vyznev

You could slightly abuse KSP's simplified aero model by clipping a bunch of extra draggy parts (like sideways fins) inside your probe. Maybe even use them as structural / decorative elements.

OTRAG

Stick a reaction wheel on there and it'll go sideways too just fine. (Well, OK, not exactly fine. It's kinda janky to control because the reaction wheel keeps "slipping" and letting the craft rotate by a few degrees every few seconds so you have to keep correcting the attitude. That might be a bug, but KSP reaction wheels work by pure magic anyway so who knows. But it flies and is basically controllable enough to land anywhere you want.)

Technically, the original version of my BYOA had only three parts — a command chair, a sepratron and a cubic strut (to serve as the root part for attaching the other two) — and used no exploits. But it was basically uncontrollable after chute deployment, so the only way to land it without breaking any of the three parts was by pure luck. (I did realize that I could trim the six-part version I posted earlier down to just five parts, since the reaction wheel can serve as the root part, making the strut superfluous. And technically I could even get rid of the grip pad too, if I'm allowed to break one of the four remaining parts on landing as the OP mentioned above. Or I just need to learn how to land more gently — hitting the ground at less than 8 m/s to avoid any parts breaking is definitely possible, just hard.)

LF-only career definitely seems doable. I'd say something like this general plan should work: Grab enough science from the launchpad (and/or the runway) to unlock the thermometer and the barometer. (FWIW, you can just get the required 20 science in a single launch of a capsule with two goo canisters if you make sure to get a crew report, an EVA report both on the ground and while jumping, and grab the data from one goo canister while on EVA and recover the kerbal and the capsule separately.) Collect EVA reports and temperature and pressure samples from enough KSC mini-biomes to unlock the Aviation node (which has the Juno engine, air intakes, LF tanks and plane wheels). Normally you'd do this with a science roller, but for this challenge I'd consider that a form of electric propulsion, and thus not allowed (since we have no permitted way of generating electric charge from LF yet). But in KSP 1.11 you can instead send two kerbals — a scientist and an engineer — on an EVA walk carrying the science instruments in their inventory. In each biome, the engineer will place the instruments on the ground and the scientist will run the experiments and collect the data. Remember to get an EVA report from each biome too. (One kerbal can collect a report from the other, allowing multiple EVA reports without returning to the vessel.) Once you have the Juno and some wheels, build a jet-powered science car and get more science (crew reports, goo samples, any EVA reports and temperature and pressure data you didn't get before) from KSC. This'll let you unlock the Science Jr. and a bunch more plane parts. Launch a plane to get a bunch of milestones and unlock survey and part testing contracts. Grind those for cash while collecting science from different biomes on Kerbin. Also remember to get materials science from all the KSC biomes. Eventually you should hopefully have enough science to unlock the nuke engine. Then you can get to orbit and continue from there. In fact, I've already done this up to the "launch a plane" part, so I have proof that the "KSC walkabout" strategy works. The only real issue (besides the general slowness and tedium, which physics warp helps with somewhat) that I've noticed is that getting temperature and pressure readings from most of the building sub-biomes (like "VAB Main Building" etc.) doesn't work because the collision mesh is glitchy and the instruments just fall through and disappear. But fortunately you can get more than enough science just from the biomes that do work. (Also, if you quicksave and reload while a kerbal is standing right next to the VAB building, they can apparently teleport to the VAB roof. This would've been more useful if I'd remembered to level them up so that they could actually use their parachute to get safely down again. But at least it did give me a milestone achievement for exceeding a velocity of 80 m/s. Considering that kerbals normally waddle at less than 2 m/s, that's pretty impressive. )

Does this count as a valid craft? This was just a test run to see how far I could get with two sepratrons, and I forgot to take a screenshot of the F3 screen so I only have the flight time (about 7:40; no video, so I might be off by a second or two there too). Anyway, I'm sure this plane can do a lot better with a few more engines. As you can see from the screenshot, it does take off without SAS, although control is very limited without it due to the lack of a tail fin. And it does sort of take off horizontally, although it must pitch up pretty quickly since it only has grip pads for landing gear.

