wumpus

I'd recommend using a cheat-enabled (with arbitrary funds and unlocked science) over an official sandbox mode anyway. The only downside is that if you have both sandbox and career modes running, you won't see the proper icons in the load screen. On the other hand, sandbox disables parts of the game, the most obvious to me is the landing percentages and cost recovery data. While the cost recovery might not matter (since you have arbitrary amounts of funds), it rarely matters all that much in career mode either but always seemed a worthy goal. In sandbox it outright refuses to tell you how much you recovered. Science will tell you how many Mits the data size is (dependent only on experimental apparatus), and nothing about the value of the data. If you like to do surface samples and see large values of science, sandbox can be disappointing (neither pre-unlocked and sandbox require tedious "science dances" in every biome you find).

Uses for flea: replace seperatrons for oversized rockets that couldn't be separated by seperatrons (but hammers are too big) An attempt to reverse course and return spend boosters to the pad before leaving the physics bubble (abandoned) I know I've wanted the spider and ant for uses where I only had thuds. But I think that was only one time. Reliants are pretty much superior in all ways to swivels if you don't need the pitch control. If they are used on small enough rockets, a mark1 can typically supply the needed control. Unfortunately, grouping with swivels is no longer an efficient use (I suspect they are still popular in the demo). Poodles are basically a lighter edition of four terriers. If you are too impatient with dealing with the TWR (or the TWR of a 3 or less terriers will *not make it to orbit*) it is a great engine. Also with a thrust four times higher than the nuke, this still helps in getting to orbit at all. Are the aerospikes useful on Eve? Just before I started, they were used everywhere and then hit with the nerf bat so hard that they were effectively useless. I suspect the changes to surface thrust changed it back to at least useful *somewhere*.

97% recovery makes it sound hard to improve costs by adding kickbacks, but SSTO rockets tend to be much bigger than two stage. Unfortunately the American holiday this weekend won't let me play much KSP, so who knows if I'll check this or not.

Remember, if you are targeting old hardware (you can do a lot more with a soldering iron or wire wrap at a few MHz, GHz is right out), you should be able to find things in a library (also hopefully significantly marked down in a used bookstore, but that might be better for browsing and reference books). You might want to try an inter-library loan or something, ask a librarian (I know there was a copy of Bebop to the Boolean Boogie in a library 30km south of me, but I doubt there are any in the system I have a card for). Also note that costs of building the hardware can be all over the place. The cost to pay the engineers to design such a full computer board (even for fairly small computers) can easily be over a million dollars, so "reasonable support services" can be very pricey, but worth it to those who are happy to shave a few days off the budget. Don't be to fixed in assuming there is a certain way to build something (some of those "pricey" services have a lot of competition and you might afford to build a small board done in a modern way that you could never troubleshoot if done by hand).

If you can plug numbers into a calculator (or use a sliderule, I'm not picky), you can usually find out how much delta-v you have. I like side-dropped kickbacks, which make the calculations difficult, but could probably come up with a descent approximation. I would much rather know the delta-v as I quickly built a rocket like the way KE tells me. I suspect most players will want mods. They will hear about delta-v and wonder why they are allowed to know how much delta-v a burn takes (and how much is left) but not how much is left in the rocket. The will see a youtube with scatterer(?) and chatterer and wonder why vanilla KSP is less exciting. They will want that *one* (or more) part that is missing. But really, I suspect that the worst of these is the delta-v. It is a common refrain for consoles. I can't begin to recommend one of my favorite games of all time (Elder Scrolls V: Oblivion) because it is so incredibly broken out of the box and *needs* mods to fix it (there are also tons of mods to improve it, but the character advancement is *broken*. There are also plenty of bugs fixed by the community). I suppose Squad merely has to look at Minecraft. If that (extremely mod heavy) game can live without mods, presumably so can KSP. I just can't recommend playing it there.

Two issues for SSTO rockets: In career, that's a lot of money you have to have in reserve for your SSTO (sure, you'll get it back, but you have to put up the payment first). Second, and most importantly: how close to you land to the pad? With mechjeb, its too easy. You just hit "land on pad" and collect your 100%. I'm curious how many players can hit close enough to the pad that they wouldn't benefit from dropped kickback stages (bunched onto two or less decouplers, natch). SSTO rockets (or S[l]STO, Single liquid Stage to Orbit) have the advantage that you only have to perform one landing. SSTS/TSTO (single stage to space, two stage to orbit) allows recovery, but requires two landings (a docking isn't faster than a landing. Even thought they will be *very* close). This makes them inconvenient to fly, if easier to make. Maybe just use stage recovery (or flight manager and don't bother getting the first stage into orbit).

