cpast

Pilots are much, much, much better-trained than drivers. Pilots of airliners have extensive experience. Beating the average pilot is orders of magnitude harder than beating the average driver. Most landings are not autoland, because autoland is higher-workload than normal landings. Autopilots cannot fly a plane autonomously; they need monitoring, and don't even try dealing with abnormal situations (they disconnect if things start going wrong). That's part of how they can be so reliable in normal operation -- they simply don't control the aircraft if things have started to go wrong. A car has an easy stable safe mode. A plane, no.

Railroad locomotives have a brake that literally works like this -- you connect the traction motor to a resistor grid, and bleed off your kinetic energy as heat.

Moving a coil of wire through a magnetic field requires you to expend more energy moving the wire than you get out of it in electric power. Where would this energy come from? In a normal power plant, you get it by burning some fuel in order to heat water into steam which then uses the heat to spin a turbine; in solar-thermal plants you use solar energy to heat water to the same end; in hydro plants you drop water from a higher altitude to a lower altitude (the water is ultimately placed at the higher altitude by heat from the sun, which causes water to evaporate at which point solar-powered winds move it until it can precipitate at high altitude); in geothermal plants you use the heat within the Earth (some of which comes from compressing dust into a planet via gravity and some of which comes from radioactive decay); etc. Nothing creates energy from scratch; everything that produces energy requires an energy input or a mass input.

So your idea is that while humans are always faulty, humans can make perfect AIs? Yeah, that's not happening. Ever. There will never be a bug-free AI. All you can get is an AI with no *known* bugs that makes better decisions in general than a human; you cannot have a perfect AI. I view the claim that we'll be able to automate all decision making (and that this will be a good idea) as unlikely in any but the very, very, very far future, but it's possible; a perfect AI is not possible.

Actually, direct experience can be quite helpful here. While a pilot might not be expected to be an aerodynamics expert, they have a great deal of experience with paying attention to airspeed, altitude, thrust, and attitude. While argument from authority is strictly a logical fallacy, in practice it's pretty silly to reject "dude, I have actual training and personal experience with this" out of hand as a strict logical fallacy. I would absolutely consider a pilot more likely to know what they're talking about here than someone without formal background or experience trying to interpret things written for someone with formal background or experience. Also, FWIW, every aviation thing I've seen that explains aerodynamic details of stalls says it's AoA-caused. What causes a stall isn't that airspeed is too low to keep your altitude. Stalling is what happens when the boundary flow over the wing separates from the wing surface, causing a precipitous loss of lift. Lift very much depends on angle of attack. Initially, increasing angle of attack increases lift. However, at some point the angle of attack is too high and the flow separates from the wing; this is not smooth. As airspeed decreases, the AoA needed to maintain level flight increases. At some point, this passes the AoA at which the wing stalls; that speed is the stall speed, but all it means is the aircraft can't maintain level flight aerodynamically (it'll stall if it tries). - - - Updated - - - Yes, but then you're not in normal law. This is basically semantics: "normal law" on an Airbus is the mode with full flight envelope protection (that can't be overridden, unlike a Boeing where extreme force on the yoke *will* let you make a command past what flight envelope protection would allow) and where the sidestick doesn't directly command control surface deflection. If you turn off the flight computers, then a side-effect is that enough has failed that the aircraft can't reliably provide most of the more advanced FBW features, so it fails down to a different control law with less protection (down to the sidestick commanding deflection of control surfaces with no envelope protection). But this isn't something you're supposed to do in general, and wouldn't necessarily be something the copilot would think of; if you *just* push the stick all the way forward, the plane will still be operating with flight envelope protection.

"In normal law." Turning off the protections is only possible on an Airbus by forcing it into some other flight law. Which, since the standard descent rate was adequate for his purposes, there was no reason to do here.

I really can't see headset-based things replacing full-cockpit sims. As I understand it, most of the *point* of those is that everything is laid out precisely like a real cockpit; I can't imagine real-cockpit-with-VR-headset is cheaper than real-cockpit-with-screens-on-windows.

