No, they won't. If energy becomed more expensive, people would just return to nuclear power. If you think that people are going to resign from driving, driers and air-conditioners, you're insane.
If it's forced on them incrementally, I think they might. Especially since the alternative with or without fusion power is to have an Evil Nuclear Plant right next to their city (fusion isn't much cleaner than fission for low-level waste).
It's a case of foolishness (hopefully) defeating foolishness.
Bottom line, if someone can come up with a cheap/light way to store electricity (either through a flywheel or through a superconducting coil), we'll still keep boosting my Exxon stock. And we'll do it by drilling in your backyard.
Fuel cells for cell phones should be appearing within the next 2-5 years or so. Prototypes already exist, and energy density is almost as good as a chemical fuel.
Fuel cells at present are a pain for things like cars because storing the hydrogen they use is very annoying (it's nowhere near as dense as a liquid). However, fuel cells that can process methanol are buildable, and will IMO probably be what brings fuel cell cars into the real world. Or you could just pump methanol into an interal combustion engine made of ceramics or otherwise made corrosion-resistant.
Methanol can't be easily produced by "recharging" fuel cells (unlike hydrogen), but you can make it by fermenting plants or by direct synthesis (burn CO2 incompletely in a hydrogen atmosphere and use fractional distillation on the combustion products).
In summary, power storage won't be a problem for long.
And don't forget the other things that are petroleum based - the industrial chemical industry , the plastics industry, and the fertilizer industry to name a few - when oil gets scarce (coming soon) everythings gets expensive.
Plastics can be produced from anything hydrocarbon-based. You can already do it easily from plant matter - petroleum just happens to be cheaper for the time being. Ditto most organic chemicals used in industry. Most fertilizers that I know of are based on minerals, not petroleum (a fertilizer provides mostly nitrogen, phosphorus, and potassium to plants, none of which petroleum is terribly rich in; typical fertilizers are based ammonium nitrate, potassium nitrate, or pick-a-random-phosephate, if I understand correctly).
Current fusion engineering designs call for deuterium and tritium to be used as the fuel. You can easily get a virtually inexhaustible supply of deuterium from seawater. Tritium is another matter altogether. The only source available for this is fission reactors.
A fusion reactor would produce more than enough neutrons to breed tritium (and hopefully helium-3 as well).
Let that stream of plasma impact some new magnetic fields. It will push on the field lines, imparting momentum to them. This could accelerate the magnets generating those lines, even if they were on the other side of the wall. If those magnets were attached to wheels, could you actually make a "magnetic turbine", directly converting the kinetic energy of the plasma into mechanical work?
Maybe a real physicist can enlighten me... Is this possible?
I'm not a real physicist either, but I faked being one well enough to get a free vacation out of it (IPhO...'94? '95? can't remember):).
It turns out that you can get electrical power directly from a plasma stream by the Hall effect (similar to what you describe). Fire a stream of plasma through a magnetic field that's perpendicular to the direction of plasma motion, and you'll get negative charges deflected one way and positive charges deflected another way. Put electrodes beside the plasma stream to collect the charge, and you have a power source. Look up "magnetohydrodynamic generators" in your favourite encyclopedia for more information.
As for fusion plants, the best approach to tapping power that I've heard of is to put a big block of lead or concrete around the reactor, let it heat up from absorbing gamma rays and other crud coming out of the plasma, and run steam turbines off of it.
The downsides are that the D/He-3 reaction has a higher energy threshold, so it requires a better confined plasma to make D/He-3 energy production plausible. Also, the supplies of He-3 on Earth are limited. Of course this disadvantages leads to a interesting convergence of interest between fusion research and space exploration. There's plenty of He-3 (deposited from the solar wind) on the moon. So He-3 mining could be powerful reason to maintain a base on the moon.
I'd understood that He3 was one of the decay products that you get from neutron irradiation of a fusion reactor's lithium blanket. A few other articles that I'd read claimed that this was the main reason a lithium blanket was used (the idea was that a H2/H3 plasma fusing would transmute enough of the blanket into He3 to provide H2/He3 burning for power generation).
Is this still a viable option, or has it been shown to be impractical?
Pretty interesting idea. But that patch of empty sky is actually full of a lot of air. The heat has to pass through a lot of air before it gets free in space. Since conduction and convection are much more effective than radiation at transmitting heat, wouldn't you just end up making your local chunk of the atmosphere very warm?
Heat lost by radiation goes up as the fourth power of temperature - this is why you'd have the coolant hot enough to glow orange:). As you raise the temperature, radiative cooling starts to dominate (transfer by conduction goes up linearly with the temperature difference).
If I understand correctly, the atmosphere's pretty good at blocking light in the thermal IR band, but you can see by example that visible and near-visible passes through quite easily - most of the sunlight shining on earth reaches the ground.
Heat the coolant up to a thousand degrees centigrade or so, and most of its emission will be in visible and near-IR (thermal IR is the band that objects at room temperature shine in).
Now, a week of cloudy days would make this system less effective, but you could get around that by using a nearby lake for heat storage until the sky clears again (though that won't be very nice for the inhabitants of the lake).
ok, so you have this plasma "floating" in the bottle at 12 gazillion degrees. the power goes out. doesn't your ball of plasma just eat it's way through anything it touches and head towards the center of the earth?
The plasma is very hot, but also very tenuous. You might have a few thousanths of a gram of matter in the reactor. As soon as this touches the reactor wall, it cools down to room temperature. Your reactor wall gets warm. That's about it.
Also, as your plasma is much less dense than air, if anything, a ball of plasma would rise.
You'd never get a ball. What happens is that as the containment field weakens, the donut-shaped mass of plasma expands until it touches the side of the container. In theory, if you had a "hole" in the containment field instead of the whole thing weakening, you could get a jet of what looks like flame, but this is next to impossible to do even if you're _trying_. The natural shape of the field is more-or-less uniform.
Plasma is just a hot, conducting gas. It follows gas law like everything else.
