Not quite... if the company they were insured under at the time of the test is on the hook for all related costs, then it doesn't matter if they hike the rates after -- you can just switch companies, and the new one will have lower rates since they don't have to handle the test-related costs. Therefore it doesn't really make sense for the original company to hike the rates after the test, since they'll just drive away the customer without getting rid of the expense. That means they have to raise the rates for everyone in advance, which really is awfully close to society paying for it.
This is exactly the case for nationalized health care. Insurance companies are about mitigating risk. Once you've tested positive (at least for some conditions), you're no longer a risk. A rational insurance company would then set your rates at the cost of treatment.
However, as a society, we expect to have a certain incidence of these genetic disorders. It's unfair to expect the individual to pay for it -- they did nothing wrong, they shouldn't be punished. We as a society either need to decide that we don't care to help these people, tough luck for them, or we need to decide that we look out for our own and pay for the health care for these sorts of disorders.
Alternatively, we could come up with some plan that said that whoever your insurance company is when you have the test, they're on the hook for all future related bills -- but that's really just the same thing as society paying for it, we've just migrated the cost from a tax into insurance premiums, and it seems to me that hiding it that way is a bad thing.
The energy content of 2 tons of TNT is about 8 gigajoules. That's rather a lot of energy. A kinetic projectile at 10,000 m/s -- mach 30 -- has 50 megajoules per kilogram. You'd need a 160 kg of projectile to reach 8 GJ. Seems possible for a shipboard system, but I bet the first applications are much smaller. Anyway, for something that size you'd want much more than a few hundred kw of generator -- even at 1MW, that's over 2 hours between shots with no inefficiencies anywhere. My personal guess is that the first deployed railguns will be moderate velocity ( 10 kps) moderate weight projectiles (a couple kg) intended basically for tank-killing and use against light ships. And even so, they'll be power-hungry, because they'll want to fire several times a minute.
100 volts of electricity to make light that can only be seen in a dark room? Would we be able to power this via a battery for any length of time, and would I get electricuted[sic] if I dropped it?
Voltage != Power. Power is voltage times current (amps). Increased voltage is not directly connected to power usage, it all depends how much current goes with it. Current CCFL laptop display lights use voltages usually > 1000 volts without any problems with battery life or electrocution hazards.
Buying things on eBay isn't bad... it's the sign of a good engineer who's being budget-minded. I work at a small aerospace company, and we routinely buy things on eBay. Usually things like valves, fittings, battery chargers and the like for general stock around the shop, but also flowmeters, high pressure compressors, dewars, etc etc. It's much cheaper to buy a flowmeter on eBay and send it out for a calibration than it is to buy it new, and the same is true for many other tools and instruments. I would fully expect significant parts in any of these sorts of contest winners to come from eBay or other used / surplus sources.
History? Since when? I've been doing a bit of PIC microcontroller programming lately, and for debugging the various communications between a PIC and various peripherals this sounds perfect. I'll be seriously investigating getting one of these. Mostly the timing of such circuits isn't the issue, it's seeing the data actually on the channel without adding lots of debug code I don't have time or space for. This sounds like a great tool for the serious hobbyist.
Atmospheric water? What atmospheric water? You're out of the atmosphere...
This probably works better with a bunch of cars on the cable and wired power transmission; instead of space-elevator length *moving* cables you just need regenerative braking. It helps some, but most payloads probably stay up there, so you only recover the car energy, which is supposed to be a small fraction of the total. The passenger trade will run on rockets, the elevator is for things like satellites, propellant, and other bulk supplies.
Then again, if you're mining asteroids, that's a different matter -- and one of the few truly compelling cases for the elevator concept in my opinion. It neatly solves the power transmission problem on the elevator, and also solves the problem of how you get the asteroid down without bringing it so close to Earth while it's in a single large lethal piece (remember, geosynchronous orbit is actually quite a long way away).
