Linux playground

Hi @Gondos ,
I keep testing Orbiter and the bugs I find are getting more and more subtle, which is great news.

Here's what I found: HSI MFD not working, compass is completely black.
In Align Planes MFD the text overlaps a bit.
https://drive.google.com/file/d/1acgGHrHiiPXROdFcX5Mo04jxgZOiNtFl/view?usp=sharing

And the most serious bug, because it produces a segfault:
Unable to open VOR markers in Orbiter's standalone MapMFD. Produces a segfault.
https://drive.google.com/file/d/1rCcvjDKDoMgQ9uSJsXZSxjeprLf5anC5/view?usp=sharing
 

Attachments

Hi @Gondos ,
I keep testing Orbiter and the bugs I find are getting more and more subtle, which is great news.

Here's what I found: HSI MFD not working, compass is completely black.
In Align Planes MFD the text overlaps a bit.
https://drive.google.com/file/d/1acgGHrHiiPXROdFcX5Mo04jxgZOiNtFl/view?usp=sharing

And the most serious bug, because it produces a segfault:
Unable to open VOR markers in Orbiter's standalone MapMFD. Produces a segfault.
https://drive.google.com/file/d/1rCcvjDKDoMgQ9uSJsXZSxjeprLf5anC5/view?usp=sharing
Thanks for the feedback :)
The crash was due to an unimplemented feature in the OGL sketchpad : drawing pixels, this is now implemented
The HSI bug was caused by a mishandling of the origin in the OGL sketchpad
These two are fixed along with a fix for the day/light projection on the mapmfd.
The overlapping issue is more problematic, I don't have it on my end but comparing your video with my output, I see that fontconfig does not use the same font on our different systems. Not sure what to do with that one, the surest way to ensure identical results would be to ship the ttf with the binary but I'm not sure it's the best solution...
I'm not pushing yet because there are still some visual issues with nanovg...
fixes.png
 
Quick question : I'm playing with the better ISS mod, is it supposed to look like this ? (I mean the violet panels really look out of place)
iss_hd.png
 
Yes, the panels look correct. At least, that's the way I did the textures, do I have 'em to brightly violet?

The thing I noticed is this: That part of the mesh is hidden by a meshgroup flag. Are flags not implemented yet?
1671051118437.png
 
Yes, the panels look correct. At least, that's the way I did the textures, do I have 'em to brightly violet?

The thing I noticed is this: That part of the mesh is hidden by a meshgroup flag. Are flags not implemented yet?
View attachment 31700
The author himself :hailprobe:
I'll look into the flags handling to see where the bug is lurking...
BTW, do you have a github account to fork from?
I had to remove the SSRMS from the scenario because I don't have a module for it. Is it the same as the one provided with SSU?
Regarding the canadarm2, is there an opensource module for it?
 
BTW, do you have a github account to fork from?
Unfortunately, I have a very unreliable internet connection so I keep everything local on my machine.
I do have a GitHub account, so I can create a repo of Better ISS for you to fork if that would be helpful? :)
If so, what is your GitHub username?

I had to remove the SSRMS from the scenario because I don't have a module for it. Is it the same as the one provided with SSU?
Regarding the canadarm2, is there an opensource module for it?
It's a standalone addon for o2010, no connection to SSU (that I know of!) Then the DLL patch makes it 2016 compatible.
Unfortunately no open source version exists that I know of. :( I would reach out to @GLS or @SiameseCat (not online since 2016) to see if they have the code.
If an open source SSRMS exists for SSU, maybe that could be used?
 
Unfortunately, I have a very unreliable internet connection so I keep everything local on my machine.
I do have a GitHub account, so I can create a repo of Better ISS for you to fork if that would be helpful? :)
If so, what is your GitHub username?
There's a link in my signature ;)
vvvvvvvv
 
Unfortunately, I have a very unreliable internet connection so I keep everything local on my machine.
I do have a GitHub account, so I can create a repo of Better ISS for you to fork if that would be helpful? :)
If so, what is your GitHub username?


It's a standalone addon for o2010, no connection to SSU (that I know of!) Then the DLL patch makes it 2016 compatible.
Unfortunately no open source version exists that I know of. :( I would reach out to @GLS or @SiameseCat (not online since 2016) to see if they have the code.
If an open source SSRMS exists for SSU, maybe that could be used?
While you're at it, there's also this puppy ;)
iss-de.png
 
Added the bigger brother while I'm at it :
ion6.png
Tried to work on SSU to get the SSRMS to compile but it looks like the latest version won't compile. I got a snapshot of the r2817, but some code tries to access member fields that don't exist... And now sourceforge is crapping up on me, I can't access the project anymore to try another version 😫
 
For what it's worth, here's the latest SSRMS sources in the SSU SVN repository here on O-F, revision 2820 dated May 1 2018.
 

Attachments

For what it's worth, here's the latest SSRMS sources in the SSU SVN repository here on O-F, revision 2820 dated May 1 2018.
Thanks, but the issue is with the libUltra. I managed to cook up something with copy/pasting stuff from SSV anyway :
ssrms.png
Now I need to set up a dialog to control it...
 
Added the bigger brother while I'm at it :
View attachment 31719
Tried to work on SSU to get the SSRMS to compile but it looks like the latest version won't compile. I got a snapshot of the r2817, but some code tries to access member fields that don't exist... And now sourceforge is crapping up on me, I can't access the project anymore to try another version 😫
Regarding Ion-6: (Actually LC-39B)
Do context sensitive bases work yet? I noticed that the pad is the default Orbiter one, not the “bare pad“ configuration from the Ion-6 addon.
Regardless, great work! Just out of curiosity, how much of the functionality of the original (Windows x86) DarkEnergy project is there in Linux?
 
Regarding Ion-6: (Actually LC-39B)
Do context sensitive bases work yet? I noticed that the pad is the default Orbiter one, not the “bare pad“ configuration from the Ion-6 addon.
Regardless, great work! Just out of curiosity, how much of the functionality of the original (Windows x86) DarkEnergy project is there in Linux?
Thanks :)
I'm not sure how context sensitive bases work so if it's not handled by the Orbiter core then it'll probably not work on the current version.
Apart from bugs, the main limitation is the custom cameras that are not available in the OpenGL client.
I had to port to XRSound, switch some GDI stuff to sketchpad, mix in some windows type removal and fix some arithmetic exceptions on the way but it was quite straightforward.
You can see the diff here : https://github.com/maxq519/DarkEner.../main...TheGondos:DarkEnergy-Spaceplane:linux
 
Back
Top