wumpus

I saw this in the "how to play without Asparagus" thread and thought it was too good for just one thread. The other important thing to take away from this is that to a remarkably high degree, designing a rocket means designing a turbopump (fuel pump). I think Scott Manley has stated something like 90% of the engineering comes down to the turbopump, but such things vary so wildly that I doubt you will ever find an "official" number. Just understand that they are by far the most complex part of rocket science and that making your turbopump more complicated is making your rocket more complicated. And making things complicated is the enemy of doing reliable engineering. And while someone pointed out that fire engine pumps aren't the same as rocket turbopumps, I think we can understand that making turbopumps working any harder is a bad thing.

I'd also suggest a few more missions. Following the Scott Manley tutorial:Unlock the Kerbalnaut building to allow EVAs. Go into space (this is complicated. Not the going there, but getting back alive is tricky. I'd suggest a 45 degree path to survive, but would need to test this first. It gets worse when you realize that Bob is optimal for science but can't enable SAS). Once in space, do an EVA and EVA report. Also learn to do the "science dance" and get the science from any instrument and then reset the instrument. A cheap way to orbit is using BACC SRB and Terrier upper stage rocket. Experts can try to recover the terrier, but I've given up on it. Again, Bob is optimal for science but lacks SAS (I don't think this rocket can bring the Science Jr along anyway). Note the "science dance" proper involves storing the science in the capsule (and Jeb and Valentina can do it as well, they just can't reset the instruments). Practice doing this by hanging onto the capsule as Kerbin whizzes by and do an EVA report as something new whizzes by. For each different biome you can collect different EVA reports, but each time you have to store them in the capsule (you can only hold one type of each scientific report ("report" or measurement)). Once you feel you have all that you are going to get, get back in the capsule and return to Kerbin (thrust at retrograde to reduce your perigee (not sure you can see your perigee at this point) to ~30km. Dump everything below the capsule (a heat shield is a good idea but shouldn't be required) and when your parachute icons turn green (*after rentry*, they might be green in the upper atmosphere but won't survive rentry if you open them then) hit the space bar to open them. Experts can amuse themselves trying to salvage the upper stage engine: I gave up after awhile and don't think it can be done with unmodded KSP with a minimal tech tree.

If this is true than the aero/physics needs another overhaul (although I suspect squad is fine with it and you should be aware of the issue and use FAR if it is too big). While a bigger rocket might have greater drag then smaller rockets, bigger rockets (at least taller rockets) have (historically, anyway) much lower drag/weight ratios. In other words, while there might be more overall air resistance, compared to the total weight of the thing being heaved into space it becomes less important. This is the reason on why the Pegasus flies: as small as it is it needs to launch at 39,000ft (12,000m). The big boys don't care and launch from sea level. Personally, I prefer high TWR rockets (and thus higher speeds). This has a few efficiency benefits, but don't try to do a "real" gravity turn with them: going hypersonic pretty much locks your attitude in place. You need to angle your rocket in the right direction any way you can before getting locked in place.

He also states he wants to get into orbit and doesn't have a lot of time to spend on KSP. Yet you insist on suggesting playing KSP in the absolute slowest way possible. To get into orbit you need: 1. >3000 m/s delta-v (the higher the better if you haven't done it before. Also make sure your TMR is at least >1.0 or you won't take off.) KER/MJ will let you know this as you build your rocket. 2. A reasonably good idea of your orbital trajectory. Head up and gradually point East (90 degrees on the navball). Make sure any tutorial is from after 1.0.4 (or at least 1.0.0) as the aerodynamics have changed enough that the previous "force your rocket to make a quick turn" just isn't going to happen. 3. A reasonably stable rocket. Use fins on the bottom and some means of control (typically the capsule will be enough, if you have even basic fins on the base). Bigger rockets will have bigger problems, but getting to orbit shouldn't require anything too fancy. If you watch Scott Manley's excellent tutorials, I'm sure he will go over the rocket equation. He will then suggest using KER or MechJeb as the way to quickly determine what the delta-v of your rocket is. Knowing the rocket equation helps. Manually calculating the rocket equation each time you change a rocket is going to use up OP's playing time quickly.