The problem is that the wires have resistance. And they're thin -- the elevator is thin. So the resistance is non-trivial. It might be tiny, but it's not zero, and over thousands of kilometers it adds up. It's not impossible, and whether it's better or worse than beamed power depends on who you ask and what assumptions you make -- ie if you can get ballistic conduction working for that length of nanotube. Also, it needs to be at very high voltage (to keep the current and I^2*R losses low), which makes the power conversion to run the motors harder (read: heavier and less efficient; doing it is straightforward enough). My personal guess is that the wires beat the lasers, but I think I'm in a minority there.
The elevator car isn't on the ground. You have to get power to it somehow. The basic options are solar generation on the car, beamed power (either laser or microwave... both have problems), or wires (which have resistance problems). So yes, you use ground based power -- the power isn't the problem, it's the power transmission.
The curve is exponential -- each incremental delta-v requires exponentially more propellant. But the *point* is cost. I don't care how much fuel it takes, I care what the ticket price is. If you have a spacecraft that's built of unobtanium and weighs nothing, you can solve the exponential problem by just adding more tankage accordingly. At some point conventional chemical rockets stop being the answer because of this exponential growth, but that point is well past just getting to orbit. Interplanetary travel is doable on chemical rockets, but probably not ideal -- ion engines and solar / nuclear thermal rockets probably win for that long term; for interstellar, something entirely different is needed. The point is that for the regime space elevators are relevant, the exponential curve hasn't killed chemical rockets as an option.
The whole thing is about economics. You want high speed for short transit times, obviously, but the important reason is to get the car off the bottom of the elevator. You need to do this because of the weight limit on your cable. Once the car is a ways up it counts less against the weight limit. This implies high speed tires, high power motors, and high power transmission rates. The reason I say Mach 3 is nothing magic, just that many elevator proponents talk about speeds around 1000 m/s.
Even if your car "only" goes at 100 m/s, that's 1kw of power *per kilogram* of mass (elevator + payload), or about 1/2 HP per pound -- which is close to the weight of current electric motors, let alone the car and payload. The power to weight isn't required to loft the car, it's required to loft it quickly. And don't forget, you have to cool the motor -- a task that gets harder in vacuum.
You still actually need to put a lot of energy into climbing. The first 1000km, where centripetal acceleration isn't helping much and gravity is still high, means about 1E6m * 10 m/s^2 or about 10 MJ/kg. Total energy is well more than that, I think about 30MJ/kg. I don't have the exact number handy, but it's big.
The problem with solar cells and any number of other power technologies is, again, power to weight and therefore elevator car speed.
I'm not trying to say any of this is impossible -- just that the challenges make rockets look really easy, especially if you have space elevator class materials available; and rockets aren't really all that inefficient from an energy standpoint, once sad materials are available to improve the mass ratio a bit.
That said, people are already starting to incorporate nanotubes in composite materials. The two hard parts are that they're really slippery and it's hard to get the matrix to stick to them, and that they tend to clump up a lot. The increased length helps with the first problem -- slippery is less of a problem if there's more surface to stick to. I don't know about the dispersion.
Nanotube composites are already impressive. You can get things with 30-50% more stiffness, 50-200% more thermal conductivity, lower thermal expansion, and other useful properties. Metal matrix composites are also impressive. Think aluminum with nanotubes added. You can get double the strength, more than double the stiffness, and double or more the thermal conductivity in something as machinable as aluminum by adding only 1-2% nanotubes. This is a *rapidly* advancing field, and it's poised to seriously change high end materials science in the very near future.
People don't seem to get this somehow. Yes, mass ratio matters. A lot. Let's look at LOX+Kerosene, a very typical combination in many ways. You get an ISP of about 3000 m/s in a medium-high performance vacuum engine (the case for most of the way to orbit). LEO takes about 9000 m/s of delta-v by the time you account for aerodynamic and gravity losses. That means the mass ratio of your rocket needs to be about e^(9000/3000) = e^3 = 20. So 5% of your rocket makes it to orbit. Yup, that sucks. LOX costs about $0.07/lb in bulk, kerosene about $0.30. So propellant costs are about $0.15/lb for propellant, or $3/lb of orbited mass.