My guess is that they needed to remove the 64-bit stuff from Steam in order to push 1.0 through the experimentals channel without supplying a 64-bit version. I don't know if Steam will let you have an architecture available for only one channel.

But the dV *difference* between Mun height and a height 500km higher is less than the *difference* between the Mun height and a height 500km lower. Also, the orbital velocity decreases with distance, which wouldn't happen if gravity was constant force to the SOI boundary (orbital velocity would *increase* with altitude were that the case). It's wu Yes. The game computes either the gravitational force or the gravitational acceleration caused by the body at the center of your current SOI at your altitude (using standard Newtonian gravity). What changes at the SOI boundary is which body KSP looks to as the sole source of gravity (as opposed to real life, where a spacecraft can be affected by multiple bodies at once). Nope. Never. Not with point-mass planets and Newtonian gravity, at least.

My understanding of Airbus is even if you *are* flying with the stick, in normal law you still can't make the plane do an arbitrarily steep descent, because flight envelope protection will keep you from running too low-G.

Isn't that, like, the exact opposite of backseat moderation?

Needlessly complicated, ridiculously expensive, illegal at least in the US (you aren't getting the FAA to approve thousands of quadcopter flights in a tiny area; that's a big part of why Amazon hasn't actually *done* their quadcopter delivery, because it's illegal). Also likely to be unreliable. Advantage over trucking: not affected by road traffic (very affected by air traffic). Low altitude over a city is the *opposite* of safe. Manitoba has lots of communities very far from everyone else, without many people there. Ground transport is unreliable there. This is a niche that airships might fill. It's not a terribly big one. It's not one that is associated with factories, because you don't build factories in the middle of nowhere with no proper transportation infrastructure around unless you have an *extremely* good reason. But they *do* seek something that will be useful. As we've been trying to explain to you, airships aren't really that outside of special cases. Have you even *looked* at the numbers? The truck-ship-truck is an order of magnitude cheaper than what the people who are selling airships claim the cost will be. The secret is that trucks don't take cargo very far. You build factories where there's good infrastructure for your needs. Also, you messed up the truck-ship-truck route. You don't unload intermodal containers to move them between places. In the US at least, trucks often use 53' trailers until they approach a port or (sometimes) railyard where cargo is put in 40' containers, but then it stays in the same container until it reaches near its destination (where it *sometimes* transfers to a 53' trailer). As it turns out, this whole process is extremely cheap, in large part because the vast majority of the trip costs very little per mile. For factories with lots of things, it's not unusual to build a branch line straight into the factory. Then, this becomes load in container -> put on railcar -> put container on ship -> put container on railcar -> unload container at destination. Legitimate uses. And a textbook definition of "niche." And we're still challenging your absolutely unsupported assertion that the cost reaches a competitive level with rail and road. Again: The people selling this don't think it's cheaper than rail or trucking or ship. They think it can beat *air* for price; their view is that this can work where infrastructure is lacking. The trouble with that is that it's often cheaper in the long run to build the infrastructure, because it gives you cheaper transport for a long, long time.

