paul_c

The guy who packed the chute has just breathed an enormous sigh of relief!

Do we know which of the guys in mission control is doing the WASD controls and presses SPACE BAR at the critical points of the descent?

What is "before"? Are you doing tests of identical rockets which at some point in the past used to fly okay, but now no longer? And when you flew them before, could you actually 'fly' them as in alter the heading, angle of ascent, etc within a reasonable range? Or was it just doing a boggo-standard gravity turn that launched them? Or did Mechjob fly them for you? IF those tests are reasonably scientific, then something must have changed. Probably the version of KSP?? BUT I suspect something is different, ie the rocket is now bigger, has more stuff, more power, extra boosters, etc etc. I am guessing. Please post a screenshot of the craft, it will give a good idea of what's going on. As a guess until I see it, I am going to guess that its more draggy at the top than the bottom. This is a recipe for disaster once you get above about 5-10km, at that stage the atmosphere is still thick enough that its significant, and the speed is high enough too. Lower down, its not fast enough to mess up; and higher up, the atmosphere is too thin to influence it greatly.

Is anyone running a book on how big a crater it makes, when it crashes?

Ballistic is probably useful for optimising suborbital hops. But for take off to orbit, its putting so much energy in for the horizontal element, that its a case of "how do you best put it in over a finite period of time".

Interesting stuff! I've never looked too deeply at it but in theory: 1. Burning in one direction, then changing AoA, then burning in another direction is inefficient. It would be better to use just one direction then stay pointing that way to do the "sum vector" of the two burns. 2. Anything not pointing directly at the horizon is inefficient, because you're not doing anything for orbital horizontal speed, just overcoming gravity. (But you need a bit of altitude gain for terrain clearance) 3. You are of course subject to gravitational forces all the time (its what orbiting is...) but the longer you are "suborbital" ie Pe<0 or Pe<terrain, the more time you need to burn to overcome the gravity vertically, ie not pointing at the horizon. That might have occurred before in a liftoff/gravity turn, or it might be needed later if you need to point the nose up near/after Ap. So in theory, the "optimal" Mun-leaving rocket would find a flat/downhill section, tip over to 89.9deg and burn an enormous thrust engine to orbital velocity real quick (like wot a dragster can accelerate 0-100 in its own length!). And the "worst" Mun-leaving rocket would have TWR 1.00001, do a long long vertical burn to give enough height, turn horizontal and eventually reach orbital speed and avoid colliding with terrain. Of course, the bigger engine cuts into the payload/fuel weight allowance, so we don't take big engines to the Mun. So the optimal angle to burn at is related to the TWR? 45deg is a good ballpark; but if you've done a burn, then wait, then do a horizontal burn at/near the Ap you're inefficient (see rule 1)? If you can do one long continuous burn at say 20deg, would that be better? Or a 'grace' period of starting say vertical-45deg for a few seconds then tipping to (say) 15deg ASAP, for terrain clearance concerns? Also of course, well done on completing the challenge!!!!!! NCD time!!!!!!

Just looked thru my archive (Diamond attempt) and this was the "best" dV flown craft. It has all the science kit, but no equipment to re-enter Kerbin's atmosphere. I did do another which could do that, but didn't really use it, preferring to play it safe by gathering the science then docking with a "returner". From top to bottom, of the "business end": Docking port Small reaction wheel Battery OCTO controller 2x small fuel tanks science jr fuel tank, solar panels, aerial, thermometer, pressure gauge, goo at the base (for drag/aerodynamics) Spark engine Of course, its designed to squeak just under the 18t/30 parts limit. I normally spend an amount of time swapping fuel (tanks) between) stages to see what produces the best dV (respecting that the Terrier needs to operate in vacuum; and the Spark is low-power). The smaller the fuel tank, the finer the control so you can get closer and closer to 18t. The Science Jr is light but bulky. Its a case of: It has everything needed; and nothing not needed. Its purpose is to travel to Mun, then do a significant plane change (to reach poles or SW Crater), then land, take off, do another plane change and dock with something in an equatorial plane.

I don't know how you'd find them??

Relay "constellation" (3 satellites) are now in place, thanks to a bunch of contracts bumping the funding up (and borrowing the Terrier for incomplete testing!! I have 1 year before I need to do it...) I figured, if I set the Ap and Pe fairly accurately, they'll all have the same orbital period and it won't drift so much over time? Battery powered (no solar) but hibernated and thus consumes no idle power. Apparently relaying signals doesn't drain batteries in KSP either.... Note the thermometer, every relay can also do the "science from Kerbin space" contracts - I have about 12 transmits per satellite based on the battery capacity. I think (I am hoping) the grindy science/contract stuff is behind me now; in that more lucrative contracts will be generated, which pay £2000-5000+ compared to the £200-500 range before; and also more will fit in with the necessary science trips I'll be undertaking, eg "orbit the Mun" etc.

