wumpus

I wouldn't call it the "best engine" (although for non-recoverable launches it should provide nearly all the thrust in first stages of Kerbin, it is *clearly* the best at that role). I honestly think I would be more upset if somehow kickback bugged out (and I couldn't revert for some reason) and had to be replaced than the Nerv. But the NERV is likely the "best engine": I'd hate to have to haul enough fuel to do any space work with poodles. I put a nerv-based rocket up early in career mode and was so happy with it never came down (just kept hitching to it and refueling). But you better believe all those rockets to the NERV-based shuttle got most of their delta-v (nearly 2000 in the first stage) from kickers.

The other issue is cooling. You might wind needing to pump a significant amount of heat out to your radiators (which would then be getting considerably hotter than your spacecraft) during battery use (I'm assuming they are used for the Oberth effect, as initialing described above. Other issues wouldn't seriously effect the engineering). Oddly enough, the NASA research suggests using capacitors for pulsed fusion (I haven't had time to read the paper, so I don't know if Fe-based batteries are an option). I had no idea anyone who was working on such a thing used actual hardware (for at least fusion tests, I doubt they really bothered where the electricity came from). Plenty of radiators, though. http://www.nasa.gov/sites/default/files/files/Adams_2013_PhI_PuFF_inProgress.pdf Note: this assumes a "five day burn" to Mars (for a 90 day transit). Presumably with the efficiencies of a Mangallayan maneuver you might be able to turn it closer to a "two day burn" that takes 10+ days to get to Earth Escape velocity (followed by a one day burn with no further efficiency gains). You can take all the time you want for the first 3000m/s to escape velocity, but if you can't do the last 1000m/s in one burn, you don't get the Oberth effect. I'd also expect that for large values of power, "batteries" would become "similar battery chemistry" flowing through a fuel cell in a way that can be recharged. Presumably this would optimize all the packaging constraints of keeping the electrodes next to each other and whatever materials need to separate them. Considering the speed battery technology is moving (nothing like billions of people who all want a thinner cell phone to drive the tech), it might take awhile for this to be an option (so far fuel cell tech keeps getting passed by battery tech. Once battery tech levels off, I suspect that fuel cells will be developed with superior J/kg. While this might take awhile, I'm not holding my breath for pulsed fusion either.

I'm pretty sure you could substitute a corvette (LS3/LT1) engine reasonably well. VTOL power (for sufficiently light aircraft) and similar power over long stretches (Lemons* data implies that hacking the autopilot to allow hundreds of horsepower for long periods will kill the engine fast, at least without expensive upgrades). More likely you would just use electric motors** and draw from a capacitor (or more likely an Fe-ion battery) for takeoff landing and a generator (or Li-based battery). * The race, not cars brought for repair multiple times. The few times teams who managed to bring in worn LS1 engines did not work well (with the implications that it was the engine), implying that such engines really can't be expected to run close to the "hundreds of horsepower for hours" that naive plane makers hope for (I'm pretty sure that homebuilt plane makers found this out a much harder way than the lemons teams, I just haven't seen the data). ** I'd certainly expect the innovation for this type of thing to happen in China and/or India. Basically somewhere that doesn't have an "FAA" with very ossified views as what an aircraft is and how to regulate it. The bit with multiple motors is to avoid the requirement for a "multi-engine" pilot's license while still having redundancy (that and electric engines don't suffer such ghastly inefficiency when running at less than 50% power).

Huh. Some quick googling implies that even "cheap magnifying glass" lens material might not be cost effective. I was expecting that to be much cheaper, like "included in a cereal box because it's cheaper than cereal" cheap. I still suspect that in quantity it would make sense, but only for PV systems not at all constrained by roof size (probably a minority of solar usage).

Don't forget that doing such just moves electricity generation from peak need (because air conditioners will be trying to cool buildings that are being heated the same way the furnace is) to less than peak need. Not such a great idea (although probably necessary to build generators at a reasonable cost). I'd be more interested in reflecting mirrors into (PV) solar panels. Since Hawaiian panels get at least twice the power as (non-desert) continental US panels (or at least where I live), I'd assume that concentrating 2-4 mirrors into one PV would make sense (that you can find mirrors cheaper than PVs). You might have to move the PVs (which would be a pain/cost/whatever) or wildly overprovide on mirrors. Or you do the same with some sort of cheap fresnel lens (again, expect to "overprovide", especially if you plastic filters important wavelengths), but this should be easier to move (or sufficiently overprovision that it doesn't matter). Note: if you go the "overprovide and don't bother moving" route, make sure that the concentrated rays that miss the PV aren't melting anything. Either run hot water lines (for cooling) or reflective surfaces.