Actually, that timeframe isn't preposterous. No trip *requires* 7 days by plane; much of a 7-day plane trip is spent going elsewhere, going to hubs, or sitting in a warehouse (I'm pretty sure you can fly from anywhere to anywhere in at most 36 hours from initial wheels-up to final touchdown). Freight can go way away from a direct line (hell, next-day packages via Fedex from LA to Seattle have a good chance of ending up in Memphis along the way). An airship at 120kts ground speed could circumnavigate the world in 7.5 days; while actual groundspeed varies significantly based on wind, you could probably pull off a 7-day one-way trip without too much trouble, and four days would be achievable in many cases. That said, I *do* question why truck-direct airship-truck is rated as cheaper than roundabout flight; the premium of a direct flight over an indirect routing is paying for the cargo to be prioritized over other things, and I can't really imagine a direct airship as being much cheaper than indirect planes (while you have more cargo to spread the fixed costs among, a direct routing involves paying a significant premium to get an expensive airship devoted to this one route). That said, one thing this could do well at is oversize cargo. Currently, the only real options for very bulky things is to disassemble them, or to put them on a barge, or (sometimes) to sling them under a helicopter. If this could handle oversize goods, that would be worth some amount of money. Indeed. Transportation methods vary quite a bit in details -- in addition to price-per-ton (using whatever packaging method is best for it), there's speed of an individual item from pickup to dropoff, reliability in terms of speed, reliability in terms of not losing the cargo, ancillary costs (e.g. facilities for handling it), how easy/hard it is to package stuff for the trip, how long it takes from realizing you need to send something until it can be picked up, how much of your stuff it can carry per week, constraints on individual items (e.g. max dimensions, max weight), etc. They differ in a lot of ways.

He's been bringing up the EU a lot as a possible target, and there are (AFAIK) no intra-EU customs (or maybe it's intra-Schengen, or something; the point is, a lot of Europe is a single area as far as customs is concerned, and you don't need to go through customs with stuff shipped within that area).

Those two clauses are in direct opposition to each other. Changing engines is the precise opposite of "optimization" and *especially* "bug fixing." The point of doing optimization and bug fixing is that you make minor changes to code. You don't do big things; you only change things if you absolutely have to. Switching engines creates new bugs. It's changing the code from something where the team has a pretty good sense how things work and what to do, and has incorporated all sorts of tweaks to fix issues, into an incompatible physics engine and a new game engine. It's not a simple task. If it were, this would be Unity 4.4 (I think). Likewise, the kind of optimization you do late in the release cycle is as unobtrusive as possible; big changes have heavy bug potential. It's entirely possible that, had this come out six months ago, Squad would have changed over. Had it come out in December, Squad might have considered investing some dev time into seeing how it works, and trying to hack a port together to see how it runs (and see if it's a bugsplosion, a won't-compile-splosion, or runs decently well). But it came out in March, well into the final push for release. At that point, no matter what Squad had said about U5 a year ago, it'd be *insane* to go all-in on U5. I wouldn't be surprised if they tried compiling against U5, but if there were *any* serious issues, you'd be crazy to put off your 1.0 to change the game engine when you're nearing release (keep in mind, Squad also had and has a much better idea of the status of KSP's development than forumers; I suspect they knew they were nearing experimentals).

The cruise speed of an aircraft (*any* aircraft) under a certain set of conditions (basically altitude/pressure, temperature, and engine power settings) is the *airspeed* at which thrust from the engines balances drag from the atmosphere. Mass of the airship is irrelevant; it doesn't determine either the thrust produced by the engines, or the drag force from the atmosphere (since we're comparing forces instead of accelerations). Put that together and you get the speed of the aircraft with respect to the air outside. Because the aircraft is suspended in the air, its ground velocity is its velocity w/r/t the air plus the wind velocity w/r/t the ground; the speed is an equilibrium, and aircraft reach that equilibrium reasonably quickly. If the aircraft climbs into a 60kts headwind, it will quickly have a 60kts lower groundspeed. Theoretically, an incredibly massive ship with low cross-sectional area might take a while to reach equilibrium again. Likewise, if it's *way* off equilibrium (e.g. re-entry), it takes a while to reach equilibrium. But at airship scales, it's not taking very long to do that. In cruise, any aircraft is equally affected by sustained winds. Where mass *does* help is for gusts. More massive things with lower cross-sectional area take longer to reach their equilibrium when the wind changes, which means that if the wind is rapidly shifting they won't be nearly as affected. However, airships do worse than planes in this respect (they have much higher cross-sectional area per unit mass). Also, the equilibrium doesn't necessarily shift as much as you might think when you increase the size of the airship. The greater payload is irrelevant; what matters is how engine power scales with size. There's a limit on the power you can put into a single propeller (you can put more power in by changing pitch, adding blades, making it spin faster, and making it bigger; the first has a limit, the second has seriously diminishing returns, the third and fourth are constrained by the tips of the propeller needing to stay subsonic), so it's unclear exactly how it would scale. If you're using the PV idea (which Aeroscraft isn't, incidentally), available power would scale with the square of the length (and thrust grows slower than power).