I thought this seemed vaguely familiar… (It's not quite the same challenge, since that one only allowed one sepratron per craft, but all entries there should qualify for this challenge too. Modulo possible physics changes in three years, at least.)

Does a Kerbal with a parachute count as a lift-generating part? If yes, let me present… KerbalX link: https://kerbalx.com/vyznev/BYOA-Rocket-Paraglider

I developed some tech a while ago that might be useful here. If it works for docking in orbit, it should work for a Gilly landing. Those are basically the same thing anyway.

https://wiki.kerbalspaceprogram.com/wiki/Tutorial:Sharing_craft_files You can also give https://kerbalx.com/CraftManager a try. I don't use it myself, though, so I can't give any personal opinion on it.

Something like this. Unfortunately, it seems that KSP 1.11 kind of broke that bike. I'm still hoping to find some way to fix it, but I haven't had much time to play with KSP lately. The major issue seems to be that rotor torque was nerfed, but that's easy enough to fix in the SPH by setting the rear wheel torque to 100%. But it seems that there was also some other change that somehow made the bike a lot less stable. It used to run nice and straight on flat ground in 1.10, with only minor and occasional course adjustments needed on hilly terrain. I could literally leave it running while I went to get a cup of coffee. But now the bike veers constantly to one side and is almost impossible to keep going straight for even a few seconds, making it only barely driveable.

Hmm, I think I just found another loophole in your rules that basically allows reducing the refurbishment penalty as low as you want, making big SSTOs competitive after all. But it's kind of a tedious one to exploit, so I'll just suggest a fix: require that all tourists must land at the target in the same vessel at the same time. (While you're at it, consider also disallowing ore mining and conversion on Kerbin. That's a separate cheesy exploit, but it happens to synergize well with this one.) Otherwise, one could just:

Either that, or maybe a reaction wheel attached to the swashplate could work. I haven't tested it yet, though. So basically like a real helicopter, then, only with fewer parts.

You mean like with an actual swash plate built using robotics parts and same-vessel interactions, as opposed to just enabling built-in cyclic control on the rotor blades? That sounds like a potentially interesting engineering challenge. Also probably liable to summon the kraken if the RPM gets too high. I guess the real challenge would be making it work with SAS...

It does rather encourage air launching, though. Alas, the way you've phrased the updated rules, I guess my original plan of circularizing at 69 km before decoupling and landing the first stage is not going to work. At least not unless you'll buy the argument that Rapiers are still jet engines even if used in closed-cycle mode, which I certainly wouldn't buy myself. But I did manage to come up with an alternative cost-saving trick that should work under the rules as written, and might even be more efficient. Maybe. Not sure how much I should spoil it yet, since I might actually have time to fly a test mission. But let me just note that drop tanks also don't have rocket engines. And that there's actually quite a lot of things besides just tanks that you can toss on a return trajectory to Kerbin once you no longer need them. (Alternatively, and just staying firmly within the presumably intended spirit of the rules, I suppose ballistic air launch should work. Basically, go as high and fast as you can with air-breathers, point your nose up at a suitable angle, fire up the rockets on the next stage and decouple it off without switching to it. Just make sure it's aerodynamically stable. By the time it leaves physics range, it should hopefully have enough velocity that you can quickly land the first stage before the second stage has reached apoapsis. And if you get the launch angle right, you might even end up in a fairly reasonable orbit after circularization.)

Well, since this thread has been resurrected, let me just drop in a fitting picture from an old post. So, uh, lessons learned: 1) command chairs can be used to force kerbals (or at least their heads) through holes they wouldn't normally fit through, and 2) judging by the facial expressions, leaving kerbals hanging from their helmet in mid-air apparently makes them kind of uncomfortable.

