vyznev

Honestly, the Dynetics lander kind of looks like it was designed in KSP to begin with. (Not saying that to disparage the design in any way — I'm just saying that a lot of the parts look kind of familiar. And mounting the lander sideways inside the fairing during launch feels like a very Kerbal solution, as does the use of disposable drop tanks for extra landing delta-v.)

BTW, a few questions and remarks about the scoring: I wouldn't mind if the scores for Tylo and Moho were swapped. As it stands, there seems to be no point in going to Tylo, since it takes more time and delta-v and bigger antennas, and it's harder to land there too. For that matter, I kind of suspect the scores for bodies with an atmosphere (Eve, Duna, Laythe, Jool) are a bit underrated. For very small landers like this, an atmosphere seems to be almost more of a hindrance than help, since even the smallest heat shields and parachutes in KSP are quite heavy (a tiny heat shield with no ablator weighs 25 kg, while a single Mk16 or Mk2-R chute is 100 kg, i.e. nearly as much as the total dry mass of my Mun lander!). I'm not 100% sure I've understood the time scoring exception for the Jool system and Eeloo correctly. Does it just mean that a mission to those planets (and their moons) always gets a time bonus of at least 10 points regardless of the actual mission time? (If so, basically the only destination where you won't be guaranteed at least 10 points for mission time is Duna, since a standard Hohmann transfer to Eve or Moho takes less than 200 days.) For that matter, is having multiple science instruments supposed to give any bonus score? It seems like you might've intended that, since there's limit on how many one can have and a footnote that they must all be distinct, but I don't actually see such a bonus listed in the scoring rules. Or am I just missing something? Overall, the scoring seems a bit coarse-grained: it's not that hard to max out the score for a given destination (noting that a transfer time of < 10 days to anywhere but Mun or Minmus is basically impossible, and so not worth even considering). Maybe the launch cost penalty of could be changed to something like -1 point for every 1000 funds, with no lower limit? Don't get me wrong, I think it's a nice challenge in any case. But I feel like fine tuning the scoring a bit more could turn it into a great challenge.

This seemed kind of similar to the Smallest Moon Lander Challenge from November 2018, so I decided to adapt my entry. Basically all I needed to do was add some landing legs (made out of cubic octagonal struts, since they're cheap and lightweight) and a booster stage for getting to LKO: More screenshots in the album at https://imgur.com/a/0c4hZ2L Notably, this craft has no reaction wheels or RCS at all, steering with engine gimbals only. That's not nearly as hard as it sounds, except that I made a slight mistake while tweaking the lander and moved the engine closer to the center of mass, which significantly reduced the steering authority I had for landing. (Did I mention this thing also has no SAS?) I still managed to make it to the surface in one piece (at least after a couple of reloads), but if I were to redo this mission, I'd definitely pull the probe core further away from the engine. Or maybe replace it with an OKTO2 so I could use retrograde hold. Anyway, if I'm not mistaken, this should give me 10 + 5 + 20 + 15 = 50 points for landing on the Mun, plus 15 points for doing it in less than 10 days. I believe this also should count as a small rocket, so that's +5 points, and the cost is well under 40k, so no penalties there. Thus, my total score should be 50 + 15 + 5 = 70 points, i.e. the maximum for a Mun landing. Of course, going to Tylo or Moho could beat that. I might try a Moho mission later… I think this is also the smallest lander so far. The total launch mass is 10,832 kg, of which the lander weighs 190 kg fueled (the dumpling tank isn't quite full) and 102 kg when dry. Ps. Craft file here: https://pastebin.com/eMqmMK71

That's bloody insane! You did skirt a few of the rules a bit there, but the sheer craziness of your contraption definitely deserves at least an honorary mention.

