the_bT

You know what, you're right. Kerbal rovers suck, there is no point in building a road if all your cars flip over 50km/h tops. How to fix that is easy though. To flip one side of the craft has to rise upwards. It can do that because wheels only bother to keep the craft off the surface, and not on it. The kerbal engineer thus concludes: we need to mount wheels not only upside up but also upside down. And for good measure upside left and upside right won't hurt either. The road needs some slight adjustments but that's quick work... check it out: I also figured out most of my troubles with calculating the positions. The problem was: I'm apparently stupid and can't keep radians and degrees apart. Also, the direction parameter in the cfg is not 0=north to 180=south, you have to subtract the longitude... or coincidentally 105° which is KSC's longitude, need to test a bit more... Keep an eye out for that if you put in your text field. My problems with height remain for the time being but it seems you can drive over such imperfections without issues. I also used box colliders for my track, I believe they are much more stable for driving on them.

Sleek looks is about the only thing that's going for this design. I tried to fix yaw stability problems for some person in the questions forum, and this was one of my more radical refactors of their original craft. The yaw instability was gone but not quite and the cargo bay is a bit to large for the max cargo weight (at least for my taste)... Still damn does it look nice! PS: clipping fuel into wing-roots is fair game

I really love how the article bashes computerized learning. Because it is right! It's right in saying that it is cumbersome and expensive and would in the end not work very well. BUT only in the way the author envisions it. The frustrating thing is: His vision is still actively chased. It seems to me few have really understood how interactive multimedia applications can be utilized to teach. You can't just copy the texts and images from the books, make some animated character say a few silly lines and call it a day. Well... you can and some do. It won't teach though, at least not more or less than the books did. The internet is irrelevant to basic education in my opinion. Such education requires nothing that could not be done offline. The only exception is education on the internet itself. Higher education on the other hand is almost unthinkable without access to internet nowadays. And why is that? One thing the article fails to predict is how the internet has sped up innovation. This high speed innovation has made a number of things possible that where unthinkable in '95. Google is an example of that, as are Facebook or Twitter. But the effect is not limited to things that would be useless offline. Consider Kerbal Space Program for example: KSP itself does not rely on the internet to function, but it would not been possible to make without it. And just by the way, it can be used to educate on the (in my opinion basic) principles behind planetary motions. In other words, the what moves around what question that so many adults get wrong so often... It also does a remarkable job of motivating people to get educated on less basic matters. That's where the internet comes back into the picture. What would you have done (offline) to answer this: forum.kerbalspaceprogram.com/threads/84342-Change-Inclination-before-or-after-I-make-my-orbit-smaller I double dare you to try that

What can I say, the back to back cockpits is genius by the strictest definition! I wish I could contribute something but the replica's I made are already in this thread (compare my swordfish, I think the nose is a tad to long )... Edit: Except there is one that isn't I think... Here goes: This is a replica of the not very well known because proposed but never actually built Blohm & Voss BV P.111 asymmetric flying boat. It's not a very nice replica, because it was really hard to balance and because it is way to curvy with all those bubble turrets It's stock + Firespitter prop engines. Reference Pic: Craft: As far as I remember it flew quite agreeable considering how it looks, landing in water was not a problem. Landing on land is impossible, as is taking off from water. Initial take off is accomplished with some sort of wheel assembly that can be decoupled. It might be possible to fix take off from water by abusing radial intakes. I discovered that they make the best floats a while after this was built... Edit2: Forgot to mention, all the crafts that I have posted here can be downloaded (along with a lot of other stuff) here.

Anti grav devices look like this (duh): Also, they work over water! (if you can propel yourself without solid ground at least )

Ya my original plan was to figure out how to do it with small segments, then write a script that takes 2 locations, finds a path between them where the slope is always > 1 and then generates a ton of instances to copy into the part.cfg. For something like that I definitely have to be able to calculate the positions accurately from lat/lon (as that's likely how the data about the slopes is going to be structured) and also place stuff at precise heights so it fits together... I don't think that's going to happen I was also wondering if it's a good idea at all, I don't know what happens if you throw a million instances at kerbtown, would be interesting to see

I went to explore Kerbin. I found this: I wonder if anyone has ever been to the same place in KSP before. After all the planet's surface has about 2.25 million km² dry land... It's almost a shame that so much of it goes to waste almost unused. Maybe a plugin that randomizes KSC position for every new game would be cool... I know there is one that can move the KSC. Hm...