Oddly enough, I've often wanted a console controller for docking. Not sure I'd use it for anything else. Count me as another one who will certainly recommend the game for PC(/Linux/Mac/whatever) over consoles if it doesn't have mods. The idea that this game will likely sell better on consoles without mods just feels weird. Prepare to explain (over and over) how to do the rocket equation with a calculator. Prepare for gamers to jump ship when they hear it (the fact that most PC players, including myself, prefer KE/mechjeb to do it for us (or just don't bother with delta-v) leaves us no room to complain about that).

My understanding is that "rocket candy" is the only solid fuel that can be made in moderate safety (that is, proper precautions and avoiding non-obvious dangers will allow you *a*chance* of being safe). Pretty much anything else and you have a large chance of blowing yourself up. This pretty much goes with any homemade explosives (I'm not sure if one or two of Alfred Nobel's sons blew themselves up, but that was a pretty good survival rate for that profession). No idea if "off the shelf" model rockets exist in the UK (I know they were in danger in the US after 9/11, and I remember pointing out how dangerous they were when attached to a Ford Explorer [2000ish US news media joke]). I'm pretty sure you can get ammonium perchlorate filled rocket engines, but I suspect you need some sort of license (tied to amature rocketry, more of proof of competence) for at least the bigger ones. If even the gunpowder-filled ones are available, they are likely your best bet until you are ready to go with hybrid nitrous oxide/rubberish fuels (and I'd still assume plenty of model rocket stages if you want that last bit of delta-v).

No SRB love? Go check the "cheap and cheerful challenge", those beasts are nearly all kickers by mass, and cheaply get mass into orbit. So three cheers for the kickstart, the cheapest way to get delta-v and TWR on the launchpad. Once in space, terriers never go out of style.

If I had attached a switching power supply to a microwave-based design that was looking for power levels deep in the noise, I'd be smacking my forehead hard. While the noise that goes along with the power won't be close to microwave frequencies, things can certainly get amplified and monkey with results. Better run it again, with a battery. Even a chain of linear regulators and filters are unlikely to filter the power down to the minute issues that you are looking for (well maybe, but when you are challenging the conservation of momentum and/or similar principles you need to nail *everything* down).