More like rule out such worlds for not having life terribly similar to Earth's. That one stepping stone between RNA reproducing it self and fossil life exists doesn't imply that it is a mandatory step, just that if it exists, it probably will be taken on some other planet (we don't know if hydrogen-eating life out competed other life on Earth or not).

Some notes: I trust google's deep pockets far more than Tesla's. I think even Toyota, GM, and Ford will think twice before making the wild claims Tesla does. Tesla eats the risk because they need to keep growing, and the risk of the massive lawsuits outweighs the risk of standing still and being caught and passed by the big boys. For a long time, NASA would build two space probes on the basis that "two-offs aren't much more expensive than one-offs" (the design being the biggest expense) and launching both in the hopes that at least one will survive (see mariners, pioneers, voyagers, etc.). I suspect that the Delta program lead to fixed launch costs lead to ending this. Redundancy is still an option. I'm not even sure why the quadcopter is more popular than a tricopter, but moving to a pent-copter should survive an engine loss (theoretically you should be able to do it with a quadcopter, but it probably puts to hard a load on one of the engines). "Gliding to safety" implies that it can find a landing strip (for values of "landing strip" including the Hudson River), I wouldn't always assume such (hopefully the navigation computer has pre-computed flight plans that are always with x-km of a "safe landing area". In the end, I remain a big believer in using lift to maintain flight. Toy drones might make the quadcopter design popular, mostly because it simplifies the issues of helicopter design (no need to angle the rotor), but the efficiency of non-fixed-wing flight remains an issue. Speaking of "non-fixed-wing", I'd love to plug the Scorpion (yet again): that was an (extremely) short takeoff and landing "drone" (old school drone, 2-4m wingspan depending on which prototype you were looking at). Overall design by Burt Rutan, detailed engineering by Freewing. Simply and efficient, it seems a great loss that it never went anywhere. In "passenger drone" configuration, it would also [still] not worry about complete lack of pilot visibility during takeoff/landing.

Is this meant to be on Mars? You simply skip most of that humidification/dehumidification process by merely choosing a non-desert climate on Earth and tapping a river filled by rain (a natural process that does all that for you). Also the subject is rather misleading. "Fuels" often imply an energy source, what you describe appears to be an energy sink (or more accurately, a low efficiency form of energy storage). Obviously for lifting off Mars (or similar), it might not matter how inefficient your process is, as long as the end concentrates enough power to have TWR>>1 And as far as "keeping it to engineering", I'm glad I wasn't the one responsible for the debacle that caused my Chief Engineer's livid rant about "price is a spec"*. If you can't get the price of this fancy system under the price of an oil well, it won't be a success. There will be oil wells still around once there exists enough power from other sources to power this expensive process. The only hope for this process is that the desire for fuel (for aerospace mainly?) and chemical feedstocks (and plastics if they use different wording) require more petro-products than the remaining oil wells can supply. * You better believe heads rolled. And more than a few laid off who had nothing to do with it, merely because of the lack of money it caused.

While I keep having visions of the eventual elliptic orbit and week-long-charging cycle (out to LTI), I'd guess that a two-three hour charge cycle combined with a 15 minute discharge cycle would probably work best (and the charging cycle would have to deal with "day/night" issues). I also suspect you will spend even more time in the Van Allen Belts (especially in the outer belts, as the craft struggles to escape with a fixed amount of delta-v per orbit). The biggest problem with this discussion is the idea that you can increase thrust by an order of magnitude without losing efficiency (well, no more than 50%). That seems to be counter to all available engineering on electric-powered spacecraft. VASIMR makes a whole lot of noise about this idea, but seems to willing to handwave away *all* problems, many of which would greatly change current practice elsewhere (long-term storage of hydrogen, cooling in space at massive power levels, and just how the Isp varies with power. None of them seem to be mentioned).

It has to be at least 320 m/s delta-v. And you can't EVA until you are sufficiently "close" (on rescue contracts. If you stranded your kerbals this isn't a problem), even though the jetpacks probably *have* 320 m/s delta-v (keep at least some in reserve to get in the capsule). As I mentioned, I could do it. I wouldn't want to try it around the Mun where it would be at least twice as expensive (according to the subway chart that I taped to my monitor around beta).