It should be fairly easy to compute from the wiki (and it isn't all that much). It also isn't terribly meaningful: To get *all* the science in KSC you need to upgrade all the buildings to level 3 (might be one or two that don't need it, but budget for all of them), then enough science to unlock all the measuring instruments. By this time you can do nearly the same thing on Minmus and get some real science returns. You have to ask yourself: How many times am I going to do this? Is it really worth it to do the full KSC sweep after everything needed is unlocked? Is there some time when I just need a little more science and can build a science buggy to pick up the local science (even if there is a lot left to be unlocked)? Do I want to do it again to get the rest? Granted: this is from someone who has three copies of nearly all the KSC science collected in mark 1 capsules waiting to be sent to Minmus science base (not looking forward to the transfer climb). After doing the grind I can't recommend it at all. Go do some landings for real science.

The scientists, doctors, and engineers tend to live in relatively urban areas. This is not a good place to survive. The ones who will really survive are jacks-of-all-trades (specialists like the scientists, doctors, and engineers are in trouble). Conspirators, denialists, politicians, and newspeople will need to be able to see things as they are to survive. Death would come way faster than they could possibly learn to see. Also don't expect human technology to go beyond mideival tech. Coal, iron ore, you name it: if you need it to fuel an industrial revolution, somebody has ripped it out of any easily accessible place (with mideival tech). Steel would obviously be available (recycle 20th/21st century junk), but the fuel would be an issue. Also don't expect population to ever recover: the feedstalks needed for fertilizers just won't be available. Lower population means lower tech growth: there just won't be as many people trying new things and specializing it the stuff needed for the new tech. Extinction level events collapse ecosystems. Being one species trying to survive in a collapsed ecosystem may well be impossible (fortunately humanity is world-wide. Don't expect your little area to have *any* humans left). But technology acts as an ecosystem as well: try reading/watching James Burke to see why (that's pretty much his only schtick. Once you get what he's saying, he doesn't have much else to say. Still worth understanding).

You should be more specific to a stock career mode game. Building such a thing in sandbox would be its own reward. Any ideas on how much *more* delta-v you have to supply to the crew and supplies (vs. a Hohmann transfer)? Somehow I don't think that an orbit that keeps hitting Earth and Mars is all that close to the ideal transits between the planets. That said, it has to be easier to accelerate the crew and supplies (presumably in a dragon/orion or similar) than a ship capable of housing them for months.

And such was a reasonably rational belief. The catch is that Congress is not designed to *want* to save costs. They want to maximize funds to go to their "doners" and constituencies, and like to be seen as providing for the common defense (regardless how insane the Air Forces suggestions are). There are many, many threads on why the Shuttle failed. But it wasn't because such a thing was all that hard, it was because Congress wouldn't pay for it. The real question is why they paid for something else.

This is more a plan to get there effectively (in more or less modern technology, but with more than a little wishful thinking*). But it is semi-current from NASA, so I'll mention it. http://www.nasa.gov/pdf/373665main_NASA-SP-2009-566.pdf * both plans require zero-boiloff hydrogen storage. There still is no known way to make such (and it *has* to stay in space. Don't ask how many more times volume of H2 you have to drop to fill up your return rocket, even with a quick turn around. The X-15 took off (i.e. the B-52 took off) with 250% of the H2 it would need once it was dropped. H2 works great for gas-and-launch. Not so good for gas-and-wait. And nerva needs H2 (well, water might work as well as H2 and LOX. Which is great if you can scoop up some water on Mars...).

