*Aqua*

Can you reproduce it? Is it always after 28 seconds? Just in case check your Windows event logs if there's something happening in that time.

Or you add another mod: Active Texture Management It compresses all textures on start up of the game and therefore reduces the memory usage by a lot.

Relevant part of the log: Just to make it clear: 1. You start the editor. 2. You add a MK2 Cockpit. 3. You delete it? 4. You add a LFO fuselage. 5. Game freezes. 6. You add a MonoProp fuselage. Somehow it doesn't make sense.

Ok, the problem is the ESC menu not saving when you want it to. Does your language include letters which are not in the English alphabet (read Claw's post)? Is your KSP install in C:\Program Files\? If yes please copy & paste it somewhere else. The Program Files directory is secured by Windows which can cause problems similar to yours. And we still need your log files.

In KSP when a building blows up the object of the intact building is swapped with the destroyed one. (Look closely, you can see that.) Unfortunately prints, tracks and marks require another technology than the destructible buildings. But! Unity most likely already has what you need. In fact you usually add a transparent ground texture in which you draw the prints, marks, etc. and display that. The marks will be flat and a giant texture or several small one will be needed. It's pretty common. Another method is to change the ground mesh. A crater will then be a real crater you can fall in and never get out again. This requires a lot of work to implement! Only a few games do that.

I think that level of commitment requires some sort of world goverment (union, federation, etc.). If there's no world goverment then everything will get complicated quickly. For example there is no nation's law which applies to Mars. If somebody commited murder on Mars which nation's law will be applied? The one the murderer comes from? The one depending on the country which build the module the crime scene is in? The country which shoot the astronaut into space? (What about an US astronaut who fly from French Guyana to Mars and murdered an Indian in a Chinese habitation module?) Or a less extreme example: Colonists will probably have babys. What do you say about the nationality of the newborn? (Italian/European Mother & Russian father living in an Australian module) A Mars colony will depend on Earth a long time, I guess a Marsian state with its own law won't form during that time. I don't know what they will do.

Please read this. It helps to quickly understand what went wrong.

Afaik is gpu passthrough still an experimental technology in VMware products. XenServer seems to have at lot of problem, too. Hm... it should be working with KVM but I won't touch that again in the near future. I still remember the nights spend bypassing a bug in the network code. *~*

Polywell is interesting. I didn't hear about that before. Does it really work? There seems to be a lot of discussion about how strong the bremsstrahlung will be.

How to obtain the logs is explained here: http://forum.kerbalspaceprogram.com/threads/92230-Stock-Support-Bug-Reporting-Guide Note that you can only load custom named saves in the ESC menu in the space center view *after* loading the general save. Custom save names don't change the save names in the main menu. Think of it like that: Load a save game in the main menu is like loading a universe. Once the universe is loaded you can then load an alternative timeline in the ESC menu. ...yeah I know it's strange.

I'm not sure if gpu pass-through really exists but it makes sense for gpu clusters for number crunching. 3 fps on the menu! Congratulations!

You didn't understand my post or I couldn't express myself good enough. English isn't my native language. If scientist can't build a fusion reactor which can generate electricity (or other form of harvestable energy) for more than a few milliseconds then there is no way Lockheed can do that 1/10 (or smaller) scale. That's simply not possible at the moment. Most projects about fusion reactors center around the problem how to ensure continous fusion for at least a few minutes and how to extract the energy. Thinking of a "smaller" reactor design is something we can (and probably will) think about after.

The real name is "Kukgaujugaebalkuk" (no I can't pronounce that) which translates to English national aerospace development administration or NADA. And there are how many ways to say the same thing with different English words? NASA, ESA, RFSA (commonly known as Roskosmos), it is always the same, so NADA is acceptable. And don't laugh at them. Whenever someone makes things wrong, a hard punishment will follow. (And no, I don't mean firing that guy.) The scientists and technicans only act on behalf of the goverment, they can see themselves there are bigger problems than flying rockets into space. But what can they do? If they refuse to work, they or a member of their family disappears. And some of you guys think that's funny?

People aren't able to set up a big working fusion reactor. There is obviously no way Lockheed can build a small one that's working and produce enough energy.

There is a reason why planes look like planes.

Once again: If you somehow manage to get KSP going in a VM, it will be horribly slow. Expect way less than 5 fps and major graphical glitches. Consumer hardware does not provide features for VM speed ups and gpu pass-through. Only expensive server hardware can do that but I don't think you'll want to pay the price of a car for that.