Yep. LESS. I meant less. Can't believe I typed more. Luckily I have coffee now.

If you are dealing with career mode silliness [part limits], I'd ditch the aero caps first (but not the fins). If you are trying to reduce gravity losses (by going fast), I'd recommend making the rocket more aerodynamic. Note, aero effects should be much [less*] significant on more massive rockets. Remember, aero drag is proportional to frontal area: 1.25m = 4.9m^2 2.5m = 19.6m^2 3.75 = 44.1m^2 Yet very rarely is a 3.75m rocket only ten times the mass and thrust of a 1.25 rocket. * thanks, Red Iron Crown

It certainly accounts for fuel useage over time, so it isn't really an instantaneous issue. It does ignore gravity losses (such as going to orbit from Kerbin), so the bits when TWR<<2 is going to "cost" more than twice the delta-v the rocket equation gives you. Once you are going effectively sideways (and your TWR is more a matter of getting into orbit before falling back down) even this doesn't matter. There is also the aero drag on the vessel, which is much harder to calculate (note how long it took Squad to even compute an effective instantaneous drag needed for a more accurate KSP).

If you are about to attempt to rescue *plural* kerbals from the Mun, be careful about the directions of the orbits. Once I decided to pick up two (likely from Minmus) when I discovered they were going opposite ways around Minmus. Luckily, I [had thought I] vastly overbuilt for that mission and manged to get them both, but reversing orbit is an expensive proposition.

There was a bit in Ignition where one fuel "created too much plasma" that would interfere with missile guidance (note that most of the fuels in Ignition are military in nature as HLOX and KLOX are pretty obvious. Getting ones that can sit on the missile-rail on a ship exposed to all weather* is another story). While I suspect that there is a certain amount of plasma, some fuels create more. It is a bigger effect coming down. I'm pretty sure Columbia was lost during the "can't communicate due to plasma" time coming down. * not sure this is the case, but it still has to be ready (i.e. already gassed up) and ready to go on the rail and stay there until the fire command is given.

The problem with this is that you are essentially asking for shell access from an ISP. Shell access was typical in the 80s, dying in the 90s, and dead by 2000. What you would do is log onto a machine (nearly always UNIX) and run the machine from your Apple (or PC, or whatever). Later, SLIP was developed (which was like an almost-working PPP), but my school only had one of those and lots of shell dialups. Eventually PPP was developed (which pretty much meant "real dialup", or full-fledged internet over the phone) and that took over. I think netcom would sell shell access for dialup into the 90s, but I don't remember it being an option when I was a field engineer traveling all over the US in 94-95. I used netcom because they had local ppp access available from all but one place I was sent to (Hicory, NC). About the only way to get shell access nowadays would to buy a general server share that didn't expect you to use specific means to sever webpages (and *of course* they wouldn't provide dialup, you would still need a TCP/IP stack. You just wouldn't need to parse all the HTML and similar). So the minimum computing power is likely going to be something that can access a full TCP/IP stack across a Starbuck's WiFi. And since I've noted that you are going to be able to get two GHz ARM chips at least as cheaply as anything else (they will likely come with an LCD control, not sure about the Atmels), you might as well directly connect with your computer. Whether it is inventions or "why can't I buy x", the answer is never about the stuff itself. Technology isn't gadgets, tools, or other parts. Technology is infrastructure. We don't have the infrastructure (there might be spotty exceptions, but I'd like to see if the odd dialup ISP still has shell accounts) to do what you are asking.

Sounds like OLPC (one laptop per child). Its biggest "success" was in Uruguay (where they managed to get every kid one for about $21/kid). Unfortunately it really didn't seem to have much of an effect on learning. Which is rather bizarre, it seems to me that if nothing else, it should make an a mind boggling e-book for such kids. Perhaps there isn't enough Spanish works in project Gutenberg (although the South American literary scene is rather strong). The OLPC webpage doesn't seem to have been updated since 2012. I'm guessing that ~$20 laptops/tablets can be had from purely commercial sources much easier.