Other things killed off by a successful Falcon landing: stupid SSTO proposals (I hope). Arguing for SSTO tends to confuse cause and effect. "Magic" ISP leads to SSTO*. Trying to do so with existing chemical reactions leads to madness. While the empty booster SpaceX landed might not have weighed more than the payload (my back-of-envelope gets them within an order of magnitude), this would require at least doubling the fuel needed in the second stage to get it to orbit, which requires again doubling the fuel (and thrust) needed in the first stage - without increasing the dry mass at all (and that's assuming some pretty generous calculations. I'd also like to see what the SpaceX engineers would have to say about cutting the dry mass of their rocket in half). Or you could just recover the first stage and be done with it. Which is still likely easier than recovering the whole thing from orbit. * yes, all air breathers can be assumed to have "magic" ISP. But even the Sabre people don't really explain why it is so much better to use the (presumably) expensive SABRE engine (instead of a normal rocket engine) for going from mach 6 to orbital velocity (and haul that whole air cooler along to orbit as well). The supremacy of the staged rocket hasn't changed a lot since Goddard's time, and air breathing stages are extremely hard to get anywhere near half the delta-v needed for orbit.

So Neil Armstrong's relatively recent death and considerably extended life (past his astronaut career) was due to "cheating" in both Gemini and Apollo (two missions where he would be dead without overriding the automatic controls)? I'd still recommend trying to learn to do as much as what mechjeb does as you can. I'd also think twice about racking up those juicy 100% returns on landing on the pad (SpaceX uses automatic landers that can hit the landing pad, but I'm pretty sure that they don't get 100% cost-free reuse).

The "wrap around" method is just the start of the quote. There really isn't any reason that asparagus have to be circular. I've switched to doing them as one long line of boosters. That way is easier to view and attach (in KSP), and only really becomes a problem if you don't fit on the VAB and/or launch pad. The Martin Company (I'm assuming non-British, it appears to be LA based and some engineers I knew who were part of the Apollo program worked at Martin in Baltimore, MD) appears to have been working on inline rockets and clearly hits the critical points of asparagus staging: 1: All rocket engines fire at once (note that ignition is tricky and this was certainly appreciated even more in 1958) 2: Fuel tanks fuel all remaining engines and are dropped when empty. But unfortunately, nobody seems to know how to make a good cross-feeding fuel supply.

The only worry I have with MechJeb and cheating is that the ability to land *exactly* on the landing pad is overpowered considering that it hits 100% reuse return. I suspect that using the "mission program" that adds homing devices on units (and will no longer give 100%) should be used if this bothers me too much. I don't feel that mechjeb outperforms me in other ways (of course I use and abuse Kerbal Engineer all I can). There is just no way I can hit the landing pad from orbit without MJ or similar mods.

While the name "asparagus staging" apparently barely predates KSP, the idea has been around a long time: And obviously it is harder than it looks like in KSP. The book I am quoting is from 1958 (presumably hastily published after the Sputnik launch). It is a nifty book I found in my parents old collection: being from 1958 it starts *everything* from first principles, without launching into extremely specialized engineering. I've looked through some fairly well stocked used bookstores for other Kerbal-type books on rocket science, but haven't seen anything close. http://www.amazon.com/Space-Flight-Early-thoughts-projections/dp/B0000CK2Z4/ref=sr_1_1?ie=UTF8&qid=1451578393&sr=8-1&keywords=space+flight+by+carsbie+adams [ps. Of course it should include trigger warnings for 1950s-type thoughts.]

I like the "kerbal" level of realism. I am also very glad that the realism overhaul set of mods exist, and when I am "done" with KSP proper I expect to spend time in the "real" thing. I want explosive staging back. To me that was the "most kerbal" part of the game and I miss it.

Best bet is to use less parts. I run a lower clocked 8320 and a 560, and haven't had much issue. My guess is that I'm not as spoiled on other games as you and your 290, so I never miss the difference. Note: I'm pretty sure the Scott Manley videos from when he was doing the Kerbin-Mun-Kerbin runs in <1 hour was *not* shown in real time, but sped up somehow during the first few stages. Wackjob sized crafts and huge space stations are always going to be a problem (and the way scaling works, you probably just keep building until it becomes a problem. And when it becomes a problem it becomes a huge problem). [Scott just made a video of building a state of the art Intel/Nvidia 32GB beast, but I think the old machine was an AMD. It seemed to work pretty well.] Also, Squad has made at least some sort of commitment to fix this type of thing with 1.1 (for unknown values of fix). The processors in the PS4/XBox1 are 8 (AMD) jaguar cores and much weaker than even the AMD FX cores. For KSP to "work" on consoles, they have to require much less single core power. Just don't expect part limits to ever go away: there will always be either a scaling issue that limits your framerate or some sort of artificial limit (such as the VAB part limit) that keeps you from banging away at the scaling limit.