If there aren't any DR related errors in the log files, everything went fine. The stock configs are very forgiving. Alt + D + R opens up a small config window*. There you can change the values DR uses for calculating reentry. For example a shockwave exponent of 1.12 will incinerate nearly everything. Experiment with the values until you find a setting you like. * Alt + WASD are trim commands. You pod will start to spin. Remove the trim setting by pressing Alt + X afterwards.

Running KSP in a VM? Forget about that. Even if you are somehow able to get it working, performance will be awful.

Beta! OMG! Thanks for putting up the detailed FAQ! It answered almost all questions I had. (I'll hold back the unanswered ones until the appropiate time has come. )

In other words: "We spent 60 years but we still don't know how to do it. Help us!"I look forward to Wendelstein 7-X. They finished building the stellarator. Next year they will start the plasma tests. The goal of Wendelstein 7-X is to determine how an ongoing fusion (for years) can be archieved. It won't be able to generate a surplus of energy, it's too small for that.

I had a few courses about AI while studying computer science. It would take to know how to efficently solve optimization problems. That's all AI is about. And that's the reason there why nobody succeeded in creating a good AI. And what kind of AI do you want to have? A weak one? This one can only solve simple problems which solving algorithms are coded by a programmer before. It can't react to problems the programmer didn't think of. A strong one? This one can solve all possible problems except one: it needs massive amounts of processing power. We are talking about super computers here. First learn about logic (classical logic, propositional calculus, predicate logic and may be lambda calculus), then knowledge-based systems ("rule-based system"), then learn about ontologies (Bayesian network, what is information?, concepts, relations, etc. basically how to structure parts of information in a meaningful way and how to process them to learn new things) and after that learn about "advanced" AIs (neuronal networks, evolutionary algorithms, data mining, etc.) and finally it is good to learn about theoretical computer sciences (automata theory, Chomsky hierarchy, Turing machines, P and NP complexity classes). *gasp* This is heavy stuff! When you know all about these you'll have a good understanding what (and how) an AI can do (stuff). AI is such a difficult topic that there was almost no progress the last 30(!) years. Today our so-called AIs pretty much work the same way they did back then. Ok, sematic webs evolved a bit because of the internet. But essentially that's it.

I don't have this mouse. What's the problem?

I'm sorry I didn't explain well.Gravitational forces of the two moons will pull the elevator. It may not fall immediatly to the ground (I guess it can withstand some pull) but every pass of the moons will pull it more out of its ideal place. It may even swing back and forth a few times but it will eventually crumble. And there is the atmosphere, too. Winds will pull the lift. Ah that's what you mean. Unfortunately it won't work exactly as you imagine as I explained above. The lift will move and you have to give it the space to move. Fixing it on the planet surface will only result extreme forces building up which will rip everything apart.The current idea of a space elevator is to fix one end on a movable platform (i. e. a [giant] ship in the ocean). The CoM of the tether will have to be at the geostational orbit and rest of the tether extends further out. At the end there is a movable counter weight for the elevator cabin. As the elevator moves up, the weight comes down and vice versa cancelling out the forces. So in the first step the freight gets to geostationary orbit height. In the second step the freight will be moved to the counter weight. As it's goes to further out the elevator cabin gets down to the planetary surface. The counter weights movement capability can also be used to balance the lift in case it starts to swing. This design is more fexibile and than the pure "centrifugal design". @Error404brain It's easier to live in space. 1) Mars has a thin atmosphere. It can't protect us like our atmosphere does. In fact Mars is clustered by craters! And we have to care about gravity as Mars has only 1/3 g which isn't quite enough for humans. It's easier to build a rotating space station and evade meteorites. 2) "Near" is relative. The Moon is near, Mars isn't. 3) We don't need to transform asteroids into space stations. We just build one in Earth's orbit, fly it to the asteroid and process it. Processing will be easy as we almost don't need to care about weights. On an asteroid an astronaut can lift tons of material with one hand. 4) There is (almost) no gravity on an asteroid and we don't have to think about to survive an atmospheric entry and how to get out of it again. Btw, who want's to terraform an asteroid?

It is miniscule yes. The lift won't fall down over night but the stress will accumulate and one day it will rip apart. I guess it will take a few years. It's not? Then what keeps it in place?

NaN = not a number Not all values of a floating point number are interpreted as a number to indicated special circumstances. 0/0 -> NaN ∞ -> NaN √-1 -> NaN etc.