Perhaps... Heat Pollution would become a problem, but hopefully that would be worse than greenhouse-effect causing gasses that we release by Fossile fuels, if such gasses do in fact cause global warming (of which I'm not fully convinced.)
Getting rid of excess heat is easy enough if you decide to spend the money to do it - just set up a solar heat plant, point the mirrors at a patch of empty sky, and run it in reverse (use a heat exchanger to heat the working fluid from your waste heat, run it through the pipes in the mirror array, and let it beam its blackbody radiation out into space).
Yes, it takes a fair bit of power to move heat across this big a temperature differential (you need to heat the pipes up until they're glowing orange to get a decent rate of heat transfer). However, you'd only need to do it in the first place if you had power to burn.
To prevent confused responses - you're not building a giant toaster-oven. The heat you're stuffing into the pipes comes from the city or industrial park you want to cool off, via heat exchanger. You're not generating it from scratch.
In other words, either we get cheap, clean, nearly free unlimited energy, or our future is gonna look uncomfortably like those Mad Max movies.
Actually, what would be more likely to happen is a slow and painful shift to renewable forms of energy and a lifestyle that consumes less power.
Fossil fuels won't run out all at once - they'll get incrementally more expensive as they become scarcer and more difficult to extract. As price climbs, people will out of necessity start using less power, and alternative forms of power generation will become cost-competitive (even if their own cost doesn't change).
You'll see more houses with photocell shingles. You'll see a mirror-based heat plant or a wind farm next to most cities. More people will take public transit, because it's cheaper than driving. House design will depend more on passive heating/cooling and good insulation than on furnaces and air conditioners (they'll still be used; just on lower settings). People will have to hang-dry their clothes instead of tossing them in the drier. And so forth.
A Mad Max style catastrophy would only happen if energy supplies ran out all at once. A nuclear war or an asteroid strike could cause this, but I doubt fossil fuel problems will.
What about the supply of deuterium? It's not perpetual...
It's close enough to perpetual that it makes no difference.
About 1 hydrogen atom in 7000 is deuterium. Hydrogen accounts for 1/9th the mass of water, which means that the mass of deuterium we could extract from the oceans is about 1/63000th the mass of the oceans.
Assuming an average depth of about 3km and a coverage of 2/3 of the Earth's surface, you have about 1e18 tonnes of water on the planet. Let's say 1e-5 of that is deuterium that you can easily extract, and you get 1e13 tonnes of deuterium.
Fusion is about 1% efficient at turning mass into energy. Let's say that your fusion reactor is 0.01% efficient in total (pessimistic estimate). This gives about 1e13 J/kg, or 1e16 J/tonne.
The total amount of energy we could extract from fusing the oceans' deuterium with these fairly inefficient fusion plants is about 1e29 J.
If we want our fuel supply to last a million years, that gives a world power consumption of 3e15 watts. Far, far more than we consume now.
If, within the next million years, we build reactors good enough to fuse light hydrogen, we'll have enough fuel to last us until the sun burns out (several billion years). Or we could just ship in deuterium from elsewhere in the solar system. Either way, barring a *huge* population expansion, we won't run out of power - ever.
In my experience with high precision RF simulations on massive SGI multiprocessor machines, the OS (IRIX), the compiler (MipsPRO), and the communications package (OpenMP) do very little for you automatically. In order to achieve decent scalability you need to take your computing needs into consideration when you develop the architecture behind your simulation. If you, as the developer, don't distribute your calculations as evenly as possible, your simulation is going to run very, very poorly.
What the original poster was referring to, if I understand correctly, was distribution of processes or (in his case) web page requests across hardware nodes. This is handled by the OS and/or libraries in the setups that I've seen (admittedly few).
I definitely agree that the task has to be properly parallelized in the first place, but that wasn't what I was responding to.
Regarding the internal bus in a SMP system being faster than the communications network in a cluster, that's what I'd thought too. Then my prof got me to run subsets of the SKaMPI and NPB 2.3 benchmark suites on our cluster (MPI libraries on top of SCore running on a bunch of dual-processor Linux boxes). Running on 2nx1 nodes was faster than running on nx2 nodes for almost all of the tests. Caveat: this was for small n, and I can't guarantee that it wasn't just due to lousy process distribution or lousy SMP communication routines in the libraries.
YMMV.
For anything with a high communications load, SMP almost certainly wins, but for moderate communications loads, my experiences have made me leery of it.
Any task that is easily parallelized and has low internal communications requirements would run more effectively on multiple servers than on one SMP behemoth.
I have several problems with this generalization. First, parallelizing over multiple servers always adds overhead (in both $$$ and performance) of its own. How are you going to spread a load over multiple web servers? You need a load balancer, either the dedicated (pricey) hardware kind or a standard server converted over to load balancing service (which doesn't get you the greatest speed or scalability in the world). Even in a scientific application that you spread over several boxes, you need some kind of load balancer or traffic cop to get an equitable distribution of work.
You make a valid point, in that load-balancing is an issue. However, I'm assuming that in the case of a web server, if you have enough traffic to need more than one server, you have enough money to buy a hardware load-balancer to spread out requests (and a hardware firewall, if management has any sense).
As for scientific applications, you need load balancing regardless of whether the processors running the threads are in one box or in several. This is usually handled transparently by the OS, the compliler, the communications library, or a combination of the above (usually all of the above). This is standard for any high-performance computing project, and so doesn't add to your maintenance overhead. It also doesn't contribute substantially to the processor workload, so I don't see it as much of a concern for scientific workloads.
Second, let's not forget that two-way (and even four-way, if you were in the Xeon market to begin with) boxes have gotten much cheaper in the past year or two. Most of the important server availability features, like hot swap drives, hot swap power supplies, ECC RAM, 64 bit PCI, etc., are almost impossible to find on 1-way systems these days.
The last time I checked, n-way systems for n > 2 were still far more expensive than n one-way systems, but I haven't checked within the past couple of months. This might have changed, but I doubt it.