A dangerous trajectory like into a river? Or on a levee? For that matter, what if the pilot turns off the computers that tell whether it's on a dangerous trajectory (the pilot on an aircraft can turn off any electrical system)?

And it's a lot easier to fish when fish populations are high. And yet overfishing is a thing.

Also this. @AngelLestat - The thing about aircraft is that they are completely suspended in the air. While over short timespans the mass and surface area controls how much wind affects it, that only affects gusts; in cruise, your airspeed is independent of wind speed, because winds are fairly consistent. For things on the ground, there's the force of friction that balances the force of the wind; for things aloft, inertia is all that resists acceleration from the wind. It doesn't take long for an aircraft that flies into a high-wind area to be back at its cruise airspeed, which means that an airship that flies into a 60 kt headwind will soon have a ground speed of 60 kts less.

No, the niche is actually pretty small. Not compared to delivery to consumers, but compared to inter-factory delivery. Inter-factory delivery involves far more cargo than an Aeroscraft can carry. Argentina-Spain is a *terrible* route for this; ships have far higher capacity (a Panamax ship can hold over 70,000 tons in one ship), and you don't care about speed for manufacturing stuff (you care about quantity sent per week, not how long it takes one individual thing to get to its destination). At most, this can make sense factory->port; this is in a small niche between cargo aircraft (which are better for time-sensitive individual parts) and ships (which are better for lots of cargo). You can't handle heavy cargo on this; you have *medium* cargo quantities, which is a small niche. And again, you don't have any support of your claim that it's unclear whether a rail line will last 25 years. History is against you here; rail lines can easily last well over a hundred years. Tomorrow, I'll be taking a train on a right-of-way completed in 1894. Rail lasts. The infrastructure cost is high, true, but you're missing the nature of that cost -- it's *hugely* an upfront cost. Once you build a rail line, you have a right-of-way set up. You have a rail line. Those things last a while; maintenance is not all that expensive. It's an investment for the future -- rail lines last a long time, so your one-time cost will be giving benefits for a long time. This is also in no way an alternative to rail for most things. Rail is very cheap for capacity; this is because trains are big. This is not big; any claim that it scales up needs to be accompanies by engineering studies concluding it does that efficiently (outside of a game, you can't double the size and expect everything else to be multiplied by 8; you have to reengineer things). You can't replace rail with a 250t craft. It augments rail, but it cannot possibly replace it. For price, not even the people selling this think it can reduce cost to less than rail -- their ultimate goal is a quarter the cost of *air*. Air freight is 40 times the cost of rail per ton-mile. The claim for this unproven technology from the people selling it is ten times the cost of rail. And that's why countries subsidize rail (and roads) -- it's worth it to build transportation networks that will have lasting benefits for decades, because efficient transportation helps the whole economy. Numbers source? You're saying they improve *faster* than one would expect from simple volumetric analysis (1.2^3 is only 1.75 or so, not 2). As mentioned, you also have to be very careful applying rules like that, because you have to reconsider the engineering of the thing to support the new size. They *do* have a size limit -- the size at which they're too big for most facilities to support, and big enough that they'll not be able to fit in a hangar to repair, and big enough that they'll be heavily restricted by air regulators (good luck flying a kilometer-long craft around; the airspace is government-controlled, not open for anyone to do whatever they want, and regulators will have things to say for a kilometer-long craft). I can't see the actual people designing this claiming zero fuel consumption. The actual design is diesel powered. A PV-powered craft isn't going to do 120kts with this kind of drag. It's not going to do 60kts with this kind of drag. You will need fuel to get acceptable speed. In fact, all they claim is like 1/4 the fuel of an airplane; this is more than rail and ship. I'm not sure you know how niches work. There is no best mode of transport. There are different bests for different purposes. Here's what seems to be theirs: delivery away from existing infrastructure (our views differ on this because the US has a big rail network already and Argentina doesn't; this would be better in Argentina than the US), of medium quantities of cargo, at reasonable speeds. It won't replace ships for very much. It might be great between ports and places inland, especially in mountainous terrain, as a shuttle. Or for medium-distance transport or stuff that needs reasonable speed, but not next-day or anything like that. Or for oversize cargo (where currently barges are the only real option) -- this will be great for oversize cargo. It's not "average good" in most tasks; it's awful at dealing with large quantities of cargo cheaply, which happens to be among the most important shipping tasks. It's awful at doing last-mile delivery. It's not the "best" -- it can be the best in its niche, but a "general comparison" does not exist. That's like asking for the best material in general.