I guess this counts as an excuse to trawl through my old forum posts for some of the crazy contraptions I've posted here before, like: a rocket with a helicopter as its first stage (built before BG came out), a one meter tall Tylo lander, a five-part LF only SSTO (flies half of the ascent backwards), a gyroplane (that's really hard to fly), a kerbal diving bell for planting flags underwater (plus a pointless torture device), a method of toggling Sepratron thrust on and off using elevons (for orbital rendezvous and docking using only solid fueled rockets), and what's probably my most prized piece of underappreciated silliness so far: an airplane that can fly upside down, backwards and sideways (and any combination of these). (Quite appropriately, that last one was made for your own earlier challenge. So thank you for the inspiration. )

Might be SAS-related. You could try turning SAS off (or switching to stability assist mode) before separation. On my non-basket runs I found that the active pod would start drifting apart from the others if I kept SAS locked to retrograde. Apparently that's because the active pod stayed in retrograde mode, but the others automatically switched to stability assist. Since the pods are aerodynamically stable and will hold surface retrograde just fine even without SAS, I just turned it off entirely before staging. I'm not sure why you're getting more than one pod flying off, though. It might be because the ship is flexing slightly due to automatic control inputs, in which case turning SAS off might still help.

Well, having said that, obviously I had to give it a try. That's 27 crewed pods with a max distance of about 3.5 meters. I believe that means the scoring works out to: 10 points for each command pod with one or more occupants: 27 * 10 = 270 For each pod destroyed during reentry, subtract 20: 270 – 0 * 20 = 270 Measure (if possible) the distance between the two most seperated pods, distance to be in Km: 3.5 m = 0.0035 km Divide the score by the distance apart in Km: 270 / 0.0035 = 77142.857 Take the total cost and divide by the score to get the adjusted cost: 28128 / 77142.857 = 0.3646 Take the total number of parts and divide by (score / 10) to get the adjusted parts: 114 / (77142.857 / 10) = 0.01478 Of course, this score is quite easy to beat: just build a bigger basket.

Yep. Down to 9.6 meters now. And I even flew the pods to orbit this time. Scoring: 10+1 points for each command pod with one or more occupants: 10 * 11 = 110 For each pod destroyed during reentry, subtract 20: 110 – 0 * 20 = 110 Measure (if possible) the distance between the two most seperated pods, distance to be in Km: 9.6 m = 0.0096 km Divide the score by the distance apart in Km: 110 / 0.0096 = 11458.3 Take the total cost and divide by the score to get the adjusted cost: 29288 / 11458.3 = 2.556 Take the total number of parts and divide by (score / 10) to get the adjusted parts: 63 / (11458.3 / 10) = 0.05498 I think any major further score improvements will come from using something like @mystifeid's box trick to pack the pods more closely together after landing. Something like 27 pods packed in a 3x3x3 cube ought to work nicely. Oh, I didn't notice that before @doggonemess pointed that out above. In that case my adjusted costs and part counts are overestimates. I'm not sure what the correct values should be, though, since I'm not sure which parts of my craft(s) count as the "orbital vehicle". Just the pods, chutes and decouplers? Those plus the completely superfluous nose cone? Or maybe all those plus the fairing, fuel tank and Terrier engine? And if the latter, should I count all the fuel in the tank for the cost, or just what I actually had left after achieving orbit? Or just what I burned to deorbit? And for the fairing, should I count just the cost of the base or the full cost of the shell (given that only the base reached orbit)?

I think most of the attempts so far have made the mistake of trying to land the pods side by side. It's a lot more effective to have them all stacked in a long line aligned surface retrograde to prograde, so that they all fall along the same trajectory. My test landing with 10 pods in a row achieved a maximum separation of 117.2 meters. And I'm pretty sure I can do better by delaying the parachute semi-deployment until later. Anyway, if I understood the score calculation right, this test run should score as follows: 10 points for each command pod with one or more occupants: 10 * 10 = 100 For each pod destroyed during reentry, subtract 20: 100 – 0 * 20 = 100 Measure (if possible) the distance between the two most seperated pods, distance to be in Km: 117.2 m = 0.1172 km Divide the score by the distance apart in Km: 100 / 0.1172 = 853.2 Take the total cost and divide by the score to get the adjusted cost: 17524 / 853.2 = 20.54 Take the total number of parts and divide by (score / 10) to get the adjusted parts: 45 / (853.2 / 10) = 0.5274

Like this?