Now lets look at the space elevator. Climbing to geosynchronous orbit is equivalent to about 8000 m/s of delta-v (roughly... don't have the exact number off hand and I don't feel like calculating it). From 1/2M*v^2, that's 32MJ/kg. That's about the energy you get from burning 6 kg of LOX-kerosene. So from an energy equivalence standpoint, you're using 6 kg of propellant worth of energy instead of 19 -- a factor of 3 improvement.
The problem with the space elevator is twofold. First, the required *form* of the energy is different. You can't just use cheap hydrocarbon fuels -- you have to convert it to electricity, and then get that electricity up to the elevator either by beaming it or along wires, and neither option is efficient in the slightest. In fact, by the time you turn the hydrocarbon fuel into electricity and then get it to the elevator car, you're under 50% efficient; being as high as 30% would take a lot of work and be quite impressive. But the rocket was 30% efficient! Space elevators are *not* particularly more efficient than rockets.
The second problem is the infrastructure of the space elevator -- the required capital investment for a certain payload rate (kg delivered per day) is higher than for the rocket (we won't even discuss non-reusable rockets). Even if you got the space elevator more energy-efficient than the rocket, this fact combined with the slower transit time, the geosynchronous orbit as the only one available, and the more complicated technological requirements, the rockets win.
Yes, the space elevator tech is harder. The ribbon itself and the beamed power are the obvious examples, but there are others. For example, the tires on the car that work against the ribbon -- you need tires that run at about Mach 3 and are good for 27000 miles. That's not even remotely easy. You need motors that have higher power to weight ratios than currently exist. Etc, etc, etc. Rockets, in comparison, are easy. Especially if you have space-elevator class building materials available -- at that point you can do SSTO with pressure fed rockets, and get rid of the pumps altogether -- the pumps being the hardest part of rocket engine development by far in a conventional design.
When people say that for space elevators you only have to provide the energy to climb up, and aren't wasting the energy carrying propellant, they often forget that it's actually a *lot* of energy to climb up, and that rockets are actually remarkably good at converting available chemical energy into exhaust kinetic energy -- some are better than 80% efficient by that metric.
It's not about the fuel prices. Never has been, and won't be for the foreseeable future. Propellant is cheap, it's the vehicle that's expensive. Elon Musk of SpaceX was recently quoted as saying propellant costs are comparable to the accounting errors.
Remember that the space elevator has to supply all the energy to the payload too, but it has to get it in a much more expensive form -- like electricity beamed from the ground by lasers or some such. Rockets aren't actually all that energy inefficient in comparison.
I used to be a huge fan of the space elevator idea, but then I started looking what those same materials do to rockets. SSTO is just the start. And remember, those materials will change rockets long before they make a space elevator.
Of course, I am a rocket engineer, so I might be a little biased, but I've also examined the problem in some detail:)
Well, they're still more slippery than wool, so that problem has to be solved too. But this is one piece of the puzzle, and it's very cool to see it coming along.
No it doesn't. Dry ice is made from commercial CO2, which comes from fossil fuels. In fact, the manufacture of dry ice releases additional CO2 beyond just what ends up as dry ice. The reason is that air is only a few hundred ppm CO2, which is not normally economical to capture and do anything with. Industrially it often comes as a byproduct of ammonia production -- natural gas, CH4, is converted into hydrogen and CO2; the hydrogen is used in making ammonia.
Putting more than one gyro in a box is the same as putting one in with a different angular momentum. Angular momentum is a vector quantity; two gyros add together and act like one with the corresponding angular momentum. So when you try to move the box it will act like a box with a gyro in it, and you won't be able to tell how many gyros there are.
Electrons orbit in orbitals, but they also spin, like how the Earth both orbits and rotates. See Quantum spin for details. Electrons and other subatomic particles have angular momentum. But, since for the most part they're randomly aligned, you don't notice -- as above, the spins add to mostly zero. I believe there are macroscopic experiments that demonstrate this, but I don't remember details off hand.