I assume decoupling the engines after landing and recovering them separately would be cheating? (I think something like "if your craft has ever left Kerbin's atmosphere" might be less prone to exploits and weird edge cases. Although it would allow full cost recovery for boosters if you decouple them low enough. Maybe a combination of the two criteria — i.e. "has no rockets, nukes or ions and has never left the atmosphere" — could work?)

I haven't had much time to play KSP lately, but I thought I'd submit an old craft I made last year. This one's obviously for category 6. In particular, whereas my last entry had way too many landing gears, this one has none. https://kerbalx.com/vyznev/Clonespeeder What it does have, instead of landing gears, is Vernor RCS thrusters. This makes takeoff and landing a bit challenging, but hopefully by episode 6 you'll have had enough practice to pull it off. The recommended procedures are: With both landing methods, the main difficulty is that this craft has only a limited amount of oxidizer and the Vernors guzzle it up like crazy. Thus, the maximum hover time is very limited. If you prefer easy mode, turn on infinite propellant in the Alt+F12 cheats menu and you can hover for as long as you like. Other than the tricky takeoff (and trickier landing), this craft is quite fun and easy to fly. It's not very aerodynamically stable on the pitch axis, so I do recommend keeping SAS on. Also, as usual with my planes, direct yaw authority is rather limited. Rolling and pitching to steer works much better.

Hmm, if we're going to exploit physics bugs, surely a Zompi drive or some other form of reactionless kraken drive would be more useful here than magic wing props? But I guess that would kind of trivialize the challenge. Mind you, "no wheels" doesn't necessarily mean "no horizontal landing." Although it does make it quite a bit more challenging, especially if the goal is not to lose parts.

Sure, if you use enough of them. And in any case, it's well known that you can use Kerbin and Eve assists to get to Jool, which will then let you escape Kerbol. In fact, I'm pretty sure that if you can get from Kerbin to a Mun encounter, the rest of the trip can be done with arbitrarily low delta-v if you just have enough patience and fine control. Basically, you'd use a Mun gravity assist to get out of Kerbin's SOI, then repeated assists from Kerbin and/or the Mun to get to Eve, then Eve and Kerbin assists to get to Jool, and then a single Jool assist should be enough to reach escape velocity. It's going to take a while, though. I wonder if there could be an interesting optimization challenge here. Maybe something like minimizing launch cost multiplied by the time to reach some given distance from Kerbol?

Do docking ports in KSP even care about relative spin about the docking axis? If not, this shouldn't be significantly more difficult than a normal docking. Actually, with no time pressure, it's shouldn't be too hard to match spins, either. It seems like the main difficulty in the scene (realistically or otherwise) comes from the urgency and the fact that the vessels are presumably deep enough in the atmosphere to experience non-trivial (and different!) amounts of drag. Which does definitely make docking harder — I know because I've tried it. And it was indeed really tense. For an extreme version of this challenge, put the target docking port somewhere not on the axis of rotation. It should still be possible, but depending on the rotation speed and the distance from the axis (and the orientation of the port) it could be anywhere from slightly more difficult to damn near impossible.

Solar electric prop planes are allowed, aren't they?

Very well, thank you. It basically stays upright the same way as a real bike: when it starts to lean e.g. left, the contact point between the wheels and the ground moves slightly to the left, which makes the bike turn left, which makes it lean right again. Also, it cheats a bit by having a small reaction wheel set to SAS only. But you can turn SAS off and it's still decently stable at low speeds (about 10 m/s feels close to optimal). At higher speeds it tends to wobble without SAS, though. The wheel geometry matters a lot. The Mk1 version only had a single row of grip pads on each wheel, parallel to the wheel axis, and while it kinda worked it was a lot less stable and harder to control. The two rows of grip pads, tilted and offset with a slight gap in between, sort of approximate the tubular shape of a real bike wheel. I could probably make it handle even better by adding more rows of grip pads to make the wheel surface rounder, but this seems to work well enough.