N = 2 was marginal, if I remember correctly.
ECC RAM support is available on several single-CPU motherboards; check your favourite vendor's site for a list of options (admittedly pricier than most of the boards, but not horribly so).
I'm assuming that hot swap power supplies aren't relevant. Your load-balancing hardware or (for a cluster) software will be able to detect malfunctioning nodes; this is essential for any cluster of significant size. A supply failing would be no different from any other component failing from a maintenance point of view (bad node is cut out of the loop by the load balancer, the hardware person gets paged, the node is swapped out and the old node serviced or gutted for parts).
PCI-64 support is a good point. If you have to support PCI-64, then it probably makes sense to build your cluster out of dual-CPU nodes, because the incremental cost of getting a dual-CPU motherboard will be low. Quad-CPU and higher will probably be less economical (quad cost diamonds the last time I checked). You'd only need PCI-64, though, if you either had a very large communications requirement (multiple very fast network cards per node), or if you were mounting a large RAID on the node (many controllers, many strings). In the first case, I can weasel out by claiming that you're outside of my stated problem domain (low communications bandwidth):). In the second case, you're looking at one of a handful of disk nodes within a much larger system (in all likelihood). For non-disk nodes, you wouldn't need PCI-64. For clusters that distributed disks over many nodes, your I/O bandwidth needs would be adequately served by PCI-32, and PCI-64 again becomes unnecessary.
It's nice to get an interesting response, though:). You've made me think about the problem in more detail.
Has anyone done a cost-efficiency comparison of dual-cpu performance vs. a simple cpu when considering the costs involved (special SMP boards, etc.) In otherwords is it more economical to buy two web servers or one smp server with tons of ram? Do certain applications (cpu intensive obviously) save money with SMP systems verus others that depend on IO throughput, etc and what applications are those?
Any task that is easily parallelized and has low internal communications requirements would run more effectively on multiple servers than on one SMP behemoth. Web serving has zero internal communications requirement, and so falls into this category. Things like ray-tracing have low communications requirements when partitioned properly, which is why you use clusters as render farms instead of massively parallel Big Iron.
SMP has overhead from coherence operations, and more complex and expensive chipsets.
SMP benefits tasks that lend themselves to shared-memory implementations. It's a lot easier to toss ownership of memory pages back and forth inside an SMP machine than it would be to send modified pages back and forth across a network. I don't have examples of this kind of task offhand, but I'm sure they exist.
All of this is for CPU-bound tasks. For I/O bound tasks, you're still better off splitting it up into multiple machines if it's easily parallelized, but again I don't have good examples to illustrate with off the top of my head.
For more information, pick up a couple of good books on parallel computer architecture and parallel programming. Your local university's bookstore will stock these.
In fact plasma and laser accelerators, while a very good idea being explored in many places, are not ready for prime time yet. Currently, they are NOT cheaper yet... either to build or to operate. And they likely won't be for many years.
My point is that for $6 billion you could, from scratch, build a plasma accelerator research facility the size of a university and fund it for years.
The payoff from doing this would be at the very least advancement in plasma accelerator research, and possibly development of a practical plasma accelerator. This would vastly benefit all future accelerator projects.
Spending $6 billion on a conventional accelerator won't provide this long-term benefit; future accelerators will still be extremely expensive. This is why I question the wisdom of *not* devoting the funding to accelerator design research, if that level of funding is available.
It seems to me that as accelerators become more and more expensive, it might be more cost-effective to devote accelerator funding to researching other classes of accelerator, instead of building larger and larger synchrotons and other conventional accelerators.
I've been following the development of plasma accelerators with much interest. They have a much greater rate of acceleration of particles, which means that an accelerator of a given energy could be much smaller (and hence hopefuly cheaper) than a synchroton. A good paper discussing the merits and problems with current plasma accelerator designs is at http://www.indiana.edu/~icfa/icfa12/node24.html .
Other promising types of accelerator almost certainly exist also.
I'm not saying that synchrotons should be abandoned because they're bad; on the contrary, they work quite well. I'm arguing that it would be _cheaper_ to invest accelerator funding in developing new types of accelerator, as opposed to building ever-larger synchrotons.
We consider the possible influence of a star light years away more than a million years from now, but fail to track most of the near-earth objects, which would pose a much more immediate threat to life on Earth than any star! Do they just pick a large object at random and say, 'Okay, how could this affect Earth?' or what?
The purpose of this research was almost certainly not to find threats to Earth - it was to get a better map of dim stars in our area, which lets us check some of our models of star formation and celestial dynamics in our area of the galaxy.
It also provides another data point for planetary system models (by giving us a better idea of how frequent close stellar encounters are, which certainly affect sytem formation), and (if they're finding stars in addition to just measuring their courses) a better estimate for the amount of normal matter in the universe (there has been debate over how much of it is bound up in dim stars and sub-stellar objects like brown dwarfs).
What most likely happened here is someone noticed that this star in their survey would come close to earth, and said "hey, neat, free publicity for our project".
Most astronomy is pure research - there won't *be* an application for most of it.
Making money with a free MMPRPG system is simple in concept - charge for access to your world, not for the software. The software is just a means for the players to access your service.
There are a few gotchas with this:
You will have to invest heavily in building and maintaining the world.
You're asking people to fork over a monthly subscription fee for the privilege of playing in your shard/continent/what-have-you. Your world had better be a lot more fun than the free alternatives that others will set up. You can do this; you just have to be willing to support several *good* full-time artists and game designers to build, maintain, and continually improve the universe that your players will be logging into.
Collaborative efforts could in theory compete with this, but in practice *good* artists and game designers who can work full-time (read: have maintaining the world as their paid day-job) can almost certainly provide a richer environment than a patchwork of volunteers could (this is art, not software).
You'll have to spin off and maintain a customer support company or division.