I think I'm at 110.6 science earned (I can't remember how much the unlocks were) but part of the fun is figuring out the order in which to unlock the technology. Some items are "wants" or "nice to have" and some are simply needed otherwise you'll reach a dead end. Mun is possible/'easy' once the Terrier is unlocked. Before then, the Swivel T45 is the smallest (and only, if you discount the Reliant T30 as useless because it as no gimbal) liquid fuel engine and its weight means you'll never achieve enough dV within the 18t limit (without other weight-savers like the 0.1t reaction wheel pack and a controller, to avoid needing the command module). I know people have done it with solid fuel etc but remember too, in Caveman there is no precision of maneouvers either. Docking port - yes its needed but its not a "blocker", since you can go to the Mun and return enough science with one decent vessel. Actually I've not done the detail planning to double-check its not a dead end again, but in previous missions I've managed a spaceship which could fly to the Mun, land, take off again and return its science in 18t/30 parts. Very stripped down, no science jr but its possible. My guideline was "3000dV left once in LKO" - 800 to get to Mun, ~200 to orbit, ~800 to land, ~800 to take off again, leaving 400 for return and corrections. And I was reliably sending a spaceship with docking port/science jr which could land on Mun, re-orbit then dock in Mun orbit. I have yet to do the detail planning for Mun stuff, since I know I'll need about £100k money to fund it all, it will come a bit later on. Definitely on NCD, there is a lot of needing to do contracts alongside to keep the funds flowing in, then an occasional "spend" to gain science. Its difficult to combine those two (repeatably).

I got so adept at "rescue" missions, I ended up with 28 Kerbals in the 12 capacity Astronaut Complex. If I lose one of these 'rescues' on a distant planet, its even?

Diamond is simple compared to NCD. I KNOW for NCD I'll need to visit and revisit pretty much all the Mun biomes 3 times - once to land & transmit; next to land/reorbit/recover using a remote probe; then probably a "mop up" extra points with a crewed mission (which will probably need 3x launches to get enoug fuel in the right place). Then repeat for Minmus.... There is a "hump" early on in NCD whereby to achieve the unlock of "Basic Science" (which unlocks many useful things - Science Jr, Stayputnik, battery, HG5 antenna, science store lunchbox) I needed to exhaustively gather almost all Kerbin science, including 3x "rare biome/situation" EVA reports eg "splashed down in the desert". This is my Kerbin science detailed tracking, I've not tried EVA holding on to a ladder etc but I believe others have had success with it. And my yellow highlighting means "not gathered", orange is "partly gathered" (eg 2x more Goo only earns another 0.2 science....but everything helps). Blaarkies is the guy to speak to on ground assembly. He's also done orbital assembly too. Orbital assembly - I found I could consistently get ~2.5t useful payload into orbit, and constructed an "upside down" spaceship with the controller/engine/etc at the bottom and the fuel tank at the top with docking ports both ends, this way it can give its remaining fuel but dispense with the superfluous parts. Satellite/spaceship minus fuel tank on top (which had docking port each end, and was detached with the stack separator):

Kerbin Badlands is easy compared to Tundra!!! It took a piece of luck to land in Tundra. I never even found Mun's "Polar Lowlands" on the last run (Diamond) and it took luck to find (Mun) "Lowlands" over the more widespread "Midlands". And "SW Crater" (and obvs Polar ones) need a big old plane change if you arrive, or have other stuff to dock with, on the equatorial plane. When I did Diamond, my Minmus trip did 6 biomes in one trip and it wouldn't have taken much to extend it to all 9 biomes (and/or crewed too). It relied on docking ports to make a larger "station" in orbit. It turned out, I had tons of excess dV at the end of the trip. The limiting factor was station-wobble, although for next time I am going to make the central spine accept 8x docked modules, this way I can keep the wobble under control. And in-orbit construction means no aerodynamic concerns (or TWR really, if you can do a multi-orbit TMI). I did 10 trips to construct it, and I'm fully prepared for 20+ dockings per interplanetary voyage.

"EVA Experiment Kit" is in "Miniaturisation", which is level 5 (90 science points). To unlock also costs £1500, then it costs £150 per kit. I've only used it once in the last challenge (Diamond) to try it out and it only worked on "landed" (I forgot it was global, not biome-specific) so didn't bother with it after, mainly because I'd taken a path of unmanned missions. These got me over the finish line. Obviously for NCD, a sum of the entirety of science gatherable from Kerbin/Mun/Minmus shows you'll need to go interplanetary. And a quick calculation shows you'd need about 300x HG5 aerials or some other weird way to get comms for other planets (even then, there is the science drop off for poor communications). So you're pretty much committed to sending a crewed mission or two.

The only science is: Temperature Pressure Goo Materials (Science Jr) Crew Reports EVA Reports EVA Experiments (I think!) (Plus "vessel returned from xxxx"; and anything that comes from contracts and world firsts - but on NCD, the 10% means I've not seen any there yet).