Good point, I hadn't even thought about that. Maybe add a rule against taking part testing contracts for parts that you haven't unlocked yet?

I did some test runs (with the "no science from minibiomes" restriction), and I don't think money is a real issue, especially since you won't be spending any on building upgrades. I didn't really pay much attention to it, but even launching (and blowing up) several extra flights and only remembering to grab a couple of early contracts, I was nowhere near close to running out. (And contracts do give science rewards. Not much, but even a little science is better than nothing.) Mun and back is something like a three-day round trip, so avoiding it definitely pays off even if you have to fly a subsonic plane to both poles and the badlands. Going to high space above Kerbin is worth doing, though — that only takes a little over half an hour, and if you really want to micro-optimize it, there are two ~10 min coasting phases that you can spend sciencing Kerbin's surface at the same time. The part count and mass limits are hard on Mun missions, though. I've pretty much concluded that my original plan with two crewed Mun landers is just not doable in caveman mode. Instead, I'm thinking of going with two probe landers with a single Science Jr. and goo can each, plus a single fly-by / orbit mission (which might as well be crewed). And maybe launch a couple of extra landers, just in case I crash one or land in the wrong biome. Still having some issues with lander stability. On the plus side, apparently my return stage is capable of horizontal takeoff.

I assume this is not quite the expression you're looking for?

Just right-click the image and select "Copy Image Location" (Firefox) or "Copy Image Address" (Chrome), then paste the resulting URL (which should end with ".png" or ".jp(e)g") into the forum editor and press enter. (Also, as an extra check, you might want to open a Private Browsing / Incognito mode window and paste the image URL there. If you can't see the image there, or if it redirects to a page showing more than just the image, then it's likely that your image host doesn't allow "hotlinking" and that other people may not be able to see your images properly on the forum. FWIW, imgur at least is known to work.) Ps. Just to make sure that you indeed know this, your peak TWR needs to be above 1.00 at least some of the time for hovering to be possible. At 0.99 TWR, you'll slowly but surely sink. Of course, as you burn fuel, your weight will go down and your TWR will go up, so starting at 0.99 is technically possible. I'd still recommend starting a bit higher so that you can immediately gain a little bit of altitude above the launch pad before throttling down.

Nice! Actually looks more like 1:13 to me on the video, but we're not (yet) down to single-second margins in this category anyway.

To add to my previous post, I'm now thinking about doing this challenge with the added restriction of not getting any science from KSC (or from any other launch site minibiomes). It should still be doable with only minor tweaks to my original plan, and would avoid the tedium of crawling around the KSC for scraps of science. My current mission plan would be something like this: Short atmospheric hop with goo canisters, land in ocean (or grasslands?) -> unlock Basic Rocketry, Engineering 101 & Survival Suborbital flight with goo, thermometers and barometers -> unlock Stability & Aviation Fly plane to one or two nearby biomes, get ground & low air science -> unlock Basic Science Orbital flight with materials bays, in parallel with more airplane sciencing of biomes near KSC during orbital coasting phases -> unlock General Rocketry, Advanced Rocketry, Electrics, etc. Launch two Mun landers (each with double goo canisters and materials bays for landed & space science), land in different biomes and return -> challenge completed! With the added restriction of not using minibiome science, leaving Kerbin is definitely required. There's just not enough science available otherwise. FWIW, in principle it should be possible to cut the time needed almost in half by only sending one-way probes out to the Mun and having them transmit science back. But the reduced amount of science from transmitted data would require landing something like 7 or more probes in different Munar biomes, which also feels way too tedious (and risky, especially without quicksaves). In fact, I'm kind of tempted to also add "no antennas" to my self-imposed restrictions — if I'm going to return the Munar science anyway, it really won't make much difference.