I bought KSP for my nephew last year. As far as I know, he has only made planes (and I've made almost exclusively rockets). Lucky for us it includes planes.

Very odd. I didn't expect those parts, but that is what came with the steam download. It was well before the Christmas Steam debacle to. Maybe I'll finally make a dummy steam account to test out the demo (it wouldn't let me "buy" KSP "again").

Reaction Engines Limited (the SABRE people) are probably less likely to produce their engine (or at least likely to build it much later). The two advantages SABRE could possibly have over Falcon are lower fuel consumption and longer operational life, and that is only on paper. Falcon 9 does all the important things and does it in space, not on paper. Any guesses as to the actual effect on costs here (for Falcon9 1.1 or 1.2)? I'd expect SpaceX's cost to go down by no more than 50%, and expect that to be pretty optimistic. The claim is that 70% of the cost is the booster production, so that leaves a fairly hard limit to the cost reduction of a factor of 30%, but that assumes no launch costs at all (the hardest limit is that there are no plans to recover the second stage, so costs can't drop by more than 90%). I'm pretty sure that ULA was created in such a way that could exist without ever needing to launch a private satellite (similar to how the 747 project could be profitable without ever carrying a passenger). Looking over the wiki-page, if Delta-IV is already scheduled to be phased out by 2017, it simply doesn't have any competition from Space-X. Space-X doesn't have enough time to establish any record of reliability with Falcon Heavy before that time (even if it gets extended for a couple of years, because government and pork). It certainly doesn't help to watch a competitors cost get slashed in half, but if you biggest customer is the government, it looks like something they've already planned for. What really looks good for Space-X is the Falcon Heavy program. Do they have two ground landing areas for the outer stages? It really looks like they can do a falcon heavy launch for nearly the price they used to pay for a falcon9 (I don't really expect them to land on a barge with the center booster. It is harder than any landing attempt made so far). No ideas how many customers there are for such a launch.

Have you noticed the buzz about drones (both personal and the hellfire-missile-totting variety)? Crewed craft (either air or space) have hard limitations on the g-forces you can pull and remain conscious/healthy. This leads to a real difference in current tech, although USSR was the only place to launch a crewed armed space station (might have been Russia by then, around when it fell), and I think all "Star Wars" weapons launched by the US have been unmanned.

I must be missing something. Constant altitude burns should require more TWR than less efficient routes. In fact, the higher the TWR, the more efficient you can make your route. I can see why you rant may be applicable, but I typically find TWR much more useful (and increases *real* deltaV) in atmospheric ascents (of course if you just visited Tylo, I can really see your point). Low TWR on non-atmospheric landing doesn't seem to be much of an issue.