I'm about to give up on this, it's just too frustrating The function that converts from the Lat/Lon input is a mystery to me. If I place an object somewhere, the GUI shows the vector position and lat/lon. If I copy lat/lon and manually calculate the vector the same way as the kerbtown sourcecode (allegedly) does, I get different numbers. If that wasn't funny enough, altitude kind of snaps to a grid or something. It's all fine if you use the buttons to move stuff up and down. But if you change the numbers directly to produce smaller movements nothing happens. Change them slightly more and the piece jumps to a new height, slightly off in the other direction. Maybe that's a limitation of the underlying KSP systems... I don't know. Clear is that it effectively prohibits using the system to place stuff that has to fit together exactly. Combining the elements to make a large road mesh is an option, but when you make large roads you need to take the curvature of the planet into account. Otherwise you have these enormous ramps at the end like the runway does. It might be possible to write a blender script that bends a mesh accordingly, but you would need to produce different models for use at different heights and it all gets way to complicated for it's value fast After all it's just a road, and rovers slightly suck don't they

That the cardinal axes are not aligned to the surface of the planet was pretty clear to me. I was just confused because I expected positioning on the surface to be 2 dimensional with latitude and longitude. The GUI actually allows placing objects that way so I went to look at how those inputs are converted. From that I found out that the 3d vector used for positioning likely works like this. It seems to be the normal vector of the surface which if you think about it defines a position on any strictly convex body. There are some strange magic constants in the conversion from latitude/longitude but I figured If i do my calculations in latitude/longitude and convert the same way the actual code does I should be fine. I also found this explanation on how to calculate a latitude/longitude pair you would arrive if you traveled a set distance in a compass heading. Currently I'm playing around with those formulas, they seem to be working well except for the heading formula. That has some issues with headings > 180 Ideally it would probably be better to avoid latitude and longitude completely and stay with the n-vector to avoid these issues. But I can't figure out a way to calculate the new position that way... Or even how I would represent direction in that system... So I will probably stay with what I already have and figure out how to fix the heading formula.

Due to my recent travels in trying to figure out what's going on with the 4 dimensional positioning, I came across this: https://github.com/HubsElectrical/KerbTown I'm an amateur with code so I don't know if this is the complete code or recent etc... Edit: Just saw, it's linked in the op too, but no idea if it contains the 0.23 fixes...

I was going to try regardless of it being dead or not, I was just curious if there were problems to look out for. I got a 0.23 fixed version from this post. It seems to be working. The only thing that confuses me a little is how objects are placed. First there are 3 axes + height to position things. That is one axis to much in my head. Secondly I don't really get where the info about placed objects goes. There is Instances data inside the part.cfg files, those contain all the data I enter when placing a part. Additionally the same date is inside the savegame in a file named KTInstances.cfg. When I change data directly in the part.cfg, stuff gets duplicated. I have not yet tried what happens if I delete the KTInstances.cfg or change it to match the part.cfg. I have the strong suspicion however that having different data inside the two files is a bad idea I'm looking into automatically placing stuff neatly aligned between two positions so I don't have to line up dozens of street pieces by hand. I hope it's possible at all. So whats what here, could anyone shed some light?

Thanks for reminding me about my education and employment status, survey!

... he said, defending the practice of shooting people in the face from a safe and secluded place I generally didn't mind the camping snipers when I last played Battlefield... they usually managed to run surprisingly quick when being shot at by a Spitfire There was this mod that put the original WW2 battlefield into modern times, middle east, helicopters an all that what shortly after became the modern shooter template and has been endlessly repeated since. There was a map with a bunch of high buildings in the middle, campsite #1 obviously, but if you buzzed it with a helicopter, a whole platoon of scared soldiers jumped from every opening in fear of being blown up by rockets... A sight to be had

What do you do when you want to fly an electric plane on Eve and have FAR installed? You make a plane that can fit into a 3.75m fairing, that's what. Once I realized that there was technically now law against asymmetric planes the whole thing got way easier. Will it fly on eve? There is only one way to find out I'm so not looking forward to rebuilding the whole thing with all hinges in the folded position . But if I just stick it into a fairing like this and fold it on the launch pad (should work I think) I can't strut it properly. Which is the kerbal equivalent of doom in a can. (Or fairing in this case.)