Even worse for a Z-80 is that it has a built-in memory controller. So not only can it only access 64k of memory, it expects to see old-school dram chips (FPM will do if you can find any small enough). Some better choices. Atmel jobs: these (and the PICs) are about the cheapest you can buy in a microprocessor. Most are 8 bit, but they come in 32 bit as well (even the 8 bits address more than 64k, so don't expect a huge jump of power, but porting and compiling might be vastly easier). Famously used on arduino boards. I think you could get a CPU that could run an early browser around 2000 for $1 each (assuming you were buying 1000 at a time and were willing to deal with only having 8 pins total. You would probably pay more buying the components needed to connect to 8 pins than just paying a few dollars more for a "real" CPU). ARM: basic phone CPU. Serious power, low price. Also relatively "open" and produced in every variety imaginable (the $25 item I linked to has 4 1.2GHz ARMs in it. Note that the cheap stuff is typically in order, which won't be close to GHz to GHz what you are used to in a PC (this is changing, but don't expect to buy it for $5 anytime soon). Jaguar: an AMD-designed low power AMD86 (x86-64 to some). Takes up around 3mm on a .28nm process (a process with a great cost/transistor ratio). Runs KSP on the PS4/XBox1 (with 8 CPUs). Another one you probably won't see for $5, but only because nobody is willing/expects to make it up in volume (given enough volume, you could easily get the CPU (and plenty of integration) under $5, but getting the rest of the computer would be another story. All in all, the minimum price is more likely between $15-$25. The catch is that there is a certain motivation to build a $5 phone (to sell phone calls) that doesn't exist for a $5 laptop. Basically you would be building the same electronics (less the RF-phone specific parts) plus a keyboard. Expect a completely cheaped out rubber keyboard that is "carbonized/conductorfied" to short out bits of printed circuit board underneath it (and old trick. But don't expect to play games like KSP where reading multiple keys at the same time matters). And don't expect an LCD bigger than the cheapest phone screen. But you would probably at least have dual >1GHz ARM CPUs (that would be comparable to circa ~2000 PCs, and way cheaper to use than Z-80s).

I have quickloaded at least once to an inevitable explosion. Didn't happen right after I saved, but 1.1 decided to load from save with the engines off then ignite the engines. The stress on the rocket caused it to immediately destroy itself.

Airports already have NIMBY/noise issues. And my understanding of pulse detonation is that it is likely to always be too loud for public use. Wouldn't it also work if you provided the oxidizer as well (Vexhaust should be higher)? I find the lack of rocket companies attempting to use this tech making me suspect it isn't going anywhere. SCRAMJETS don't work below ~mach 4.5, and presumably pull past mach 10? (ok, the record so far is 9.4). That isn't air travel, that is suborbital ballistic travel. For seriously long (trans-oceanic) travel, it should make sense merely avoiding drag for the parts in space. But so far that type of speed is only available (for consumer values of available) from Virgin Galactic and Blue Origin (and they only get you halfway there). The price tags so far can't justify the speed. No idea if you could get the price down with regular flights (that mach 9.4 craft was single use).

I think on the "things I hate about KSP" thread, players were *still* complaining about hitting the launch tower on the way up, long after it was gone.

While there have been plenty of documented supernovae visible in the sky over the years, I don't think any were close enough to kill anybody. I'm pretty sure you can compute the odds of it happening to well over 1/Million years (for widespread fatalities). On the other hand, solar flares are reasonably common. Fortunately, unless they are directly at the Earth we don't have much issue. However in 1859 we had an incident that certainly shocked a few telegraph operators and would knock us back before 1859 (because we don't have telegraph or gas lights). I'm curious how much it would cost to have enough spare parts to quickly survive such an issue. To point out a simple issue, I'd expect 99%+ cars wouldn't work (only classic and cars stored underground would work, and getting more gas would be nearly as difficult).

I'm all but certain that the domestic Chinese/Indian market has plenty of examples of a <$100 laptop. I know that years ago AMD was designing a low-cost/high integration CPU designed to run WinXP (back when it was being discontinued, but vastly popular in developing countries). http://www.microcenter.com/product/453347/A741_A33-5G-8-7R (note this is $25 in stock local to my location). http://reviews.microcenter.com/3520-en_us/0448361/azpen-innovation-azpen-innovation-a741-tablet-black-16-9-lcd-ips-7-1024x600-display-allwinner-a33-1-2ghz-quad-core-cpu-512mb-ram-8gb-storage-android-4-4-kit-kat-os-expandable-up-to-32gb-via-microsd-card-reviews/reviews.htm (I think this is the same thing. Hard to tell with Azpen). Some notes about the tablet. I have one [a different model: this company seems to be the only to sell a 10" screen without a huge premium], the battery died in under a year (still have to order the thing, I've finally found a more reliable place than alibaba to order from), although that's what I get for buying "refurbished". While I'm sure you could connect a keyboard to the linked model for less than $75, the one I have has a "real" USB port (not OTG) which means that it connects to a $5 keyboard (no idea if there are $5 keyboards with touchpads to cover for the mouse, that thing is barely capable as a tablet and not for use as a laptop). So these things *do* exist (for values closer to $5 than $100), but you basically get what you pay for. Often "cheap computer" is wildly better than "no computer" (and for me, "cheap computer" typically beat "getting up and booting a full power desktop" for such things as surfing and watching Scott Manley explain KSP).