Notes on the demo: [I played the demo over the holidays, and just got back to the machine with my password manager] There are a ton of tutorials for getting to orbit and the Mun in KSP. Nearly all of them are for much later editions than the demo (I'm guessing you knew that when you posted the orginal post). The basics: Getting to Orbit. You need to get to at least 70,000 meters of altitude and 2000m/s horizontal velocity. The horizontal velocity is the hard part (see snide sniping in the recent Bezo/Musk spat). To get to the Mun you do this and then need another 800 or so m/s just to get out there, then you need to decelerate and land, and take off... The parts (whole list found here: http://wiki.kerbalspaceprogram.com/wiki/Demo_parts) [rocket engines] - T30 (non-swivel) - Likely your main engine until you are in space. T45 (swivel) - Just like above only it has the ability to thrust in [limited] specific directions. Use one of these in the center and make the rest T30s. LV-909: highly efficient/low thrust rocket. Once you are in space this is the only rocket you should use. It might even replace the T45 in the center, depending on how you design your rocket (don't do this in the current game, around 1.0.0 it was nerfed for atmospheric use). RT-10 (the solid rocket booster). Note that you can't build these around another solid booster, you can put a bunch around a liquid rocket (the three above are all liquid rockets). This can get annoying as one of the main points of SRBs is low cost (the other being high thrust), and you are stuck either attaching them via side decouplers or ditching an entire rocket stage when the SRBs burn out. Fuel tanks - mostly obvious. You probably only care about the rocket fuel and not the RCS fuel. SAS: Put one of these on your ship if you want to lock your heading in place (I like to). Hit the "T" button to lock heading, then launch. After about 10,000m, hit T again to unlock and point to 45 degrees (heading east i.e. 90 on your navball). Hit "M" for map view (and go down and click at the bottom to bring your navball up) and then then burn until you are high enough (70,000m is minimal, about 100,000m makes things easier but takes more fuel). If you don't lock your heading, do pretty much the same only with adjustments on course. Note that the full version does everything here differently, and the rockets flop around off course enough to want SAS enabled almost all the time. RCS: If you want a real challenge, try doing a rendezvous. Note that you won't have the parts for a docking, but getting close is more than half the battle. Otherwise you really don't need to bother with either RCS thrusters or fuel. Decouplers: How you drop the first stage when it is burned out. You have vertical and horizontal decouplers, horizontal decouplers is how the magic happens. Also note the cost of the things: if you want to play the game for a little while, reducing the number of horizontal decouplers is a good way to reduce overall cost. Adaptors/Couplers/Struts: 3-1 adaptor: Connect three for the price of one decoupler. Probably comes most in handy for attaching SRBs. Fuel coupler: Probably the most important part in the entire demo and easily overlooked. This allows "asparagas staging" (an old kerbal specialty, if less important in current releases). The magic is this: click on you lower stage (but horizontally attached) liquid fuel tank and drag to a higher stage fuel tank (the order is important and determines which way the fuel goes, right click on the part for a reminder in game). Then adjust the staging so that all rockets fire at the same time. Once you set this up, all the fueled engines fire at once, but only drain the lower stages. Once the lower stages are drained, the lower stage rocket shuts off (out of fuel) and is jettisoned. The upper stage fires on at full fuel load and then drains itself as it goes. This allows much larger rockets in the demo than vertical staging (the rockets can only lift a few large fuel tanks on their own, any larger needs lots of rockets in parallel to lift it all). One of the tricks I came up on my demo holiday was placing a small fuel tank beside the rest of the rocket, and then surrounding it with four SRBs. With the 4 SRBs (per side), it hardly needed an engine for itself, but with the fuel tank being drained by 3-5 other engines (in "higher" stages) it would drain before the SRBs burned out. The resulting rocket was shot at a fairly high speed over 5000m and with nearly full fuel tanks. Struts: add these to keep your rocket if it wobbles about. Anything with multiple SRBs attached will often get ripped straight off the rocket and shot upward without some struts attached. Ground support/launch clamps: Launch clamps are a good idea in general, and help avoid a nasty bug where parts touching the base get stuck to it. Note that in the full release (in career mode) launch clamps have effectively zero cost since they are recovered on the launch pad. Nose cones: Include only if you think the looks need it. Due to the horrendous aero/physical model in everything before release (unless you added the FAR aero-fixing mods), these actually add drag. Winglets: AV-T1 (non adjustable) - don't bother. Unless you are really having trouble keeping on course (and the SAS should help more), these shouldn't be needed. AV-R8 (adjustable) - note that these let you guide your rocket and thus have some use. Just understand that they typically go on the ends and get discarded pretty fast (but that shouldn't matter if you have a T45/LV909 on the next stage). Parachutes (only one, and remember to pack it. I don't think it can save more than the capsule itself). Ladders: Bring these along so you can walk on the Mun. Not needed in the release, and I think there should be a means of using the RCS pack on the Mun (it told me the tank was full, but I couldn't access it), but a good idea. Landing gear: missing on the wiki (which meant I hadn't looked for it on my successful trip to the Mun). Always a good idea. Note: it might make a lot of sense to have both a "decent stage" and a "takeoff stage" on the Mun (Apollo did this, for obviously different reasons). When I landed on the Mun, I did a quicksave. After loading I found one of my engines stuck to the surface (if I launched with it connected it would rip off and my unbalanced ship would quickly crash). Fortunately I had extra engines surrounding my craft (I didn't know landing gear was in the demo) and was able to stage and drop the connected engine. Ship building hints: (note the above posters likely included better specific designs, the following are more general and how to find a good design). The obvious one is build horizontal and use asparagas staging. Note: while it might get mentioned in earlier guides on KSP (and was until recently planed for Falcon Heavy) it is hardly that magical. The real benefit it has is not "wasting" the mass of rocket engines by allowing their thrust to be used by more than one stage. The thing you learn in rocket science (i.e. KSP) is that a little efficiency gains by doing this go a long way. I'd recommend starting at the top and working down (the game helps enforce this by handing you the capsule first). Build enough of an orbiter/lander/rendevous-ship/whatever and then figure out what type of booster you need to get it into orbit. The space[only]-craft should be powered by LV-909s (and can often get away with just one. You might attach "drop tanks" to the side to get to the Mun. Gravity isn't going to pull you back, any extra thrust isn't needed to get there and just adds mass you will need to lift/accelerate). The most efficient means to build a booster is essentially a row of (pairs of) boosters. Start with a T-45 and a couple of large tanks. Then attach horizontal decouplers and add a pair or similar boosters (two T-30s and two large fuel tanks). Note: Hit the "double symmetry" button and try to add the decouplers in the center of one of the fuel tanks (the next set of fuel tanks likely will only connect in the center. Getting these things to attach is one of the bigger frustrations of KSP). Connect your fuel lines from new to old, then group the staging. Repeat with as many boosters as it takes to get into space. Note that if you build your lander with 3-fold symmetry (to make it more stable during landing), it works fine to build the rest with 3-fold (or more) symmetry. 2-fold is the most efficient, as it pulls from the least number of tanks at once, but it isn't that big a deal. A word of warning: while mods exist to make everything easier (my rockets were likely silly-over powered due to a lack of Kerbal engineer), the Mun in the paid release bears little resemblance to the flat surface of the demo. You might want to land on Minmus first once you buy the full release (although for some reason career mode won't offer you Minmus until you land on the Mun. Something to do with Minmus having an inclination I guess (note: I'm pretty sure Minmus didn't make the demo. It is a smaller moon much further away than the Mun, but easier to land on being flat and having a much lower gravity)).