If you're trying to get Joe Teenager to install and use your software, you're going to need a nice, packaged box with a good manual, and a fully-staffed tech support pool. What the user pays for when they buy the box is the ability to call you and have you hand-hold them through installing the software and connecting to your servers. You're *going* to have to sink a large amount of money and manpower into the division that handles this - if you have crappy support, most of your users will give up, and spread the word that your boxed software was impossible to use. Your servers will sit idle, and you'll lose a lot of money.
Subscription to your service should be optional, not mandadory. Ideally, you'd have about three versions of the box on the shelves, with different price tags - one that's just the client (with support for using the client), one that's a client bundled with a subscription to your service (with support for both), and one that's the client and _server_ with support for maintaining the server (charge them through the nose for this, because they're going to be *using* this tech support).
You'll have to spin off and maintain an ongoing software development division.
You don't have to be in control of the project itself, but you do have to have a paid team working on the software. This lets you address bugs customers find promptly (which you'll need early-on if you want to *keep* the customers), and continue to tweak the install and server selection routines (which you'll need to keep making more idiot-proof, and adapt to new server/client versions), and add features to the software that would help you make a more interesting game world.
All of this will cost a *vast* amount of money to set up, but will have a sustainable revenue stream coming back in. You'd have to do most of this regardless of whether the game was open or closed. Opening it means you save a bit on maintaining the software (because others' improvements and bug fixes are folded in for you), but forces you to work harder making your game world attractive (because anyone can set up their own world to compete with you).
Technical edges don't last past release anyways, so as far as I can tell you'd lose nothing by making the software itself Open or Free. Your IP is in the artwork and world design.
Computers have been working in pretty much the same way since then and it kind of reminds me the way cars continue to use the same archaic technology used back in 1900, only greatly refined.This is simply because it's easier (and cheaper) for manufacturers to maintain a single common base for their products for as long as possible before throwing it away and starting all over again. The choice, I believe, is ours: Mass production or revolution?
Well, the way that machine code works is vaguely similar, except that the hardware implementation has changed on a *fundamental* level since then, and we have several layers of increasingly abstracted software on top that couldn't exist back then, and have developed several new branches of engineering involved in the design of hardware and software for computing devices...
No, no innovation or revolution there.
Sorry if I'm sounding a bit harsh. It's just that I've seen the "we've been using [foo] forever, we're shackled to it and should switch to something new and revolutionary" argument a few times now, and it almost always a) ignores a lot of fundamental change that's gone on over the years both in [foo] and in the design and use of [foo], and b) fails to propose an alternative.
What would be the "revolution" that you hope for?
Optical computers? The computer hardware layer would change utterly, but the design concepts used wouldn't, and the software would show almost no change at all. You'd also be stuck with a feature resolution no smaller than one wavelength of light (about half a micron to a micron).
Nanomechanical computers? Same deal (though without the feature size limit). You'd still be implementing digital logic, so all of the upper layers of the design and use of computers would stay the same.
The only thing that's fundamentally different both in design and use is quantum computing, and I'll bet that even that would have strong similarities on several levels.
Technological progress rarely happens by revolution - it's by evolution of existing ideas and devices, sometimes put together in new ways. Over time, the result of these changes can be profound, but the idea that you *must* turn the world on its head to move forward is a misconception.
Sorry if I'm being impolite; this is just something that's bothered me for a while:).
Me, I'm looking forward to being able to change eye or hair color with a swallowed capsule!
You'll be waiting a bit longer, then.
This experiment involved producing mice with what amounts to an on/off switch built in to their pigment genes. The dietary supplements just toggled the switch.
The research project demonstrated one possible switch mechanism, and showed that such switches could fairly easily be spliced into arbitrary genes, if you're in a position to modify an organism's DNA (easiest before conception).
Your pill would have to rewrite all of your DNA to accomplish the same effect. If you can do that, you don't need to bother with the switch.
Still a neat experiment, though. This will make it easier to study the function of genes, and easier to study certain types of genetic disease.
The real issue is that you can't just throw away half of the data. If all you care about is magnitudes, then yes, you can take the magnitude of the energies for a given frequency. However, if you want to reproduce the original signal, you can't.
This would be the case if I was throwing away all of the real or all of the imaginary data, but I'm keeping half of each.
The spectrum really is perfectly symmetric for the real component and antisymetric for the imaginary component. Deviations from this symmetry would cause an imaginary component to exist in the input signal, which can a priori be assumed not to be the case. Phase in the real signal is encoded in the ratio of real and imaginary spectrum components for a given frequency.
The handwaving argument from information theory is that because I'm only encoding N values for N samples (the real components of the samples), I only need N components out to retrieve all of the information (the real and imaginary components of half of the spectrum). The full spectrum has a factor of two redundancy.
Again, the DCT makes the same assumptions I do in throwing away this data (it just does some additional tweaking on top of it). I can show the derivation if you like, but you probably already have it on file.
Your signal processing library is more complete than mine:). My library just does 1d DFT and 1d FFT, though the n-dimensional versions are numerically equivalent (you just have to chop up and reassemble the 1d spectrum).
BTW - I'm looking for a way of doing the equivalent of a FFT on a non-power-of-2 number of samples. Zero-padding produces an interpolated spectrum containing more information than I need. Any other scheme I can think of takes as much work as a DFT of the samples would. Any thoughts on this?
BTW, I just wanted to say that its been a pleasure discussing things with a person who actually knows what they're talking about and doesn't dodge questions:)
Likewise.
I should probably email you with my email address, as these articles will be archived soon, preventing further comments (especially the antimatter propulsion one). Is your listed email address accurate (modulo spam-removal)?
I've never heard of lunar mass drivers before, but I'm quite interested:) Care to elaborate?
A mass driver is a ground-based device that accelerates cargo to escape speeds. Usually they're based on electrical, magnetic, or electromagentic principles (configured as giant railguns, coilguns, or other such devices).
The advantage to this is that you don't need to carry any fuel at all with your cargo, so the only energy consumed is that imparted to the useful cargo. This would be atou 60 MJ/kg launching from Earth, or about 3 MJ/kg launching from the moon.