I will take one for my surface trip(s) to the Mun, if ever I get there! I know I need a pilot for the interplanetary stuff (or a spaceship that looks like a hedgehog with a ton of aerials) so I don't want to risk them, but I suspect I need the science points. I am assuming once on the Mun they can get out and gather EVA landed and EVA flying (by jumping) reports?

Are you going for a 1.11 NCD attempt?

As I see it, the only notable change is "EVA Experiments", which adds 25 science points per biome/situation; but its quite late you can unlock it anyway so its something for the end-game.

(The pics show fine for me) In my first career I got to the stage of doing "extract xxxx ore from Minmus and return it to Kerbin orbit". I had a huge base station with mining kit on Minmus so I used to send a craft with empty ore tanks (and a fuel tank) with a little engine, designed just to get into orbit then leave Minmus SOI to fulfil the contract. The thing ended up about 40t and 1.1TWR once it was full of ore, and the base was at the west edge of Greater Lakes (so it was on lowlands - better ore concentration - but still flat). The take off and departure had the whole lake to gain height so I used to give a 10 sec squirt upwards then shoot along the lake pointing only about 5-10deg above the horizon and gradually build height to just make it over the hills, its a very fun way to take off when you are confident it will work! Its good fun on the Mun too but for safety, I'd always be checking I will clear the terrain comfortably, if you land in a crater its quite deep. If you ever need to land in "Canyons" its aligned pretty much east-west so that's always an exciting departure, not unlike Innsbruck Airport.

It didn't take long to earn another 20 science points with the Science Jr. I have done most local biomes and flight/upper/space etc. So, I was able to upgrade to "General Rocketry". This buys the T30 Reliant. I tried it once, its pretty useless for my needs because I don't actually need its extra thrust and lack of gimballing in a single engine design leads to almost complete lack of control the reaction wheels in a command module can't overcome. BUT....it means a "test Terrier" contract might pop up. And indeed it did (it wants it tested on the launchpad - something which would never naturally occur, given its optimised for vacuum). So....... A careful, fairly well thought out (I only have the funds to try it once) redesign of my current biggest rocket resulted in this: Plenty of dV should get it to the Mun and back. Note, the command module - without it, there is no way to control roll etc, I need more science points to buy the reaction wheel on its own; and the blunt Stayputnik on top. My first NCD shot at the Mun. Note the deliberately short Ap, I am aiming for a flyby only, with a not-too-dissimilar resulting Kerbin orbit after, to get home: Success! 750km of Pe will be nice for a flyby: Not much fuel left but hopefully enough to tweak the Pe down for aerocapture, at the top of the homewards orbit: All science gathered into the lunch box. I'll return the CM home too, and a few instruments, whose £/kg are high: All for......12.2 science: No further science points but £11k of money is a nice little bonus too:

With Kerbin --> Mun, in a general sense its parent-->child. You never actually "fully leave" the SOI of Kerbin, yes you enter the SOI of Mun but that's enclosed within Kerbin's SOI. So its one 'step'. With Kerbin-->Duna you have to go via a solar orbit, so its child-->parent-->child. One analogy I like to use is snooker. Kerbin--> Mun is equivalent to hitting the cue ball precisely, so it hits another and pots it. Kerbin-->Duna would be a "plant shot" ie hitting the cue ball to hit an intermediate, to hit another into the pocket.

Ice caps here we come: Yay! Enough science points for the next upgrade, which buys a number of useful things including Stayputnik, battery, science jr, science storage. I'll gather the easier/cheaper Science Jr science first, then I'll need to build up funds again to take it further. I've already blown about 6 of them up, they are delicate and don't like landings! I can land a rocket on its engine nozzle, then if it falls over the Science Jr blows up.

A very general/vague rule of thumb I use, is that if the ratio of the radii of the orbits are >12, its worth considering a bi-elliptic (to change the size of the orbit on roughly the same inclination). (Also if you want to do a BIG inclination change, like 60deg or more...) Neither of these cases occurs with Kerbin-->another planet, so, its not considered for interplanetary.

A little luck as come my way at last. By chance, found the "Tundra" region: I did walk to the green area too, but didn't reach another biome - I suspect it was highlands/grasslands anyway (no new science there). Badlands. Note the "marker" (I left the heatshield) there from last trip: I wasn't too far from a Badlands lake so Jeb took a walk to investigate: Northern Ice Shelf. Suborbital, but using the orbit-capable vehicle. Reentry isn't too bad, there's excess fuel for a slowdown and the more vertical entry gives better precision on where I'll land: I think I just need to revisit Tundra and Ice Caps and I'll have just enough science for the upgrade. Not sure.... Run out of money again so its back to contracts for a bit anyway.

An oval shape is only very slightly less inaccurate than a fixed water level, to simulate tides. The height of a tide is mainly based on the bathymetry of the sea floor in the local area. It is possible to predict but it would be an order of magnitude harder; but it could be pre-done then modelled. Of course, water would no longer be modelled as "level" and need to incorporate a slope, flows, a water cycle, clouds, viscosity, etc too....don't hold your breath.