The first thing I would suggest you learn about is wire wrap. As the name implies, it is a means of connecting wires by simply wrapping the exposed conductor around a terminal several times. Quick to attach and de-attach, it is a favorite means of prototyping (especially at the frequencies 6502 and Z-80 operate at, much less useful with modern gear). I've seen plenty of older computers that were *shipped* using wire wrap, so it isn't just a "cheap hobbyist" method. While wire wrap is great, expect to learn how to solder. It takes a little practice and the basics involve heating both sides hot enough so that when you apply the solder to them (not the iron), the solder melts and flows on them. Don't get the leads to chips too hot. A good solder joint is shiny (well used to be, I'm less sure with modern silver-based solder). When I was introduced to professional electronics, my professor forced us to by textbooks (which seemed expensive at the time, but nothing compared to now), mostly to assign homework. We were told that the text book was sort of right, the book containing the data sheets was supposed to be right, but the actual computer (much like the cat in psych) was *always* right. A quick test is often easier to do and harder to misunderstand than reading through a datasheet. Still, you need to know how to do the test and a higher level book is a good place to start. Books from that era are almost certainly out of print (although "Soul of a New Machine" is probably good for inspiration, and almost certainly available). Go through a used bookstore and see if you can find anything that looks good from that era (I scored a "TTL databook" once. While absolutely nobody builds anything that it describes, the current stuff all has the same pinout and is largely backwards compatible). One thing that should cover the basics is https://www.amazon.com/Bebop-Boolean-Boogie-Third-Unconventional/dp/1856175073 (note this is third edition. The second edition lists a $12 used book (might not include shipping) and is certainly written since 2000, thus not being a problem). For PCB design, I certainly would recommend downloading the Eagle CAD package. While it completely lacks the polish of professional CAD, it should be able to get the design done. You can also build something like 10cm by 10cm boards (check before starting) without paying anything for this software (and smaller circuit boards will always be cheaper to have manufactured anyway). Assembling a circuit board can be more tricky. One small startup I worked at had built circuit boards by manually placing parts on the board and then cooking the whole thing in a toaster oven. I never got the hang of placing the tiniest components, but an earlier guy could. Note that there are a lot of hidden costs in this, as you need things like solder mask stencils to have something to stick you components on and cook (and don't use a toaster oven for food that has cooked lead). If you want to get really creative and build your computer from the architecture up, I'd recommend looking into FPGAs. In any event, I'd look into FPGAs and CPLDs if only to reduce part count. Basically, this is a way to design your own chip in "software". Typical ways to program involve languages like Verilog and VHDL and the chip vendors probably have *something* you can download and convert into the necessary bitstream (although they will be designed primarily to work with some seriously expensive software). Also learn what microcode is/was. Microcode basically lets you program a pile of chips to be a computer, and was the basis of computer design until things like Verilog and VHDL could compile down to a whole chip (microcode took up a lot of a few of my electrical engineering classes at school and at least on professor quoted a computer designer as "never making another non-microcoded computer". Microcode was effectively obsolete by the time I graduated, but still used in the dark corners of modern Intel machines for all those instructions that haven't been used since the 90s). If this seems like cheating and not sufficiently "hardcore" compared to using your soldering iron, don't forget that the pros in "Soul of a New Machine" were half "microkids" who were doing the microcode and the hardware guys were slapping down "PAL here" (a CPLD forerunner) every time they needed something complicated. For overall computer design, as far as I know "the book" is still Hennesy and Patterson: Computer Architecture, a Quantitative Approach. Note that it is used in graduate level classes as well as upper level undergraduate courses, so you might be spending some time in the first few chapters furiously trying to build up the background. It isn't cheap, but sufficiently advanced libraries should have earlier editions (I'd expect current ones to be checked out by students). They also wrote an introductory book called Computer Organization and Design, and if you can get it first I'd recommend it. Unfortunately I bought it after studying the first and was greatly disappointed (so I really don't know if it is any good or not). And don't try to build anything huge as your first soldering project. The lessons KSP teaches you about fake rocket design carry well over into other design fields. Make sure whatever you build is small enough that you can figure out why it doesn't work. When engineers first apply power to a design it is called a "smoke test". If no smoke comes out of the device, they are happy. They often don't expect it to work. Then the real work goes into discovering exactly what it is doing and adjusting the design/manufacturing so that it does what it is supposed to, and iterate from there. If you have multiple errors in you device (design flaws plus soldering globs plus unsoldered pins) it will be vastly harder to troubleshoot than a single failure (which can often be divided and conquered, no matter how invisible the error). PS: I remain convinced the the "best" architecture for that era would have been a stack-based architecture with a fairly large DRAM array on chip. The point of a stack architecture would be for the single addressing (both one source and destination would typically be top of stack) so that a single word would come off the DRAM array (you can only use DRAM if you can deal with only one port). The DRAM array would be useful for other things, as an instruction/loop buffer (probably too small for a real cache), for a stack buffer, and (tiny) data cache. Not sure if fast page memory would be early enough to make it make sense (it needs *some* form of burst memory reading), but it would certainly hit a brick wall around the time multiple issue RISC designs came out (early 90s), so perhaps we are better off without such a design.

Cost? You ought to see the size of "big boy" model rockets. If a hobby attracts a certain type (often middle-aged men. Competitive middle aged men) they will spend enormous sums of money if only to show they can. Getting to orbit might be like fielding a yacht, but don't expect cost to stop everyone (although the "amatures" might be like John Carmack instead of Musk and Bezos). Balloon launch? The less atmosphere you fly through, the less you are fighting the scaling effects of aeronautics. If your ISPs are equal (they're not), you should get the same mass ratio as the "real" rocket programs (thus launching nanosats instead of things measured in tons). Don't forget you can optimize the nozzle for [near] vacuum, that should help a lot (unless you need a lot of stages, which I suspect you would). Additional legal issues. While the idea that someone could launch an anti-satellite device at will, you have to remember that from 1945-1989 several nations were concerned that someone could launch a missile from *anywhere* (such as from a submarine) and destroy multiple cities. I'd expect that a means for instantly detecting such a rocket are already in place.

Except that any thermal generator will have the same Carnot limits (because there theoretically exists the inverse heat pump). Getting those last few percents near Carnot are likely to get *really* hard. Not that the maintenance crews wouldn't *love* to get rid of high pressure steam. I'm not holding my breath more a replacement (and the idea of basing a power plant on nanotech seems like distant science fiction. But it sounds like reverse-osmosis water treatment is being built on similar scales, so who knows) While I certainly try to pound on cooling issues (cooling is a big thing with electronics on *Earth*. Now try it in vacuum), mass issues are probably going to keep solar panels being used in favor of RTGs this side of Mars, and don't be surprised if you see large reflector use. The plutonium supply is also an issue, but I think NASA got the DoD to whip up more plutonium (they were carefully rationing it for awhile).