On Earth, you'd have to worry about heat shielding on the cargo, keeping a barrel around the launch path so you can take out the air or otherwise reduce turbulence, powering the device, and just finding somewhere to put it. On the moon, you have no atmosphere to interfere with things, lots of space to build, and lots of space for solar power generators of various types. Sending material from the lunar surface to lunar orbit or interplanetary space is beautifully easy. This is why the moon is often proposed as a source of raw materials for space-based construction.
Moving material in from the asteroid belt would be expensive, because there's a great difference in gravitational potential energy between the belt and earth's orbit. You could use a near-earth asteroid to reduce this problem, but the moon's in a very convenient location and facilities there would be useful for many purposes, so IMO it's probably the best choice to supply any construction near Earth.
Re. mining, it would actually be pretty easy to attach a mining facility to an asteroid, either on the surface or inside the asteroid itself.
If the asteroid is in danger of crumbling, you can always turn some of its material into fiberglass rope and wrap the asteroid in webbing to keep material from drifting.
The weak gravity of the asteroid itself is still enough to cause most crumbled material to return eventually (a few hours for a medium-sized asteroid).
Dear god, a tonne of anti-matter???? We're trying to get to stars, not take them out!. All kidding aside, the proposed mission to alpha centauri mentioned in the article - the only mission for which they proposed using pure antimatter/matter annihilation - was to take 1 gram of antimatter, to push a several hundred ton spacecraft at 0.4c to alpha centauri.
As you noted, they're building an antimatter-triggered fusion drive; different beast. Most of the energy comes from fusion, and most of the craft mass is reaction mass. They're still being very optimistic about their numbers.
If my old "interstellar drives" post is still on file, it gives the numbers for fusion drives with varying degrees of optimism. The most optimistic would get you a delta-V of 0.14C at 50% fuel.
That sounds like it's in line with the numbers you're quoting. If you can perfect fusion technology, you won't even need the antimatter.
The problem is that they're assuming that all of the fusion energy can go to kinetic energy of the reaction products. For most fusion reactions, you lose most of it to gamma rays. Even if you use one of the more difficult reactions and get all of the kinetic energy in the plasma, the plasma will cool *extremely* quickly due to radiative heat emission (rate of radiative heat loss is proportional to the fourth power of the temperature). You'll have to divert the exhaust to its final direction before substantial energy is lost. Good luck.
Thus, I question how realistic their assumptions are.
A more realistic fusion drive would just set off fusion bombs and count on most of the energy being emitted as light and gamma rays. Put your ship on one side of the blast, and photon pressure pushes it in the desired direction (at photon drives' horrible energy efficiency).
This is probably how a realistic pure-antimatter drive would work, too.
For all practical purposes, without a space elevator, a solar sail is economicly impossible (if it is even feasable, given stability issues), if the issue of lifting material from the earth is required (just think of how much a day's worth of water costs for an astronaut, even including water recycling). The only way to build such a thing would be to do it in space, from asteroids and the like.
The sailcraft itself would be light enough to build with material from Earth, though you'd still be much better off using a lunar mass-driver to supply it.
The laser array would have to be built using space-derived mass (though a lunar mass driver would probably be a cheap enough solution). However, if you can afford to build the laser array at all, you can afford to put the required facilities up there.
it requires a one-time investment in focusable mirrors orbiting near the sun. You take an asteroid and then spin it incredibly quickly (how quickly depending on the dimensions of what you want). You then keep light constantly focused on it at high temperatures. It'll steadily melt inwards, and flatten into a disc (which can then be cut up into sheet metal) after being cooled
It turns out that space mirrors are quite cheap, actually. You don't need any support structure; just lots of aluminized mylar and lots of spray-foam for structural ribbing.
Trying to melt and form the entire asteroid at once would probably result it being spun apart into globules (take a hose and shake spinning water blobs into the air for an idea of how this behaves). You could certainly use solar furnaces to aid in melting smaller quantities to smelt or to form by other means, though. This would be by far the cheapest way to melt large amounts of material.
Larry Niven proposed an interesting variant of your idea in a few of his sci-fi novels - putting water in the core of an asteroid, and then heating it to get it to expand into a bubble of rock to build a colony in. While a neat concept, it would probably also fail due to stability problems.
Slashdot Mirror
No, they won't. If energy becomed more expensive, people would just return to nuclear power. If you think that people are going to resign from driving, driers and air-conditioners, you're insane.
If it's forced on them incrementally, I think they might. Especially since the alternative with or without fusion power is to have an Evil Nuclear Plant right next to their city (fusion isn't much cleaner than fission for low-level waste).
It's a case of foolishness (hopefully) defeating foolishness.
Bottom line, if someone can come up with a cheap/light way to store electricity (either through a flywheel or through a superconducting coil), we'll still keep boosting my Exxon stock. And we'll do it by drilling in your backyard.
Fuel cells for cell phones should be appearing within the next 2-5 years or so. Prototypes already exist, and energy density is almost as good as a chemical fuel.
Fuel cells at present are a pain for things like cars because storing the hydrogen they use is very annoying (it's nowhere near as dense as a liquid). However, fuel cells that can process methanol are buildable, and will IMO probably be what brings fuel cell cars into the real world. Or you could just pump methanol into an interal combustion engine made of ceramics or otherwise made corrosion-resistant.
Methanol can't be easily produced by "recharging" fuel cells (unlike hydrogen), but you can make it by fermenting plants or by direct synthesis (burn CO2 incompletely in a hydrogen atmosphere and use fractional distillation on the combustion products).
In summary, power storage won't be a problem for long.
And don't forget the other things that are petroleum based - the industrial chemical industry , the plastics industry, and the fertilizer industry to name a few - when oil gets scarce (coming soon) everythings gets expensive.
Plastics can be produced from anything hydrocarbon-based. You can already do it easily from plant matter - petroleum just happens to be cheaper for the time being. Ditto most organic chemicals used in industry. Most fertilizers that I know of are based on minerals, not petroleum (a fertilizer provides mostly nitrogen, phosphorus, and potassium to plants, none of which petroleum is terribly rich in; typical fertilizers are based ammonium nitrate, potassium nitrate, or pick-a-random-phosephate, if I understand correctly).