I'm being realistic about it. DVD doesn't prevent me from doing what I want to do; it doesn't even make it hard. I'm not going to take an idealistic principled stand and not watch any movies; I will, however, choose the least inconvenient format, and I'm willing to do so in the face of incentives like higher picture quality. I'm also not about to just pirate whatever I feel like; I don't agree with what the MAFIAA is doing, but I don't think it's ok to infringe their copyrights either -- I'll either pay up and deal with it or I'll not watch / listen to their stuff.
There's something called PSRR, or Power Supply Rejection Ratio -- aka how well the DAC does at rejecting power supply influence on its output. This characteristic varies with the quality of the DAC, and is dependent on the frequency of the noise. They do better at rejecting low frequency noise than high frequency. And guess what? your PC has lots of high frequency noise. I've looked at dirty mains power on a scope, I've looked at "sine wave" inverter output, and I've built ultralow noise power supplies for analog signal work. It's far, far easier to clean up a signal that doesn't have lots of high energy high frequency components to it, which is exactly what the noise on a PC power rail looks like. Sine wave inverter output looks mostly like a sine wave with a triangle wave laid on top of it, which is actually fairly easy to clean up with some simple inductors and caps -- the higher frequency components are low energy, and the medium frequency components filter easily enough. It's easier to clean up even relatively dirty mains power than it is to clean up the output of your average computer supply.
As you say, handling the internal sources of noise is a solved problem. Proper supply bypassing and using a high quality D/A, combined with a decent board layour and a solid ground plane, are usually all that's required. Ones like the PDM output of my amp actually engage in techniques known as noise shaping -- by making sure *all* the noise generated is well beyond the range of hearing (think 100kHz+), you don't have to worry about it much -- the speakers won't reproduce it, and the tiny fraction they do you can't here anyway.
Panasonic SA-XR25. Looks like it can be had for under $200 these days. When I compared it to other amps a bit over a year ago it compared favorably to almost anything under $1k.
Slashdot Mirror
Some of us are kinda fond of them, you know...
Tag 'boycottroland'
Not quite... if the company they were insured under at the time of the test is on the hook for all related costs, then it doesn't matter if they hike the rates after -- you can just switch companies, and the new one will have lower rates since they don't have to handle the test-related costs. Therefore it doesn't really make sense for the original company to hike the rates after the test, since they'll just drive away the customer without getting rid of the expense. That means they have to raise the rates for everyone in advance, which really is awfully close to society paying for it.
This is exactly the case for nationalized health care. Insurance companies are about mitigating risk. Once you've tested positive (at least for some conditions), you're no longer a risk. A rational insurance company would then set your rates at the cost of treatment.
However, as a society, we expect to have a certain incidence of these genetic disorders. It's unfair to expect the individual to pay for it -- they did nothing wrong, they shouldn't be punished. We as a society either need to decide that we don't care to help these people, tough luck for them, or we need to decide that we look out for our own and pay for the health care for these sorts of disorders.
Alternatively, we could come up with some plan that said that whoever your insurance company is when you have the test, they're on the hook for all future related bills -- but that's really just the same thing as society paying for it, we've just migrated the cost from a tax into insurance premiums, and it seems to me that hiding it that way is a bad thing.
The energy content of 2 tons of TNT is about 8 gigajoules. That's rather a lot of energy. A kinetic projectile at 10,000 m/s -- mach 30 -- has 50 megajoules per kilogram. You'd need a 160 kg of projectile to reach 8 GJ. Seems possible for a shipboard system, but I bet the first applications are much smaller. Anyway, for something that size you'd want much more than a few hundred kw of generator -- even at 1MW, that's over 2 hours between shots with no inefficiencies anywhere. My personal guess is that the first deployed railguns will be moderate velocity ( 10 kps) moderate weight projectiles (a couple kg) intended basically for tank-killing and use against light ships. And even so, they'll be power-hungry, because they'll want to fire several times a minute.
100 volts of electricity to make light that can only be seen in a dark room? Would we be able to power this via a battery for any length of time, and would I get electricuted[sic] if I dropped it?