You're claiming that something involving "Advances in nano/micro fabrication" and "carbon nanotubes" is easier and cheaper than steam? Than 19th century tech (ok, high efficiency, high pressure steam was *expensive* 19th century tech, but still). Maybe it will happen someday, maybe not. But large scale steam is pretty efficient, and well known. Except that nobody is paying for the walking to Mars trip. And any extra delta-v you spend to get there, you have to spend again to capture (aerocapture with Hohmann speeds is controversial. Don't even think of pulling it off with "fast Mars" in real life). And not only that, nobody is planning on using VISIMR for plain old Hohmann (or faster). But if they had it, you could at least do a Hohmann. It is more or less exactly what you want to get to Mars. Ask yourself why nobody bothered to do a fast trip to the Moon? Presumably anybody could eat 3-10 times the delta-v just as easily.

Why use solar panels? You have low gravity, zero wind, and an available heatsink. So concentrate the solar and use a heat engine. How much solar can you get out of 1t of mylar? Nukes in space don't have a great history. Mostly because of the efficiency issues of NTGs. I'd also really like to see how to build a fission [or fusion, for sufficiently far future] reactor with a heatsink that relies on blackbody radiation (nuclear thermal rockets were cooled by the fuel. Making power without using *lots* of hydrogen is another story). Building a nuclear reactor on the Moon might make sense, but I'd have to ask where all the current Earth tech is that is ready for such a leap*. Spacecraft and rockets are made out of aluminum since they are made at the bottom of an 9km/s gravity well. New designs are likely carbon fiber, and I'd expect more bits of titanium sooner or later. Build outside of that well and cheap asteroid steel suddenly looks better than somewhat more expensive lunar aluminum (unless you really need to lose that dry weight. Which might happen more often than you would think). * I have this crazy vision of using comets to create a "cooling reservoir". Basically the requirements would be to not boil it, and to be able to stabilize the reservoir's temperature to useful levels over the entire lunar day/month. Possibly put the reservoir in a deep underground cavern carved by a nuclear explosion (how quickly would it evaporate in a crater?). I suspect you need to want a *lot* of power before you will give up solar as the main source of space power, at least this side of Mars.

Because your electrical propulsion system won't lift off of Mars? Because solar power can be generated cheaply on a [dwarf] planet surface in ways not in other places in space (no wind bothering huge reflectors, ability to use the ground as a heat sink, no atmosphere interfering with the solar energy)? Because you can't bottle up that solar power and send it to the asteroid belt? No idea what the demand for aluminum as a structural metal would be (already in space), I suspect you can get high grade steel (expect to bring a few alloying metals, "starmetal" only was a great advantage vs. copper and bronze) from asteroids. And you will almost certainly mine the asteroids first (and thus have the harder time disrupting the market for space metals).

TL;DR - PB666's response. Huh? There are a ton of problems with "fast Mars missions". The biggest issue is that if VASIMR can get out of the lab, it makes "slow Mars missions" possible nearly the next day. Then we can worry how to speed up Mars missions. You need extreme power. The proposed solution sounds like it would fix a significant number of the world's problems. Even on this forum, I'd put Mars missions far down on the list of things I'd do with a high efficiency nuclear engine (no idea if the non-nuclear bits can be produced cheap enough to crank up the efficiency of hybrid cars by 25-50%). You need heat management. This doesn't begin to cover the issue that every Watt generated has to leave via black body radiation. Trying to maintain life support on a high power system cooled by black body radiation makes life interesting (and more power, which requires more cooling, and so on). And all this ignores the fact that if VASIMR can get itself out of the lab it might make even slow (Hohmann transfer) missions possible. Right now we have: Fast: requires magic. Might as well require warp drive. Slow: Hohmann transfers. Takes months, and hundreds (I suspect thousands*) of tons of [chemical] fuel (in orbit for $millions/ton). Nobody has yet found a source of funding to get this done. If VASIMR can scale up to provide this level of TWR (basically provide a bit more than 1000 m/s delta-v per Mangalayan maneuver orbit, or at least ~1100m/s on the last one), it will be *huge* in getting us to Mars. Even discussing "fast travel to Mars" makes no sense if you remove most of the issues with "slow travel to Mars" and replace them with huge new issues just for high speed. Get to Mars first and then work on going faster. Mind boggingly slow: ion engines and gravity transfers. Cheap, but way too slow for human transfer. If fuel and supplies travel this method, things might happen. * I think you need a hundred tons of xenon just for the slowest trip possible for 100 tons of life support (which is the current NASA plan).