Nothing that oil fields are vital for.
Current fusion engineering designs call for deuterium and tritium to be used as the fuel. You can easily get a virtually inexhaustible supply of deuterium from seawater. Tritium is another matter altogether. The only source available for this is fission reactors.
A fusion reactor would produce more than enough neutrons to breed tritium (and hopefully helium-3 as well).
Let that stream of plasma impact some new magnetic fields. It will push on the field lines, imparting momentum to them. This could accelerate the magnets generating those lines, even if they were on the other side of the wall. If those magnets were attached to wheels, could you actually make a "magnetic turbine", directly converting the kinetic energy of the plasma into mechanical work?
...'94? '95? can't remember) :).
Maybe a real physicist can enlighten me... Is this possible?
I'm not a real physicist either, but I faked being one well enough to get a free vacation out of it (IPhO
It turns out that you can get electrical power directly from a plasma stream by the Hall effect (similar to what you describe). Fire a stream of plasma through a magnetic field that's perpendicular to the direction of plasma motion, and you'll get negative charges deflected one way and positive charges deflected another way. Put electrodes beside the plasma stream to collect the charge, and you have a power source. Look up "magnetohydrodynamic generators" in your favourite encyclopedia for more information.
As for fusion plants, the best approach to tapping power that I've heard of is to put a big block of lead or concrete around the reactor, let it heat up from absorbing gamma rays and other crud coming out of the plasma, and run steam turbines off of it.
The downsides are that the D/He-3 reaction has a higher energy threshold, so it requires a better confined plasma to make D/He-3 energy production plausible. Also, the supplies of He-3 on Earth are limited. Of course this disadvantages leads to a interesting convergence of interest between fusion research and space exploration. There's plenty of He-3 (deposited from the solar wind) on the moon. So He-3 mining could be powerful reason to maintain a base on the moon.
I'd understood that He3 was one of the decay products that you get from neutron irradiation of a fusion reactor's lithium blanket. A few other articles that I'd read claimed that this was the main reason a lithium blanket was used (the idea was that a H2/H3 plasma fusing would transmute enough of the blanket into He3 to provide H2/He3 burning for power generation).
Is this still a viable option, or has it been shown to be impractical?
Pretty interesting idea. But that patch of empty sky is actually full of a lot of air. The heat has to pass through a lot of air before it gets free in space. Since conduction and convection are much more effective than radiation at transmitting heat, wouldn't you just end up making your local chunk of the atmosphere very warm?
:). As you raise the temperature, radiative cooling starts to dominate (transfer by conduction goes up linearly with the temperature difference).
Heat lost by radiation goes up as the fourth power of temperature - this is why you'd have the coolant hot enough to glow orange
If I understand correctly, the atmosphere's pretty good at blocking light in the thermal IR band, but you can see by example that visible and near-visible passes through quite easily - most of the sunlight shining on earth reaches the ground.
Heat the coolant up to a thousand degrees centigrade or so, and most of its emission will be in visible and near-IR (thermal IR is the band that objects at room temperature shine in).
Now, a week of cloudy days would make this system less effective, but you could get around that by using a nearby lake for heat storage until the sky clears again (though that won't be very nice for the inhabitants of the lake).
ok, so you have this plasma "floating" in the bottle at 12 gazillion degrees. the power goes out. doesn't your ball of plasma just eat it's way through anything it touches and head towards the center of the earth?
The plasma is very hot, but also very tenuous. You might have a few thousanths of a gram of matter in the reactor. As soon as this touches the reactor wall, it cools down to room temperature. Your reactor wall gets warm. That's about it.
Also, as your plasma is much less dense than air, if anything, a ball of plasma would rise.
You'd never get a ball. What happens is that as the containment field weakens, the donut-shaped mass of plasma expands until it touches the side of the container. In theory, if you had a "hole" in the containment field instead of the whole thing weakening, you could get a jet of what looks like flame, but this is next to impossible to do even if you're _trying_. The natural shape of the field is more-or-less uniform.
Plasma is just a hot, conducting gas. It follows gas law like everything else.
Perhaps... Heat Pollution would become a problem, but hopefully that would be worse than greenhouse-effect causing gasses that we release by Fossile fuels, if such gasses do in fact cause global warming (of which I'm not fully convinced.)
Getting rid of excess heat is easy enough if you decide to spend the money to do it - just set up a solar heat plant, point the mirrors at a patch of empty sky, and run it in reverse (use a heat exchanger to heat the working fluid from your waste heat, run it through the pipes in the mirror array, and let it beam its blackbody radiation out into space).
Yes, it takes a fair bit of power to move heat across this big a temperature differential (you need to heat the pipes up until they're glowing orange to get a decent rate of heat transfer). However, you'd only need to do it in the first place if you had power to burn.
To prevent confused responses - you're not building a giant toaster-oven. The heat you're stuffing into the pipes comes from the city or industrial park you want to cool off, via heat exchanger. You're not generating it from scratch.
In other words, either we get cheap, clean, nearly free unlimited energy, or our future is gonna look uncomfortably like those Mad Max movies.
Actually, what would be more likely to happen is a slow and painful shift to renewable forms of energy and a lifestyle that consumes less power.
Fossil fuels won't run out all at once - they'll get incrementally more expensive as they become scarcer and more difficult to extract. As price climbs, people will out of necessity start using less power, and alternative forms of power generation will become cost-competitive (even if their own cost doesn't change).
You'll see more houses with photocell shingles. You'll see a mirror-based heat plant or a wind farm next to most cities. More people will take public transit, because it's cheaper than driving. House design will depend more on passive heating/cooling and good insulation than on furnaces and air conditioners (they'll still be used; just on lower settings). People will have to hang-dry their clothes instead of tossing them in the drier. And so forth.
