want more stars in the night sky?

Postby wkitty42 » Sat Jan 26, 2019 6:40 pm

eatdirt wrote in Sat Jan 26, 2019 2:05 pm:Maybe the subject has been put forward long ago, but, as an amateur astronomer, I can clearly tell that the number of visible stars by night is not enough, and by quite a lot.

you don't like pi? ;)

Code: Select all
   18.44 [INFO]:astro        Loaded 3141 stars
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 5782
Joined: Fri Feb 20, 2015 3:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 14.04.5

Re: Would you want more stars in the night sky?

Postby eatdirt » Sat Jan 26, 2019 7:58 pm

you don't like pi? ;)


lol, yes I do actually. But that number is the size of stars.gz, which is a subset of the BSC, and most of these stars are actually not represented due to the magnitude 4 cut, the mag is the last number in this file. With a 6.1 cut, there are about 5000 stars to be plotted!
eatdirt
 
Posts: 381
Joined: Wed Aug 15, 2018 2:06 pm

Re: Would you want more stars in the night sky?

Postby www2 » Sat Jan 26, 2019 9:34 pm

This is only my 0.02€
I think0 that the 500 star limit is from the days that computer run with 300Mhz Pentium II cpu and a raise to 5000 is not so taxing on current systems.
www2
 
Posts: 260
Joined: Thu Apr 16, 2009 1:58 pm

Re: Would you want more stars in the night sky?

Postby wkitty42 » Sun Jan 27, 2019 5:35 pm

eatdirt wrote in Sat Jan 26, 2019 7:58 pm:
you don't like pi? ;)

lol, yes I do actually. But that number is the size of stars.gz,

well, damn... i guess the log message is in error since it doesn't say what it is really trying to say...

eatdirt wrote in Sat Jan 26, 2019 7:58 pm:which is a subset of the BSC, and most of these stars are actually not represented due to the magnitude 4 cut, the mag is the last number in this file. With a 6.1 cut, there are about 5000 stars to be plotted!

so what are we looking for? all the stars or just those that could realistically be seen from the ground and while flying? i can see where a lower magnitude would be desirable for space fliers and maybe mag 5 for commercial airline fliers... mag 7 would be the lowest i'd go but it would be nice if they were dynamic to reflect realism so that more stars can be loaded as you go higher and less as you go lower? and i guess the weather should affect what can be seen as well... i know that i see stars in the FG sky during the day and at times that i do/would not expect to see them...

this page should be of some help... http://www.icq.eps.harvard.edu/MagScale.html
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 5782
Joined: Fri Feb 20, 2015 3:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 14.04.5

Re: Would you want more stars in the night sky?

Postby eatdirt » Sun Jan 27, 2019 8:16 pm

just those that could realistically be seen from the ground and while flying?


Yep, this what the 6.1 magnitude is, the faintest stars you can see under a good dark sky (in space, you could certainly see a bit more, but that's another topic).

I'll try to post on the mailing list.
eatdirt
 
Posts: 381
Joined: Wed Aug 15, 2018 2:06 pm

Re: Would you want more stars in the night sky?

Postby wkitty42 » Sun Jan 27, 2019 9:21 pm

i saw both of your messages on the list... curt is the only one, so far, to comment there...
personally, i don't see a problem with raising the limit... i was just questioning/asking about realism... i'm 15 miles from everywhere... large populations are 50+ miles away... i know i can see ~4.5mag in my area but have to watch out for street lights... i don't know what a realistic mag number would be for high altitude because i've not been in a real plane at night in many years...
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 5782
Joined: Fri Feb 20, 2015 3:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 14.04.5

Re: Would you want more stars in the night sky?

Postby eatdirt » Mon Jan 28, 2019 10:03 pm

Ok, for those interested, I've opened a ticket with the code modifications to do:

https://sourceforge.net/p/flightgear/codetickets/2106/

Cheers,
Chris,
eatdirt
 
Posts: 381
Joined: Wed Aug 15, 2018 2:06 pm

Re: Would you want more stars in the night sky?