Voltage != Power. Power is voltage times current (amps). Increased voltage is not directly connected to power usage, it all depends how much current goes with it. Current CCFL laptop display lights use voltages usually > 1000 volts without any problems with battery life or electrocution hazards.
Buying things on eBay isn't bad... it's the sign of a good engineer who's being budget-minded. I work at a small aerospace company, and we routinely buy things on eBay. Usually things like valves, fittings, battery chargers and the like for general stock around the shop, but also flowmeters, high pressure compressors, dewars, etc etc. It's much cheaper to buy a flowmeter on eBay and send it out for a calibration than it is to buy it new, and the same is true for many other tools and instruments. I would fully expect significant parts in any of these sorts of contest winners to come from eBay or other used / surplus sources.
"Apparently, politicians... [don't] always work out."
History? Since when? I've been doing a bit of PIC microcontroller programming lately, and for debugging the various communications between a PIC and various peripherals this sounds perfect. I'll be seriously investigating getting one of these. Mostly the timing of such circuits isn't the issue, it's seeing the data actually on the channel without adding lots of debug code I don't have time or space for. This sounds like a great tool for the serious hobbyist.
Atmospheric water? What atmospheric water? You're out of the atmosphere...
This probably works better with a bunch of cars on the cable and wired power transmission; instead of space-elevator length *moving* cables you just need regenerative braking. It helps some, but most payloads probably stay up there, so you only recover the car energy, which is supposed to be a small fraction of the total. The passenger trade will run on rockets, the elevator is for things like satellites, propellant, and other bulk supplies.
Then again, if you're mining asteroids, that's a different matter -- and one of the few truly compelling cases for the elevator concept in my opinion. It neatly solves the power transmission problem on the elevator, and also solves the problem of how you get the asteroid down without bringing it so close to Earth while it's in a single large lethal piece (remember, geosynchronous orbit is actually quite a long way away).
The problem is that the wires have resistance. And they're thin -- the elevator is thin. So the resistance is non-trivial. It might be tiny, but it's not zero, and over thousands of kilometers it adds up. It's not impossible, and whether it's better or worse than beamed power depends on who you ask and what assumptions you make -- ie if you can get ballistic conduction working for that length of nanotube. Also, it needs to be at very high voltage (to keep the current and I^2*R losses low), which makes the power conversion to run the motors harder (read: heavier and less efficient; doing it is straightforward enough). My personal guess is that the wires beat the lasers, but I think I'm in a minority there.
The elevator car isn't on the ground. You have to get power to it somehow. The basic options are solar generation on the car, beamed power (either laser or microwave... both have problems), or wires (which have resistance problems). So yes, you use ground based power -- the power isn't the problem, it's the power transmission.
The curve is exponential -- each incremental delta-v requires exponentially more propellant. But the *point* is cost. I don't care how much fuel it takes, I care what the ticket price is. If you have a spacecraft that's built of unobtanium and weighs nothing, you can solve the exponential problem by just adding more tankage accordingly. At some point conventional chemical rockets stop being the answer because of this exponential growth, but that point is well past just getting to orbit. Interplanetary travel is doable on chemical rockets, but probably not ideal -- ion engines and solar / nuclear thermal rockets probably win for that long term; for interstellar, something entirely different is needed. The point is that for the regime space elevators are relevant, the exponential curve hasn't killed chemical rockets as an option.
The whole thing is about economics. You want high speed for short transit times, obviously, but the important reason is to get the car off the bottom of the elevator. You need to do this because of the weight limit on your cable. Once the car is a ways up it counts less against the weight limit. This implies high speed tires, high power motors, and high power transmission rates. The reason I say Mach 3 is nothing magic, just that many elevator proponents talk about speeds around 1000 m/s.
Even if your car "only" goes at 100 m/s, that's 1kw of power *per kilogram* of mass (elevator + payload), or about 1/2 HP per pound -- which is close to the weight of current electric motors, let alone the car and payload. The power to weight isn't required to loft the car, it's required to loft it quickly. And don't forget, you have to cool the motor -- a task that gets harder in vacuum.