The electricity is then transmitted to some sort of appliance which turns it into heat + work of some sort. You can't get away from the heat. But that is not the driving factor behind global warming so I don't think it's of concern when discussing global warming. The point is that we need to save emissions because our problem is that heat delivered by the sun is kept by the slightly changed atmosphere instead of being radiated into space. It should not matter in what way the energy is delivered to the surface, the problem is the same I think. That being said, saving emissions is hard to do if you start by launching several thousand tons into orbit. How much electricity do we have to generate until we get that initial emission investment back? Considering maintenance, do we make emission profit at all?

Hm, I have trouble believing that any human made object can reflect enough additional sunlight onto the earth that it would make much of a difference. After all, isn't using a mirror to redirect light onto earth pretty much the same as enlarging the area that is hit by sunlight. Half the surface area of earth is in sunlight at any given time. Even if you pretend the earth was a flat disk and not a sphere you get an area of about 1.28*10^14 m². What kind of mirror can compare to that? Another thing: Regarding global warming, is it of concern how the energy is beamed down? In the end the energy will be released mostly as heat anyways, will it not?

@Crown: It doesn't seem to happen in my install. This is the craft I build with some cubes added so it's components are a bit clearer: It performed just as you described, engine and tank exploded on landing. I reverted to launch and everything was normal. My kOS install differs from "factory settings" only in that I removed "MM_Stock_Processor.cfg" from the kOS folder because it threw errors. Maybe you could try removing/renaming that file and see if that helps... Edit: I forgot to mention that I use the latest pre-release which kinda makes this all pointless probably

I'm going to try to do what you did and see what happens. Will report back.

Uh, quick question: Is kerbtown still alive? Or rather does it work in 0.23.5 without any or at least any serious issues. If there are minor issues, what would I have to watch out for? I'm considering making some generic road/runway/taxiway pieces and spicing up my KSC a bit...

I admit to being under-informed on quantum physics, but how does the ability to communicate information at such high speeds improve the precision of a clock? Even if its just the imprecision you get when a time synch signal is transmitted and received with a delay... That can be averted by using multiple signals from different sources like for example the GPS system does. So, can anyone enlighten me? Edit: or is it intended for communication and the term clock is just an allegory?

This is nonsense. There is truth to the idea but the statement is way to simplistic. The fact that the USA (or any country) is spending lent money doesn't imply that they can't afford to spend that money. Almost any modern business operates on money lend to it by investors, does that mean they cannot afford to operate? The problem is not that governments everywhere spend billions of dollars. The problem is they do not have to spend it wisely (and subsequently often don't). Space programs are probably among the wiser investments. In my opinion, paying someone to find out stuff about space and thus causing need for rockets and modern materials/robotics/computers is one of the better ways to create work. I would expect Research funding to increase in the future. After all, war has been proven to be inefficient to stimulate the economy in this age, and what else can you do? Repair crumbling bridges? Never going to happen.

Considering the fact that it is not so much a bug but rounding errors in the floating point variables used to calculate the orbit I'm not so sure it can be fixed completely. All that can be done is minimize the effect by forcing low timewarp (thus smaller values and smaller error in calculations) through SOI changes. It would basically be a light version of KAC implemented stock.

What parts you use on your vessel is irrelevant, use whatever you want. There seems to be an issue if you add the kOS part module to a part that already has a conflicting part module added to it. The issue is currently avoided by having a kOS part that contains only the kOS part module and nothing else. At least that's how I understand it. My KSP install is medium heavy on mods currently. The module manager cfg that comes with kOS caused errors (all of its entry's), so I deleted it. All it does is adding the kOS part module to all command pods and I decided I would not need that. Just out of curiosity though, could that be related to the conflicting part modules issue? I do not have mechJeb installed, are there other known conflicting plugins?

You can enable/disable Flags/Debris/Bases/etc. in a small menu that pops up top center if you point your mouse there in map view. The issues with editing maneuver nodes I have too and have not found a solution yet.

Gotta launch one yourself