Since the technological issues were "solved" in the 1970s, any remaining issues are presumably economic. One fairly insurmountable issue is fuel usage. Consider the lack of SST for most military jets "just to get there" (as oppose to using speed to defeat or evade enemy action). Getting there "first with the most" is typically critical for military adventures, but as far as I know, the F22 is the only plane designed for long-term speed "to get to the action first" (it seems unlikely that the blackbird was expected to get to sites before anyone could cover them up, although it is possible). Since military budgets generally don't have the "chicken and the egg" issues of starting up a technology, I'd assume that the economic issues can't simply be solved with throwing technology at them. So the economic issues come down to "is the extra speed really worth it". My guess is that a significant chunk of jet passengers are fairly high paid employees flying on their employers' dime, it all comes down to roughly the ratio of the cost to paid (as salary) to the employee in fight vs. the cost paid to fly said employee. Right now these numbers justify jet flight (of course, management might unionize if you shipped them via greyhound) but not SST. People who can afford SST flight seem more likely to splurge on larger areas during flight and simply overlap the "downtime" via electronic work/recreation. My guess is that SST will happen when the (time saved)*(hourly cost of expensive flown employees) is less than added cost to fly SST. That may not happen for awhile. You also need to somehow allow for "overlapping" the downtime of sitting in what are effectively steerage-class seating (for above-first-class prices), VR headsets might help, but might not be popular while seating that squished together. Also note that while sonic booms pretty much killed the idea of domestic SST, TSA pretty much stuck a stake through its heart. Even before 9/11 (and before the OKC bombing), train was often the preferred means of traveling between NYC and DC (an obvious important "mostly overwater option" flight) and one of AMTRAK's only profitable routes: just getting to the airport adds a ton of overhead that makes any benefit from SST disappear fast.

A few more notes: While I learned to get to orbit long before beta (back when "gravity turn" meant "go fast, then turn East"), I still think rendezvous/docking is much harder than learning to orbit. Of course, I still remember the joy and wonder I felt as I saw the trajectory pop out of Kerbin, showing that I was going to stay up! Fortunately they seem to have fixed of most of the "going into space" difficulties (this was probably harder than orbiting in 1.0.0). I just tested a few means "of getting into space" and only an early attempt (with 250km height) got Jeb killed, the rest were hair raising, but safely landed (if you want to call a <1sec "safe" window to hit the chutes "safe"). I will admit that orbit seems much harder then when I learned. My preferred "no brainer" orbital craft (mainly uses a BACC single stage) has been nerfed into tricky measures (use launch clamps to get off the pad at an angle) and it took two tries to get up with a similar "no brainer" liquid fuel rocket (and even then had barely any delta-v to return). One nice thing is that with a decent (30km) entry pe, you at least have plenty of time to hit the parachute button (considering how much the player has invested into getting into orbit, it would be nasty to kill Jeb now). What makes orbit so hard is that you both have to build a rocket that will get there, and correctly pilot your rocket (the second wasn't a big deal when many forumites first did the job). You might want a "class built" (with at least some means of checking the rocket equation. If that means the teacher plugging numbers into a cell phone calc app, so be it) rocket to practice getting into orbit. Perhaps an unmanned job (make sure you include extra reaction wheels, those probes are wimpy) and then let the class build a rocket to get Jeb/Val into space. If you are teaching a class, it might make sense to show a clip (I'm sure there is one on youtube) of the early history of the Vanguard program. That 4" flight was one of the better ones (later ones did a lot of damage). Going into space means blowing up a bunch of rockets (spacex was both lucky and good: they only blew up three falcon[1]s before going into orbit. They also pretty much admit that if they blew up the fourth, they would have gone out of business).