You still actually need to put a lot of energy into climbing. The first 1000km, where centripetal acceleration isn't helping much and gravity is still high, means about 1E6m * 10 m/s^2 or about 10 MJ/kg. Total energy is well more than that, I think about 30MJ/kg. I don't have the exact number handy, but it's big.
The problem with solar cells and any number of other power technologies is, again, power to weight and therefore elevator car speed.
I'm not trying to say any of this is impossible -- just that the challenges make rockets look really easy, especially if you have space elevator class materials available; and rockets aren't really all that inefficient from an energy standpoint, once sad materials are available to improve the mass ratio a bit.
Note that nanotubes != CF.
That said, people are already starting to incorporate nanotubes in composite materials. The two hard parts are that they're really slippery and it's hard to get the matrix to stick to them, and that they tend to clump up a lot. The increased length helps with the first problem -- slippery is less of a problem if there's more surface to stick to. I don't know about the dispersion.
Nanotube composites are already impressive. You can get things with 30-50% more stiffness, 50-200% more thermal conductivity, lower thermal expansion, and other useful properties. Metal matrix composites are also impressive. Think aluminum with nanotubes added. You can get double the strength, more than double the stiffness, and double or more the thermal conductivity in something as machinable as aluminum by adding only 1-2% nanotubes. This is a *rapidly* advancing field, and it's poised to seriously change high end materials science in the very near future.
People don't seem to get this somehow. Yes, mass ratio matters. A lot. Let's look at LOX+Kerosene, a very typical combination in many ways. You get an ISP of about 3000 m/s in a medium-high performance vacuum engine (the case for most of the way to orbit). LEO takes about 9000 m/s of delta-v by the time you account for aerodynamic and gravity losses. That means the mass ratio of your rocket needs to be about e^(9000/3000) = e^3 = 20. So 5% of your rocket makes it to orbit. Yup, that sucks. LOX costs about $0.07/lb in bulk, kerosene about $0.30. So propellant costs are about $0.15/lb for propellant, or $3/lb of orbited mass.
Now lets look at the space elevator. Climbing to geosynchronous orbit is equivalent to about 8000 m/s of delta-v (roughly... don't have the exact number off hand and I don't feel like calculating it). From 1/2M*v^2, that's 32MJ/kg. That's about the energy you get from burning 6 kg of LOX-kerosene. So from an energy equivalence standpoint, you're using 6 kg of propellant worth of energy instead of 19 -- a factor of 3 improvement.
The problem with the space elevator is twofold. First, the required *form* of the energy is different. You can't just use cheap hydrocarbon fuels -- you have to convert it to electricity, and then get that electricity up to the elevator either by beaming it or along wires, and neither option is efficient in the slightest. In fact, by the time you turn the hydrocarbon fuel into electricity and then get it to the elevator car, you're under 50% efficient; being as high as 30% would take a lot of work and be quite impressive. But the rocket was 30% efficient! Space elevators are *not* particularly more efficient than rockets.
The second problem is the infrastructure of the space elevator -- the required capital investment for a certain payload rate (kg delivered per day) is higher than for the rocket (we won't even discuss non-reusable rockets). Even if you got the space elevator more energy-efficient than the rocket, this fact combined with the slower transit time, the geosynchronous orbit as the only one available, and the more complicated technological requirements, the rockets win.
Yes, the space elevator tech is harder. The ribbon itself and the beamed power are the obvious examples, but there are others. For example, the tires on the car that work against the ribbon -- you need tires that run at about Mach 3 and are good for 27000 miles. That's not even remotely easy. You need motors that have higher power to weight ratios than currently exist. Etc, etc, etc. Rockets, in comparison, are easy. Especially if you have space-elevator class building materials available -- at that point you can do SSTO with pressure fed rockets, and get rid of the pumps altogether -- the pumps being the hardest part of rocket engine development by far in a conventional design.
When people say that for space elevators you only have to provide the energy to climb up, and aren't wasting the energy carrying propellant, they often forget that it's actually a *lot* of energy to climb up, and that rockets are actually remarkably good at converting available chemical energy into exhaust kinetic energy -- some are better than 80% efficient by that metric.