A Mad Max style catastrophy would only happen if energy supplies ran out all at once. A nuclear war or an asteroid strike could cause this, but I doubt fossil fuel problems will.
What about the supply of deuterium? It's not perpetual...
It's close enough to perpetual that it makes no difference.
About 1 hydrogen atom in 7000 is deuterium. Hydrogen accounts for 1/9th the mass of water, which means that the mass of deuterium we could extract from the oceans is about 1/63000th the mass of the oceans.
Assuming an average depth of about 3km and a coverage of 2/3 of the Earth's surface, you have about 1e18 tonnes of water on the planet. Let's say 1e-5 of that is deuterium that you can easily extract, and you get 1e13 tonnes of deuterium.
Fusion is about 1% efficient at turning mass into energy. Let's say that your fusion reactor is 0.01% efficient in total (pessimistic estimate). This gives about 1e13 J/kg, or 1e16 J/tonne.
The total amount of energy we could extract from fusing the oceans' deuterium with these fairly inefficient fusion plants is about 1e29 J.
If we want our fuel supply to last a million years, that gives a world power consumption of 3e15 watts. Far, far more than we consume now.
If, within the next million years, we build reactors good enough to fuse light hydrogen, we'll have enough fuel to last us until the sun burns out (several billion years). Or we could just ship in deuterium from elsewhere in the solar system. Either way, barring a *huge* population expansion, we won't run out of power - ever.
In my experience with high precision RF simulations on massive SGI multiprocessor machines, the OS (IRIX), the compiler (MipsPRO), and the communications package (OpenMP) do very little for you automatically. In order to achieve decent scalability you need to take your computing needs into consideration when you develop the architecture behind your simulation. If you, as the developer, don't distribute your calculations as evenly as possible, your simulation is going to run very, very poorly.
What the original poster was referring to, if I understand correctly, was distribution of processes or (in his case) web page requests across hardware nodes. This is handled by the OS and/or libraries in the setups that I've seen (admittedly few).
I definitely agree that the task has to be properly parallelized in the first place, but that wasn't what I was responding to.
Regarding the internal bus in a SMP system being faster than the communications network in a cluster, that's what I'd thought too. Then my prof got me to run subsets of the SKaMPI and NPB 2.3 benchmark suites on our cluster (MPI libraries on top of SCore running on a bunch of dual-processor Linux boxes). Running on 2nx1 nodes was faster than running on nx2 nodes for almost all of the tests. Caveat: this was for small n, and I can't guarantee that it wasn't just due to lousy process distribution or lousy SMP communication routines in the libraries.
YMMV.
For anything with a high communications load, SMP almost certainly wins, but for moderate communications loads, my experiences have made me leery of it.
Any task that is easily parallelized and has low internal communications requirements would run more effectively on multiple servers than on one SMP behemoth.
:). In the second case, you're looking at one of a handful of disk nodes within a much larger system (in all likelihood). For non-disk nodes, you wouldn't need PCI-64. For clusters that distributed disks over many nodes, your I/O bandwidth needs would be adequately served by PCI-32, and PCI-64 again becomes unnecessary.
:). You've made me think about the problem in more detail.
I have several problems with this generalization. First, parallelizing over multiple servers always adds overhead (in both $$$ and performance) of its own. How are you going to spread a load over multiple web servers? You need a load balancer, either the dedicated (pricey) hardware kind or a standard server converted over to load balancing service (which doesn't get you the greatest speed or scalability in the world). Even in a scientific application that you spread over several boxes, you need some kind of load balancer or traffic cop to get an equitable distribution of work.
You make a valid point, in that load-balancing is an issue. However, I'm assuming that in the case of a web server, if you have enough traffic to need more than one server, you have enough money to buy a hardware load-balancer to spread out requests (and a hardware firewall, if management has any sense).
As for scientific applications, you need load balancing regardless of whether the processors running the threads are in one box or in several. This is usually handled transparently by the OS, the compliler, the communications library, or a combination of the above (usually all of the above). This is standard for any high-performance computing project, and so doesn't add to your maintenance overhead. It also doesn't contribute substantially to the processor workload, so I don't see it as much of a concern for scientific workloads.
Second, let's not forget that two-way (and even four-way, if you were in the Xeon market to begin with) boxes have gotten much cheaper in the past year or two. Most of the important server availability features, like hot swap drives, hot swap power supplies, ECC RAM, 64 bit PCI, etc., are almost impossible to find on 1-way systems these days.
The last time I checked, n-way systems for n > 2 were still far more expensive than n one-way systems, but I haven't checked within the past couple of months. This might have changed, but I doubt it.
N = 2 was marginal, if I remember correctly.
ECC RAM support is available on several single-CPU motherboards; check your favourite vendor's site for a list of options (admittedly pricier than most of the boards, but not horribly so).
I'm assuming that hot swap power supplies aren't relevant. Your load-balancing hardware or (for a cluster) software will be able to detect malfunctioning nodes; this is essential for any cluster of significant size. A supply failing would be no different from any other component failing from a maintenance point of view (bad node is cut out of the loop by the load balancer, the hardware person gets paged, the node is swapped out and the old node serviced or gutted for parts).
PCI-64 support is a good point. If you have to support PCI-64, then it probably makes sense to build your cluster out of dual-CPU nodes, because the incremental cost of getting a dual-CPU motherboard will be low. Quad-CPU and higher will probably be less economical (quad cost diamonds the last time I checked). You'd only need PCI-64, though, if you either had a very large communications requirement (multiple very fast network cards per node), or if you were mounting a large RAID on the node (many controllers, many strings). In the first case, I can weasel out by claiming that you're outside of my stated problem domain (low communications bandwidth)
It's nice to get an interesting response, though
Has anyone done a cost-efficiency comparison of dual-cpu performance vs. a simple cpu when considering the costs involved (special SMP boards, etc.) In otherwords is it more economical to buy two web servers or one smp server with tons of ram? Do certain applications (cpu intensive obviously) save money with SMP systems verus others that depend on IO throughput, etc and what applications are those?