Airplanes don't use containers. In related news, airplanes aren't used for heavy cargo transport -- the lack of containers means they're bad for moving lots of freight. Trucks don't use containers in some cases when they aren't being used for intermodal operations, but a trailer can actually also go by train (even without the container, you can stick a trailer on a flatbed). You should know that "go between plants of a firm, carrying medium quantities of cargo at medium speeds" is a fairly small niche -- rail is actually quite good for that in countries with rail networks, because while an individual thing might take a while to get from one plant to another, you care more about the overall flow of cargo between the two. Now, I don't know about Argentina; it's extremely plausible that this could work out well in a place with poor rail and road networks. I'm speaking from a much more US-centric point of view. However, I will point out that building rail on the assumption it'll be useful for the next 25 years is a very safe bet. The US and European networks started being built in the 19th century. Technology doesn't change physics -- steel wheels on steel rails are low-friction, and the ocean lets you move huge amounts of stuff very cheaply and easily. In 25 years, rail and ship will be dominant modes of transit. In 100 years, they'll still likely be dominant.

No, it wouldn't. You cannot use this without truck support in general, because it doesn't support last-mile delivery. You have, in three paragraphs, first conceded that you aren't making the final delivery, and then forgotten that you aren't making the final delivery (and so you need to transfer to whatever is making the final delivery). If you're outside the city, you still need to transfer to/from trucks; nothing that can't do last-mile delivery can avoid the need to move cargo from one mode to another. You can't just use any open space, because you have to comply with zoning and air traffic regulations (that are somewhat particular on where you can land stuff). And Aeroscraft only claims 120 kts at 8,000 ft. Aircraft hit >400 kts at 35,000 ft (where winds are also higher). This will be slower than aircraft unless you have dedicated on-site landing facilities for it. But freight rail can handle much more cargo than this can. If you have a regular stream of cargo, you don't care how fast a single thing gets to its destination; you need throughput, not latency. Freight rail in the US is very widespread; many large shippers have rail branches ending on their property (there's rail service on-site). Trains don't, incidentally, use regenerative braking to generate power. They have dynamic braking, but dissipate that power in resistor grids. Why? Because you aren't braking all that much, when you are you're mostly using air brakes, and you already have a good power source (generators, or external electricity). A PV airship will not use fuel cells. That is a *terrible* efficiency idea. Fuel cells are bad as batteries. Electrolysis-fuel cell has really low efficiency, much lower than battery charge-discharge. So let's assume this is a sensible idea, which uses batteries. You don't have a PV-battery ship, you have a PV-powered ship. This isn't going to be fast, and it will be *very* expensive. There's a reason this hasn't been proposed by Aeroscraft.