It's not about the fuel prices. Never has been, and won't be for the foreseeable future. Propellant is cheap, it's the vehicle that's expensive. Elon Musk of SpaceX was recently quoted as saying propellant costs are comparable to the accounting errors.
Remember that the space elevator has to supply all the energy to the payload too, but it has to get it in a much more expensive form -- like electricity beamed from the ground by lasers or some such. Rockets aren't actually all that energy inefficient in comparison.
I used to be a huge fan of the space elevator idea, but then I started looking what those same materials do to rockets. SSTO is just the start. And remember, those materials will change rockets long before they make a space elevator.
Of course, I am a rocket engineer, so I might be a little biased, but I've also examined the problem in some detail :)
Well, they're still more slippery than wool, so that problem has to be solved too. But this is one piece of the puzzle, and it's very cool to see it coming along.
No it doesn't. Dry ice is made from commercial CO2, which comes from fossil fuels. In fact, the manufacture of dry ice releases additional CO2 beyond just what ends up as dry ice. The reason is that air is only a few hundred ppm CO2, which is not normally economical to capture and do anything with. Industrially it often comes as a byproduct of ammonia production -- natural gas, CH4, is converted into hydrogen and CO2; the hydrogen is used in making ammonia.
See Carbon Dioxide for details.
Actually, you missed on both counts.
Putting more than one gyro in a box is the same as putting one in with a different angular momentum. Angular momentum is a vector quantity; two gyros add together and act like one with the corresponding angular momentum. So when you try to move the box it will act like a box with a gyro in it, and you won't be able to tell how many gyros there are.
Electrons orbit in orbitals, but they also spin, like how the Earth both orbits and rotates. See Quantum spin for details. Electrons and other subatomic particles have angular momentum. But, since for the most part they're randomly aligned, you don't notice -- as above, the spins add to mostly zero. I believe there are macroscopic experiments that demonstrate this, but I don't remember details off hand.
Unfortunately, one of the original 10 team members died in a freak accident involving a rolling barrel during construction.
I'm being realistic about it. DVD doesn't prevent me from doing what I want to do; it doesn't even make it hard. I'm not going to take an idealistic principled stand and not watch any movies; I will, however, choose the least inconvenient format, and I'm willing to do so in the face of incentives like higher picture quality. I'm also not about to just pirate whatever I feel like; I don't agree with what the MAFIAA is doing, but I don't think it's ok to infringe their copyrights either -- I'll either pay up and deal with it or I'll not watch / listen to their stuff.
Yes. Class D includes PWM, PDM, and "class T" amplifiers.
There's something called PSRR, or Power Supply Rejection Ratio -- aka how well the DAC does at rejecting power supply influence on its output. This characteristic varies with the quality of the DAC, and is dependent on the frequency of the noise. They do better at rejecting low frequency noise than high frequency. And guess what? your PC has lots of high frequency noise. I've looked at dirty mains power on a scope, I've looked at "sine wave" inverter output, and I've built ultralow noise power supplies for analog signal work. It's far, far easier to clean up a signal that doesn't have lots of high energy high frequency components to it, which is exactly what the noise on a PC power rail looks like. Sine wave inverter output looks mostly like a sine wave with a triangle wave laid on top of it, which is actually fairly easy to clean up with some simple inductors and caps -- the higher frequency components are low energy, and the medium frequency components filter easily enough. It's easier to clean up even relatively dirty mains power than it is to clean up the output of your average computer supply.
As you say, handling the internal sources of noise is a solved problem. Proper supply bypassing and using a high quality D/A, combined with a decent board layour and a solid ground plane, are usually all that's required. Ones like the PDM output of my amp actually engage in techniques known as noise shaping -- by making sure *all* the noise generated is well beyond the range of hearing (think 100kHz+), you don't have to worry about it much -- the speakers won't reproduce it, and the tiny fraction they do you can't here anyway.
Panasonic SA-XR25. Looks like it can be had for under $200 these days. When I compared it to other amps a bit over a year ago it compared favorably to almost anything under $1k.