Do you have any more science in the lab (i.e. did it tell you to stop when it filled up the first time)? If so, keep filling it to 500 (or 750 if you upgrade to 1.1) to increase the rate the scientists generate science. Just review science and hit the yellow button to convert science to data.

Did you use identical drop tanks for a reason, or just didn't want to deal with the idea of 10 stages made up of different drop tanks (Squad keeps tanks at remarkably similar efficiency)? The reason you should ignore mass ratio (except as an issue of total mass/cost you put on the pad) is that they need to be averaged as logarithms, just like Jeb and Tsiolkovsky intended. Equal delta-v isn't guaranteed to be optimal (nobody has shown a general solution, and I suspect there isn't one short of a truly hairy beast that takes aero into consideration), but it is typically one of the best places to start.

Deleted

I can't imagine the issues in retrieving it. Maybe if NASA wants to do another test with an inflatable heat shield they could use the HST as ballast for the way down. The next closest mission I can think of would be mining experiments, but I'd assume that any mining returns would be wildly more compact (basically a huge ingot, or possibly shaped ore) and then use litho/hydrobraking. I can't see anyone funding that (the inflatable heat shield is remotely possible, but would have to already be almost ready to go when the gyroscopes failed. So don't really think it could happen). If you just want parts that have been in space that long, wait a bit longer for ISS replacement parts (because it hasn't been up as long) and examine them (assuming they went down safely). Of course, these were in a much more benign environment, but I'd expect plenty of parts on the outside to keep the whole thing going, and they will probably need replacement more often. Finally, I'm pretty sure that quite a few parts were replaced and came back from HST on the Shuttle's last mission there to refurbish it. They should still exist (and have already been examined, I hope).

Which isn't a problem for an 8m rocket. It is also a much smaller issue than "run for your life explosions" (not sure what happened on the ground at scaled composites). Here is an attempt by a technology/rumour website to field a balloon-assisted rocket into space: http://www.theregister.co.uk/science/lohan/ They moved from the UK to the USA in hopes of a less regulated environment, but are still sitting around waiting for the FAA/NASA to grant approval (I think the FAA is the problem). Oddly enough, if can really stick something into orbit you are under NASA regulation and the FAA has much more limited influence (it may also help to be buy a launchpad at KSC, not sure if those rules follow in launching at non-NASA locations). Somehow it might be interesting trying to declare an 8m rocket when traveling to Trinidad/Tobago or other locations unlikely to have space-based regulations. Launching in other countries might be easier. Building/moving the rockets there is another story (and anybody capable of managing an ITAR issue might not be able to work again if they ran afoul of those laws).

I'll have to check this. Several editions ago it didn't seem to work (kerbals can only carry one of each type of experiment, so unlimited as long as they have unlimited different experiments. Of course with five biomes or less, kerbals + seat weigh less than a capsule. Other tricks: Having trouble keeping your capsule falling protected side down? 1.25m structural fairings are cheap, light, and easy to stack raising your center of pressure without raising your center of mass. "Leadership Initiative" emphases "boldly going where no kerbal has gone before" over contracts *and* gives up to a 50% bonus to "field science". Two gotchas: it completely nerfs contracts and requires a huge reputation to be allowed to start it (especially for 100%, something like 700 reputation). Plenty of ways to pile on the milestones (especially if you took the leadership initiative above). Take a spacewalk in a new SOI, that's a big milestone. Transfer crew? Another milestone. Dock two vessels, another milestone (you could just pop a multi-part vessel apart). Have crew modules on each vessel? Bigger (building a space station) milestone (this can be done on planetary surfaces as well). 1.1 Appears to be more favorable to igniting SRBs above KSC. Still expect this to work only if total weight>srb weight (for upper stage SRBs).

They do. Not so sure about Blue Origin (note the recent "two parachute" landing), but SpaceX lands the capsule with parachutes (at sea. Presumably because NASA said so, although I'd be surprised if they wanted to bet lives on their landing tech). Landing BIG craft with parachutes gets even worse.

Paper spacecraft are always better than real spacecraft. STS (1972) made 100+ flights. Being unbelievably expensive meant that Congress would face a huge "sunk cost fallacy" in trying to kill it (and probably lead to all those flights). Be it politics or marketing, getting people to pay for things changes how things get done in weird ways.