Project Orbiter Galaxy

[dumbo2007]

Ah ok now I get it, so the grid does define the cube in which my view is centered, and the stars that I see "not-above" the grid are actually from the surrounding cubes

Yes I understand about the pseudo-seamless thing. I am planning to do something like that for the Bump module terrain generation too, where I generate the terrain for surrounding rectangular patches too , apart from the one on which a rover is resting.


---------------------

Ok i turned off surrounding cubes and navigation is easier around the core

Hmm...I am still getting the crash though if I load an empty system. This time I tried to click the galaxy button after loading an empty system and Orbiter crashed.

Yeah I just went through the quickstart section ....have to RTFM soon

---------------

ok so I have been RTMing, and I have a question regarding this line in Section 4.3 Reaching another star the real way

Once the engines are cut, there are three possibilities:
either you have already crossed the distance limit specified in your Ogalaxy.cfg, in which case the new system will immediately be loaded and you'll find yourself some years later, in aproach to your targeted star, and should start to plan your injection burn.

So the new system is loaded by stopping the simulation and going to launchpad, after which the user clicks on the !OGSwitch ? Or does the new system get loaded without the launchpad intervening at all, which would be awesomeness !

 

[jedidia]

So the new system is loaded by stopping the simulation and going to launchpad, after which the user clicks on the !OGSwitch ? Or does the new system get loaded without the launchpad intervening at all, which would be awesomeness !

Quitting Orbiter is necessary, since there is no way to dynamically load a system at runtime. However, you should not have to restart Orbiter manually, it should do that by itself. I take it that is not the case for you, then.

I have a hunch on why that might be: In the launchpad, go to Extras->debugging options->Orbiter restart options. I guess this currently set to de-allocate memory (Orbiter default setup). Set it to "Respawn Orbiter process" and Orbiter should re-launch with the correct scenario by itself when switching systems.

Hmm...I am still getting the crash though if I load an empty system. This time I tried to click the galaxy button after loading an empty system and Orbiter crashed.

That IS strange. Can you reproduce it in the standalone program?

I am planning to do something like that for the Bump module terrain generation too, where I generate the terrain for surrounding rectangular patches too

Depending on the size of the patches, I think you won't get around multi-threading when you want to have this in the actual game.

 

[donatelo200]

Hey ben awhile since i've used orbiter but anyway i've noticed that the clouds on the planets all seem to be inside the planet itself instead of slightly outside....... I figured out the cause. The cloud resolution is very low making the edges sharp and bringing most of the clouds below the sphere.

 

[Columbia42]

The new release is great! Texture generation is much faster, better textures, no more black textures, lots of bugs fixed, just generally awesome! Thanks for a truly revolutionary orbiter addon.

 

[jedidia]

The cloud resolution is very low making the edges sharp and bringing most of the clouds below the sphere.

Thanks for pointing out, I'll have a look at it. It's possible that I fixed the clouds to a low resolution for some tests and forgot to take it out again.

One thing's strange however: I know that terrestrial planets have their clouds rendered properly, is that the case for you too?

---------- Post added at 01:11 PM ---------- Previous post was at 10:39 AM ----------

Bingo! Apparently I fixed the texture level for clouds to 4 sometime in the past for some tests, but only in the standalone. So I didn't notice the bug because I was majorly working with the plugin, which exports clouds just fine (or at least they should. Can you confirm that you were using the standalone program to export those systems?)

Thanks for the heads up, will be fixed. Artlav also provided me with an update of the generator shortly after release that is even faster than the current one and also can export clouds via GPU (they're currently exported on CPU no matter your settings), so this might just be the excuse I need to upload a first patch, but I'd like to wait for a confirmation of the crash Dumbo2007 reported. Does anyone else expierience a crash when they open a system without planets in the system overview and then push "next" or "previous system" or the galaxy map button? (you can try with Sirius, right there in the Sol-cluster).

If not, I have no idea what that's about. If yes, I'm probably having some undefined behaviour going on somewhere...

 

[Artlav]

Artlav also provided me with an update of the generator shortly after release that is even faster than the current one and also can export clouds via GPU (they're currently exported on CPU no matter your settings), so this might just be the excuse I need to upload a first patch,

If you haven't found any bugs with the new generator, then the last version should be release-worthy.
Having clouds do "blink, Lv8" instead of "vvVVRROOOOOOOOOOORRRRVvv, Lv4" would be a noticeable advantage.
Otherwise, i'll be gone from 26th and well into january, so list the bugs ASAP.

 

[dumbo2007]

Ok, so I am trying to pin down the exact steps which may lead to the crash.

So I am starting with a clean install of Orbiter Galaxy(Only ran the new Orbiter Galaxy with it a few times now):

1. I open the DG Mk4 in orbit scenario, press F1, inside the DG, load Insterstellar Nav MFD, click Map and I get the stellar catalog window.
2. Click Orbiter window Orbiter, press ESC, then back to Orbiter Galaxy window.
3. So I change the view around for a bit and spot Altair and focus on it, and start generating the textures for it.
4. It takes a while generating 23 textures varying between level 1 to 4(default settings).
5. After its done I am back in the planetary details window of Altair.
6. Then I click the >>> button and I can see the Del Pav system which is the next system after Altair. :thumbup:
7. I press ESC to get back to the stellar window(I am still zoomed in on Altair)
8. I change the camera view around a bit and then I drag my mouse to the right side to roll out the search box.
9. I type in Sirius, I get the focus there, double click Sirius when the mouse hovers close to it and I can see the blue disc showing its plane.
10. I am now in the empty planetary details window. I press >>> and CTD.
    