Postby wkitty42 » Tue Jan 29, 2019 2:18 am

OMG! it's full of starz!

seriously, this is nice! i just ran the patch and put the starz database in place and the first things i see when i start up over KRDU are what? Orion, Taurus, and the Pleiades... GoodJob!<tm>

now, where's the TLE code so i can load my 16000+ satellites into the sim and see them flying over? :mrgreen:
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 5782
Joined: Fri Feb 20, 2015 3:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 14.04.5

Re: Would you want more stars in the night sky?

Postby abassign » Tue Jan 29, 2019 8:38 am

I would like to remind you that, for a better simulation of the starry sky, it is necessary to tie the number of visible spels according to the brightness of the background of the sky. It makes no sense to see 6000 stars + the Milky Way or the Magellanic Clouds if you go low over a city ... I think the first priority is to manage the actual background brightness of the sky correlated to the urbanized areas, the presence of fog, clouds etc ... that could give a much more realistic working basis (maybe brightness modulated according to a slider as already done for example for the determination of the beginning of the snow) and then only after showing the stars actually visible . This also applies to clouds that from above should show a diffused brightness depending on the situation on the ground.
abassign
 
Posts: 824
Joined: Mon Feb 27, 2012 5:09 pm
Location: Italy (living 5 Km from airport LIME)
Callsign: I-BASSY
Version: 2018.3
OS: Linux Mint 19. x

Re: Would you want more stars in the night sky?

Postby Thorsten » Tue Jan 29, 2019 8:57 am

I would like to remind you that, for a better simulation of the starry sky, it is necessary to tie the number of visible spels according to the brightness of the background of the sky.


As I've written on the mailing list, that's a complete red herring - the brightness variation of the background sky is fairly irrelevant if the darkness adaption state of the eye is not computed - which in turn depends mostly on ambient light in the cockpit.

You won't see a single star even if you are in space and Earth is dark around you if you dial your screens to 'bright' in the cockpit of your spacecraft.
Thorsten
 
Posts: 11213
Joined: Mon Nov 02, 2009 8:33 am

Re: Would you want more stars in the night sky?

Postby eatdirt » Tue Jan 29, 2019 12:45 pm

Hi guys,

I did not intend to start a discussion to what and how we should simulate, but I would say that this is going far more than what I was proposing, just updating an old database to get the sky I am used to :)

A realistic simulation, for me, is 2 things:

- The environment, this is Physics.
- The perception, this is Biology.

Before modelling what we feel, I think it makes sense to model what is there. We can see stars up to mag 6-7, let's put them there. Of course, with a bulb in your face or the full Moon and the sun, that's not going to happen. But with a bulb in your face, your contrast vision is changed and many things should be changed as well, also in the way you would see the cockpit. If you stare at a red light for a while, your vision will be green for a while afterwards, etc...

I agree that all this could be simulated, and that would be utterly cool. Actually, for the ambient light, that could be possible. The paper I mentioned above contains biological factors fitting, so, in principle, you could imagine simulating retina sensitivity adjustment and pupil dilation response from ambient light to dynamically modify the magnitude threshold, could be even done according to line of sight... But such a model should not be implanted only for stars, it should be on top of many sceneries I guess.

Anyway, I am not enough competent to do this these days, but I'll be happy to help :)
eatdirt
 
Posts: 381
Joined: Wed Aug 15, 2018 2:06 pm

Re: Would you want more stars in the night sky?

Postby Thorsten » Tue Jan 29, 2019 12:58 pm

Before modelling what we feel, I think it makes sense to model what is there.


Sorry, but basically no. Pretty much any real-time rendering strategy is heavily biased towards rendering something like we see and utilizes visual illusions and cheats heavily to give you the impression that you see a real scene. Starting from 'what is really there' you could divide your framerate by a factor of 10 or more.
Thorsten
 
Posts: 11213
Joined: Mon Nov 02, 2009 8:33 am

Next

Return to Scenery

Who is online

Users browsing this forum: No registered users and 2 guests