Indeed, ships can be beaten in many cases. That's because, as I mentioned, the idea of the "ultimate" freight carrier is meaningless -- there's nothing that's best in all cases. Ships dominate for cheap high-capacity travel where the need to go on water doesn't impose too high a detour on the path (even if you could go via land, a ship can still be better for very large cargo; that's how NASA ships its largest rocket parts). Ports are not cheap, but they serve enormous cargo loads, and you don't really need that many of them (since you aren't delivering directly to the destination, the ship is only one step in the pathway). As for transferring cargo: In most cases an aeroscraft would also have to transfer cargo to land transport (as you mentioned). However, transferring cargo is one of the things that ships and trucks and trains *excel* at: if you don't support intermodal containers and aren't carrying a single kind of bulk cargo (like oil carriers or car carriers), your transport method isn't even worth talking about for cheap mass freight. The challenge for aircraft (including this) is that intermodal containers don't tend to play nicely with them. 40-foot intermodal containers tend to way between 20 and 30 tons (max weight is 30.5t), which means you can't take very many of them on your craft (and note that a 40-foot container weighs 4 tons empty, so if you load them lightly then the weight of the containers is a significant fraction of cargo weight). Compare to a train, which can handle 1 or 2 40-foot containers per car, or a ship. The reason is that cargo is heavy, and you need to deal with it after unloading or before loading. It's a cost for all systems; it's not special to this one. As for container handling: You wouldn't be loading/unloading stuff to trucks without ground infrastructure; you need stability when putting a container on a truck. Container cranes are not light; you lose a lot of cargo capacity if you only had 250t to begin with. You can't unload most cargo in lakes As for what places don't have flat areas: Most places that want lots of cargo, as a matter of fact. It's a bit moot, though, as since it's not built for last-mile you'd be going to a specially built spot anyway. Yeah, lighter-than-air stuff is actually *really* good against hostile fire. It's not pressurized as a rule, and it's big, so it loses lifting gas very slowly. But this *also* has a niche. Container ships' niche is massive cargo transportation from one place to another more distant place where using water isn't a big detour. This can't fill that. Trucks' niche is last-mile transport. This can't fill that. Trains do what container ships do, but over land and with more flexibility (because railyards are easier than ports). This can't do that; again, it simply doesn't have the capacity. The niche for this is medium quantities of freight, as well as oversized freight, at low speeds. It fills a niche kind of like barges, but over land and faster (less fuel-efficient, though). Other numbers I've seen put rail at more like 40 g equivalent CO2 per ton-mile, and ships at like 11 g (e.g. NRDC). The source Wikipedia cites is per ton-kilometer and gives *ranges*, and I'm not sure how faithful the Wikipedia numbers are for that. If you bring up hydrogen fuel cells, I think I can bring up electric trains. They're more powerful than diesel anyway (the reason they aren't universal in the US is that electrification requires maintenance, and lots of US rail lines are through the middle of nowhere), and have zero emissions just like hydrogen fuel cells. Like hydrogen fuel cells, everything depends on how the (electricity/hydrogen) is produced. (note: a fuel cell is a poor choice if you're going to fuel it via onboard electrolysis; actual batteries are *much* more efficient for this. Fuel cells are only good when you fuel them with hydrogen from elsewhere, because fuelling with hydrogen is faster than charging a battery). However, you won't run reasonable speeds on PV cells; the actual Aeroscraft will use diesel, because PV doesn't actually give that much power (and at the size of an Aeroscraft, you have a *lot* of drag).

This is incorrect. The US ran experiments for sonic booms in preparation for an SST. The majority were OK, but an extremely vocal minority were not. A huge part of cancelling the SST was that these experiments, plus the fact that the government paid little compensation to those affected, led to public outrage. Supersonic flight absolutely *does* cause serious noise issues (just note how every single time fighters go supersonic on an intercept, people notice and complain).

I can't see a personal spacecraft being cheaper to operate than a private plane. Those are not cheap (to put it lightly; for context, a brand-new midsize luxury car seems to cost around the same as a 40-year-old small Cessna, which will have higher fuel and maintenance costs than the car).