Oskar-

I'm currently using 1.3.1.1891 LimuxPlayer en-us, and I recall getting a consistent segmentation fault quitting the game since 1.3 (Not on earlier versions though). I was hoping this would have been a usually error and someone reporting it and having it fixed. But after picking up playing again, I can't find any mention of this error or fixes for it on the forums. Anyone have some ideas on were to start with solving this? Some clues (short enough to add directly to the thread I'm guessing): ********************** ./KSP.x86_64 Set current directory to /home/oskar/ksp/KSP_linux_1.3.1 Found path: /home/oskar/ksp/KSP_linux_1.3.1/KSP.x86_64 Mono path[0] = '/home/oskar/ksp/KSP_linux_1.3.1/KSP_Data/Managed' Mono path[1] = '/home/oskar/ksp/KSP_linux_1.3.1/KSP_Data/Mono' Mono config path = '/home/oskar/ksp/KSP_linux_1.3.1/KSP_Data/Mono/etc' displaymanager : xrandr version warning. 1.5 client has 4 screens displaymanager screen (0)(DVI-I-1): 1920 x 1080 displaymanager screen (1)(DVI-D-0): 1920 x 1080 Using libudev for joystick management Importing game controller configs Segmentation fault (core dumped) ********************* dmesg gives a "KSP.x86_64[4894]: segfault at 7f01641b9be8 ip 00007f00d30c2874 sp 00007ffda783ecf0 error 4 in liblingoona.grammar.kerbal.so[7f00d2fe5000+1ca000]" ********************* First and last lines from KSP.log (doesn't seem like there is anything interesting in the log file): ******* Log Initiated for Kerbal Space Program - 1.3.1.1891 (LinuxPlayer) en-us ******* Kerbal Space Program - 1.3.1.1891 (LinuxPlayer) en-us OS: Linux 4.4 openSUSE project 42.2 64bit CPU: Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz (4) RAM: 7982 GPU: GeForce GTX 670/PCIe/SSE2 (2048MB) SM: 30 (OpenGL 4.5 [4.5.0 NVIDIA 384.98]) RT Formats: ARGB32, Depth, ARGBHalf, Shadowmap, RGB565, ARGB4444, ARGB1555, Default, ARGB2101010, DefaultHDR, ARGBFloat, RGFloat, RGHalf, RFloat, RHalf, R8 ... ... [LOG 01:55:32.147] [MessageSystem] OnAppInitialized [LOG 01:55:32.147] [MessageSystem] Reposition 0.08033587 10446 [LOG 01:57:58.175] [UIApp] OnDestroy: MessageSystem

(edit: added embedded images) This is one of mine, from way back in 0.19 http://imgur.com/urymqyH And with the six oranges in a 265km orbit http://imgur.com/JiI451T

I'll jump in here to just lay out the basics in a way that I figured out to help me understand why this occurs. That was actually for race car performance simulation stuff, but physics are of course the same for everything. Power = Energy / time Energy = Force * distance velocity = distance / time Force = mass * acceleration (now some of those variables represent physically measurable entities, and some are purely imaginary mathematical values, but that's not important here) So: P=E/t, and E=F*s, so: P=F*s/t, but v=s/t, so: P=Fv, but F=ma, so: P=mav Power = mass * acceleration * velocity So, if you produce some acceleration on some mass, then the faster that mass is going, the more power is used/produced/converted. It doesn't matter if it's a rocket, a volleyball, a race car, a proton in a particle accelerator, or a bullet. It doesn't matter if this acceleration is produced by throwing some mass in some direction, by a magnetic force, by gravity, or whatever. Rockets conveniently work by pushing on mass that is initially traveling at the same speed as the engine, so the acceleration is constant (for a given force and mass), regardless of v. So the end result is; if you burn at double the velocity, your engine produces twice as much power! (with the same amount of fuel used, with the same Isp, naturally)

Toroid planets seem to be physically viable, which basically means that there are bound to be lots of them out there somewhere. In the Kerbal system perhaps? One of those puppies having a moon with a really whacky orbit would be incredibly cool!