For various reasons, I tried the demo over the holidays. One thing that hits you when you go from 1.0.5 to 0.18 (insert actual demo build here) is that while you lose a ton of polish and physics accuracy (not to mention planets and other goals), you gain a certain Kerbalness. The demo build just has you, three kerbals, some rockets, and gravity. This is pure KSP. While I highly recommend the full version (even if you insist on 0.90, it certainly beats the demo and pays Squad's bills), I would suggest that the devs try firing it up occasionally and try to figure out what makes this game so great (however good later features are, looking at them obscures the forest for the trees). One thing that I would recommend re-including from earlier editions would be explosive staging. It might not really have a place in the modern career mode, but it so uniquely kerbal that I think it should remain in the game (and no, I really haven't looked into the heat model to see if it is even remotely an option, although if firing a rocket into a SRB isn't going to heat it up we have greater issues with the heat model than even mentioned above).

If you have an older instalation and recently returned to KSP, check your file integrity! I had parts of different versions and started a new 1.0.4 game and found it impossible. Upgraded *everything* [not mods, just got steam to get it right] to 1.0.4 and suddenly surviving a "just into space" became possible (although it still tried to kill poor Jeb, much worse than almost any other mission I'd seen [haven't landed on Eve]).

Therein lies the problem. I've tried throttling a jet down where it "should" be (about 1/10 power), and it will barely get off the ground. I'd assume that turbojets "ought" to flame out around 10000m or run out of thrust around 600m/s (the thrust is the key. On Earth we can't get much past 3/10s of orbital velocity. Doing so on Kerbin makes spaceplanes viable). If you have a problem with kerbal jets, you will probably need to install realism overhaul. Just "fixing" any one issue leads a deep rabbit hole of other issues. Fix the size of Kerbal and then you can "resize" the rockets for the real thing and the jets work "right" as well.