So I played around with this a little bit and did some math. My observations so far are: Just rolling around the KSC can get you enough science to unlock all the available science modules and reach orbit. The time this takes is mostly limited by your roller piloting skills, but around 15 minutes seems reasonable. Doing a single orbital flight (and hopefully reaching high space) should give you enough science to unlock a couple of useful tier 5 nodes. This will take about an hour of in-game time. (If you wanted to really optimize this, you could launch a second flight for repeat science while the first one is on its way to apoapsis. Not sure if it's worth it, though.) After that, if I did the math right, two Mun landings in different biomes should be enough to complete the challenge. To save time, you'll want to launch them in parallel. You don't necessarily need to bring a Kerbal if you've unlocked the OKTO, although the crew reports will get you some extra science, but you probably do want to return the science instead of transmitting it, even though it costs time. However, all you really need to return is an experiment storage unit; stranding a Kerbal on the Mun is perfectly acceptable in this case. Doing missions is not really required, although you can easily grab a couple of free exploration missions along the way if you like. So, basically, the biggest in-game time sink will be the flight to the Mun and back, which will take about three in-game days on a Hohmann transfer trajectory. It's probably worth trying to optimize that by spending some extra delta-v on the transfers. As for possible alternative strategies, if it was possible to complete the challenge without leaving Kerbin's SOI, that would probably be the winning strategy by far — even if it required squeezing every single drop of science from every biome on Kerbin and every KSC building. But I don't think it is. Alternatively, a high-speed flyby of Minmus (and another one of the Mun) might be worth considering, although I suspect that transmission losses may reduce the science yield too much. I need to do more math on this. On the positive side, once you're past Minmus, it won't take that much longer to leave Kerbin's SOI and get some interplanetary science too. Just don't forget to bring enough antennas. Edit: Never mind. I did some more math, and it looks like you can in fact do the challenge just by sciencing the heck out of Kerbin, if you get both on-the-ground and flying science from as many biomes as possible. My numbers are a bit uncertain because I'm not quite sure how many of the KSC building minibiomes are accessible for ground science without upgrades (and also whether or not you can get any "flying low over" science from them; I've seen it happen, but only on a few occasions; it may be a glitch), but even with conservative assumptions it looks doable. The only question is how many flights to distant biomes you'll have to make.

In an effort to breathe some new life into this challenge now that it's been selected as TOTM, and also to return it a bit closer to its original roots, I'm excited to introduce a new leaderboard category: Ultra Low TWR. To qualify for this category, your TWR must not exceed 1.1 at any point during the run. The simplest way to check this is to click the delta-v indicator on the left side of the screen, which will expand the staging menu to show the TWR of your current stage (and the predicted TWR of any future stages). I also very much recommend recording a video of your run, since the F3 display unfortunately does not show your peak TWR. To check that this was indeed doable (and to finally make a somewhat competitive entry to my own challenge) I flew a test mission with a final time of 1:27: As a practical tip, I recommend having some engines on your vessel that are not controlled by the main throttle. On my craft, the Spider engines are set to a fixed throttle that gives a TWR of 0.9 on (nearly) empty tanks, with the Spark engine providing the rest of the thrust, giving me much finer throttle control in the 0.9–1.1 TWR range. Having some way of stopping all engines for landing is also recommended — I bound all the engines to the "Brakes" action group, so I could just press B to temporarily cut all thrust. Obviously, my flight path is far from perfect and should be quite possible to beat. I look forward to many of you doing just that. Happy flying! Ps. Well done, @sollarflare! I've added you to the leaderboard. Alas, while your video looks like it might almost qualify for the new Ultra Low TWR category, without a visible TWR indicator I can't really be sure. But you could always try another run… With the practice from the first run, doing it again shouldn't take nearly as long.

Maybe I shouldn't ask this in advance and reveal my sneaky plans, but… is destroying the runway while landing allowed? The reason I ask is because I think there's probably going to be some significant time savings available in optimizing re-entry and landing, and in any case it's the part of the mission that one should plan first since it's executed last. And my current plan involves re-entering with a fairly massive craft (to minimize drag losses on the way down), aiming for an impact on the runway, and staging a few seconds before collision to detach a tiny lander that can decelerate using sepratrons and survive the remaining impact velocity. But that's going to mean that the rest of the re-entry stage is going to hit the runway just before the lander, and probably destroy it unless I turn the "indestructible facilities" setting on. And avoiding that seems difficult, since aiming for an impact off the runway would require the lander to maneuver after staging, which would either slow down the landing a lot or require inhumanly fast steering. Unless maybe I aimed for a spot just past the end of the runway, and relied on the deceleration burn to slow down the lander enough horizontally to hit the runway. Hmm… (Also, does the parenthetical note after rule 3 imply that the craft must actually have landing gear? Because with a landing plan like I outlined above, they'd probably just be dead weight.) Edit: Also, can I turn off Kerbal G-force limits? I'm not sure if they're normally on or off by default.