    By the way is it possible to model binary star systems yet with the lights that Orbiter now provides. Maybe have another very bright light source in a vessel to act as the 2nd star.
    In reference to this thread : http://orbiter-forum.com/showthread.php?t=21345&highlight=binary+star+system

 

[jedidia]

"blink, Lv8" instead of "vvVVRROOOOOOOOOOORRRRVvv, Lv4"

:rofl:

Otherwise, i'll be gone from 26th and well into january, so list the bugs ASAP.

None I would have noticed. I agree that the latest version you sent me is release worthy, I just wanted to fix a few more bugs on my end before putting up the first patch. Thanks, and enjoy your vacation (assuming you're away for vacation, that is...)

Ok, so I am trying to pin down the exact steps which may lead to the crash.

Thanks a lot. So the crash does not happen if you don't export a system first? Doesn't make any sense at all to me currently, but I'll get to the bottom of it!

 

[dumbo2007]

So the crash does not happen if you don't export a system first

Well I haven't got the crash yet without exporting.

 

[jedidia]

Crash confirmed on multiple systems that have no planets if the next system button has been used before . Not sure yet what's going on here, but as far as I can tell exporting or not exporting has nothing to do with it. Which is fine, because that would have been very confusing.

Alas the crash happens somewhere outside my code, i.e. most probably a forbidden memory access under the circumstances, but since VS can't provide me a debug point inside my code I'll have to check a multitude of pointers to get to the heart of the matter.
On a side note, if anyone was wondering why some labels suddenly apear in green after a search for a star, that's a bug too. It seems that Orbiter Galaxy is calculating jumproutes when loading a cube even though Jumppoints aren't activated yet. Will be fixed too.

---------- Post added at 05:16 PM ---------- Previous post was at 05:02 PM ----------

More Trouble: Crash seems to only occur in the plugin, not the standalone. Can you confirm that?

---------- Post added at 05:57 PM ---------- Previous post was at 05:16 PM ----------

Never mind, sought and destroyed. Patch is up!

 

[dumbo2007]

Wow ...you made it before Christmas :thumbup:

------------------------

Ok I can confirm the bug is resolved . All this virtual star gazing has got me really interested in actual star gazing again.
Was out at the terrace tonight wielding my laptop, trying to spot some of the stars mentioned in the System Overview screen(luckily it doesn't freeze here)
and clear winter skies. :thumbup:

 

[jedidia]

By the way is it possible to model binary star systems yet with the lights that Orbiter now provides. Maybe have another very bright light source in a vessel to act as the 2nd star.
In reference to this thread : http://orbiter-forum.com/showthread....ry+star+system

Sorry, didn't see this. Local lightsources won't work, as local lightsources won't light up planet surfaces. Only the global lightsource can do that, and there can be only one (currently).

That said, adapting stargen to produce halfway believable close binaries would be a daunting task. Something I'm willing to take on at some point in the plausible mid-future when most other work is done and honing the generator is the only thing left, but only if the orbiter architecture at that point will allow me to implement it without a hack (or at least with a hack that works very reliable). Otherwise it's just not worth the effort.

 

[Linguofreak]

Nitpick on the 6.3 patch: The capitalization on the directory names is different in the patch than in the 6.2 archive and the Orbiter2010 distribution. This is slightly problematic for those of us running on Wine, as Linux filenames are case sensitive (so we have to manually merge the directories).

The capitalization in the 6.2 archive seems to be fine, though.

 

[jedidia]

oh horrors! Yes, I have not been too observant in that area. I'll put some more attention to it next time, sorry.

 

[Columbia42]

Is there any chance that atmospheres for gas giants will be implemented in the future?

 

[jedidia]

Is there any chance that atmospheres for gas giants will be implemented in the future?

That depends on what you mean by atmosphere. If you mean an outer layer for aerobraking like in the planets of the solar system add-on, then yes, that is somewhere on the list...

 

[donatelo200]

Why not the whole atmosphere.....In other word why not creat the atmosphere so that it is simulated all the way down the the liquid layer. I have with my Ups And system and HD somethn system.

 

[Columbia42]

That would be so cool! I'm downloading your system right now to try it out.

 

[jedidia]

Why not the whole atmosphere.....In other word why not creat the atmosphere so that it is simulated all the way down the the liquid layer. I have with my Ups And system and HD somethn system.

Could you shortly explain how you did that? I' guess I'll still have to download the system and take a look at it, but a preliminary description might help me figure out if it's currently an option or if there's too much trouble involved.

 

[donatelo200]

Well in one of the books i read Jupiter has pressure of a million atmospheres at the liquid layer. I then inffered this would be true for all the gas giants so the preassure is insaign and for the liquid layer i just created a dull blue texture due to rayleigh scattering. As for the clouds i just created a cloud layer for the gas giant apperence. Neptunian planets are similar but the pressure i read is significantly smaller at a mer 100,000 atmos.......P.S. i need to get back to work to the Ups And system