Inspired by @king of nowhere's run, I also decided to give this a try. I actually flew this mission in Science mode (since I happened to have an existing save with all tech nodes up to tier 4 unlocked, just for challenges like this), but I made sure to respect the part count and launch mass limits and not to use any SAS modes other than attitude hold, prograde and retrograde. Up to Popestar if they want to count this as a valid entry or not. Anyway, here's my lander (craft file on pastebin): It's based on what's basically my standard early-game manned Mun lander design, with a five part core (parachute, command pod, decoupler, FL-T200 fuel tank, Terrier engine) and a pair of side-mounted FL-T400 drop tanks, stripped of any unnecessary stuff like science experiments and landing gear. The booster stage consists of two FL-T800 tanks (and a partially filled FL-T100 at the bottom, just to max out the mass limit and squeeze a few dozen extra m/s of delta-v out of it), a Reliant engine and four basic fins for stability. Obviously, nobody would even think of launching such a… ballsy rocket in real life. But it works great in KSP. Mission screenshots below (more on imgur): All in all, this wasn't that hard of a mission, although the delta-v margin certainly is a bit tight. As king of nowhere noted, even just a few tonnes more allowed launch mass would make this significantly easier. Still, even with these constraints, I made some noticeably unoptimal choices during various parts of the flight and still made it home without having to resort to gravity assists or EVA pushing or any other tricky maneuvers.

Since no-one else seems to have entered yet, I decided to have Val grab her motorcycle and head for the mountains. Now, while cool and fun to ride, this bike is hardly the fastest thing on Kerbin's plains, and I'd done no route planning besides "let's aim for that mountain pass, it looks potentially climbable." But it's pretty stable, so I could leave it running while surfing the web, and after a little under half an hour of leisurely cruising Val had made it to the foothills and reached the first altitude target of 750 meters above sea level. Of course I didn't stop here, but kept heading for the mountains. As it turns out, this bike is a pretty decent little hill climber. Its top speed is mainly limited by the maximum RPM the rear wheel can handle before starting to glitch, not by engine torque, so it'll barely slow down when going uphill and can handle slopes of up to 20° quite easily. Thus, only seven minutes later I was already at 1500 meters and climbing. And in just a few more minutes (most of which were spent getting the bike back upright after I accidentally flipped it over on a steep slope) I was well on my way to 3000 meters… …when the bike flipped over again and this happened. (You can't really tell from the still image, but the bike is actually happily cruising down the slope with nobody at the controls!) As it turns out, the bike came to a stop still in perfect condition (well, expect for the missing headlight that was lost in an earlier crash ). But it's not gonna drive itself back uphill without a pilot, and it'd take way too long for Val to walk down to fetch it (especially since she keeps tripping if she tries to run down the slope), so I'm ending this run here. Maybe I'll do another one later and record a video of it. (I tried to do it this time, but because my laptop is old and I was playing at a too high a resolution, I was only getting about 1.5 FPS out of it so I gave up and decided to just do a test run without video.) BTW, riding this bike in the hills is totally awesome! At the risk of engaging in self-promotion, if you have Breaking Ground I highly recommend downloading it and giving it a try. The controls take a bit of getting used to, since it basically steers by weight shifting (bound to roll controls) and countersteering, like a real motorcycle, but once you get the hang of it, it's a lot of fun and can handle some pretty extreme maneuvering. (Tip: I recommend driving mostly at about 70% throttle, which gives a speed of about 20 m/s; while the RPM controller will let you go a bit faster than that, the handling gets a lot worse. Also, briefly tapping the brakes will very effectively stop an unwanted wheelie. Just don't hit the brakes too hard or the bike will flip over the front wheel instead! ) Ps. I believe this should qualify for the Purist and Check yo staging achievements, and would've almost qualified for Flawless if I hadn't lost the headlight to a stupid steering mistake early on while still in the grasslands. I don't think this qualifies for Certified Organic non-GMO, although technically there are no wheel parts on this vehicle — just rotors and grip pads. (Also, I'm not 100% sure how to actually interpret the description of Certified Organic non-GMO. Did you maybe mean to forbid rotary rather than reciprocal motion of any kind? That would make more sense to me, and would definitely make this vehicle not qualify for it.)