No, the copter has one battery on the right side and one large fuel cell on the left. (Of course, the copter doesn't really qualify anyway, since it's VTOL. )

Just to be clear, since you didn't mention which entries you were referring to, I'd like to note that mine does qualify under that rule. Even the left and right wings, while almost identical in appearance and aerodynamics, are built out of different wing pieces. Check the craft file if you don't believe me.

Hmm. I usually use the actual chase camera mode for landing (I'm not sure, but I think your camera mode got reset back to auto when you left IVA?) and the arrow keys for any manual viewpoint adjustments. It's not as precise as using the mouse, but works well enough. You could also consider rebinding prop pitch to main throttle, so that you can control it with your left hand. I saw someone do that in another thread recently, and it seems to make a lot of sense for prop planes.

Technically, all I have right now is the 50 points for Basic Skylon, because that's the only objective I've actually completed so far. I may try to complete some more later when and if I have more time. And I kind of expected the answer for the internal tank question to be "no", but I figured it wouldn't hurt to ask. I guess I could always design an "extended version" with a bigger cargo bay, but then it wouldn't look quite so much like the real Skylon designs any more.

Well, I made this thing: Nothing very fancy, really — it's a medium Mk3 cargo bay, two fairings stuffed full of fuel tanks, a pair of stubby wings and Rapier engine pods on the end of each wing. Plus landing gear and canards and a tail fin and RCS ports and all the other usual stuff. A single pair of Rapier engines wouldn't have anywhere near enough thrust to get this thing moving, so each "SABER" pod has seven Rapiers (and precoolers) clipped into it. That's the one slightly "cheaty" part of the design, and I consider it justifiable. it doesn't really cheat or exploit the drag model any more than is necessary to get the engines to look and work like they should. Takeoff with a full fuel load is a bit awkward, as the canards are a bit undersized and don't have quite enough authority for proper takeoff rotation. Running off the end of the runway works, though. I might need to tweak the control surfaces a bit more to see if I can make it better. For landing, I recommend using the lift-augmenting flaps that can be toggled with action group 2. (Group 1 toggles engine mode.) In the latest version I also added some airbrakes, and made the ailerons double as lift spoilers when braking. With no payload it can reach LKO with over 1600 m/s of delta-v left (probably even more with careful fuel/oxidizer allocation). I haven't really flown any missions with this thing yet, beyond verifying that it can indeed get to orbit, re-enter and land, but with appropriate payloads it should be able to perform quite a few of the challenge tasks listed. It can't carry a Jumbo-64 fuel tank to orbit, though — while it has more than enough payload mass capacity, the cargo bay isn't big enough. Unless maybe I can count one of the two such tanks hidden inside the fairings? More screenshots: https://imgur.com/a/2tfsPHl Craft file: https://pastebin.com/mUP2LjLt

As it turns out, it does. Allow me to present the Asymmetricopter! Craft file: https://pastebin.com/m2DJYV1F (Note: I/K control main propeller baseline pitch — not sure if it makes sense to call it "collective" when there's only one blade — and J/L trim the tail prop pitch. Stage to start the engines and the fuel cell. Basically it flies like an ordinary helicopter, except more wobbly. SAS is strongly recommended, although it's kind of marginally flyable without it.) Ps. I belatedly realized that this doesn't actually qualify for the challenge, since it takes off vertically. I mean, I guess you could take off more or less horizontally if you wanted to, but it'd be more difficult and serve no real purpose. (Real helicopters sometimes do it to gain extra lift by basically running ahead of their own downwash, but KSP's aerodynamics model isn't advanced enough to include such things.) Oh well, I'll leave it here anyway.