Any task that is easily parallelized and has low internal communications requirements would run more effectively on multiple servers than on one SMP behemoth. Web serving has zero internal communications requirement, and so falls into this category. Things like ray-tracing have low communications requirements when partitioned properly, which is why you use clusters as render farms instead of massively parallel Big Iron.
SMP has overhead from coherence operations, and more complex and expensive chipsets.
SMP benefits tasks that lend themselves to shared-memory implementations. It's a lot easier to toss ownership of memory pages back and forth inside an SMP machine than it would be to send modified pages back and forth across a network. I don't have examples of this kind of task offhand, but I'm sure they exist.
All of this is for CPU-bound tasks. For I/O bound tasks, you're still better off splitting it up into multiple machines if it's easily parallelized, but again I don't have good examples to illustrate with off the top of my head.
For more information, pick up a couple of good books on parallel computer architecture and parallel programming. Your local university's bookstore will stock these.
In fact plasma and laser accelerators, while a very good idea being explored in many places, are not ready for prime time yet. Currently, they are NOT cheaper yet ... either to build or to operate. And they likely won't be for many years.
My point is that for $6 billion you could, from scratch, build a plasma accelerator research facility the size of a university and fund it for years.
The payoff from doing this would be at the very least advancement in plasma accelerator research, and possibly development of a practical plasma accelerator. This would vastly benefit all future accelerator projects.
Spending $6 billion on a conventional accelerator won't provide this long-term benefit; future accelerators will still be extremely expensive. This is why I question the wisdom of *not* devoting the funding to accelerator design research, if that level of funding is available.
It seems to me that as accelerators become more and more expensive, it might be more cost-effective to devote accelerator funding to researching other classes of accelerator, instead of building larger and larger synchrotons and other conventional accelerators.
I've been following the development of plasma accelerators with much interest. They have a much greater rate of acceleration of particles, which means that an accelerator of a given energy could be much smaller (and hence hopefuly cheaper) than a synchroton. A good paper discussing the merits and problems with current plasma accelerator designs is at http://www.indiana.edu/~icfa/icfa12/node24.html .
Other promising types of accelerator almost certainly exist also.
I'm not saying that synchrotons should be abandoned because they're bad; on the contrary, they work quite well. I'm arguing that it would be _cheaper_ to invest accelerator funding in developing new types of accelerator, as opposed to building ever-larger synchrotons.
We consider the possible influence of a star light years away more than a million years from now, but fail to track most of the near-earth objects, which would pose a much more immediate threat to life on Earth than any star! Do they just pick a large object at random and say, 'Okay, how could this affect Earth?' or what?
The purpose of this research was almost certainly not to find threats to Earth - it was to get a better map of dim stars in our area, which lets us check some of our models of star formation and celestial dynamics in our area of the galaxy.
It also provides another data point for planetary system models (by giving us a better idea of how frequent close stellar encounters are, which certainly affect sytem formation), and (if they're finding stars in addition to just measuring their courses) a better estimate for the amount of normal matter in the universe (there has been debate over how much of it is bound up in dim stars and sub-stellar objects like brown dwarfs).
What most likely happened here is someone noticed that this star in their survey would come close to earth, and said "hey, neat, free publicity for our project".
Most astronomy is pure research - there won't *be* an application for most of it.
There are a few gotchas with this:
You're asking people to fork over a monthly subscription fee for the privilege of playing in your shard/continent/what-have-you. Your world had better be a lot more fun than the free alternatives that others will set up. You can do this; you just have to be willing to support several *good* full-time artists and game designers to build, maintain, and continually improve the universe that your players will be logging into.
Collaborative efforts could in theory compete with this, but in practice *good* artists and game designers who can work full-time (read: have maintaining the world as their paid day-job) can almost certainly provide a richer environment than a patchwork of volunteers could (this is art, not software).
If you're trying to get Joe Teenager to install and use your software, you're going to need a nice, packaged box with a good manual, and a fully-staffed tech support pool. What the user pays for when they buy the box is the ability to call you and have you hand-hold them through installing the software and connecting to your servers. You're *going* to have to sink a large amount of money and manpower into the division that handles this - if you have crappy support, most of your users will give up, and spread the word that your boxed software was impossible to use. Your servers will sit idle, and you'll lose a lot of money.
Subscription to your service should be optional, not mandadory. Ideally, you'd have about three versions of the box on the shelves, with different price tags - one that's just the client (with support for using the client), one that's a client bundled with a subscription to your service (with support for both), and one that's the client and _server_ with support for maintaining the server (charge them through the nose for this, because they're going to be *using* this tech support).
You don't have to be in control of the project itself, but you do have to have a paid team working on the software. This lets you address bugs customers find promptly (which you'll need early-on if you want to *keep* the customers), and continue to tweak the install and server selection routines (which you'll need to keep making more idiot-proof, and adapt to new server/client versions), and add features to the software that would help you make a more interesting game world.
All of this will cost a *vast* amount of money to set up, but will have a sustainable revenue stream coming back in. You'd have to do most of this regardless of whether the game was open or closed. Opening it means you save a bit on maintaining the software (because others' improvements and bug fixes are folded in for you), but forces you to work harder making your game world attractive (because anyone can set up their own world to compete with you).
Technical edges don't last past release anyways, so as far as I can tell you'd lose nothing by making the software itself Open or Free. Your IP is in the artwork and world design.
Computers have been working in pretty much the same way since then and it kind of reminds me the way cars continue to use the same archaic technology used back in 1900, only greatly refined.This is simply because it's easier (and cheaper) for manufacturers to maintain a single common base for their products for as long as possible before throwing it away and starting all over again. The choice, I believe, is ours: Mass production or revolution?
:).
Well, the way that machine code works is vaguely similar, except that the hardware implementation has changed on a *fundamental* level since then, and we have several layers of increasingly abstracted software on top that couldn't exist back then, and have developed several new branches of engineering involved in the design of hardware and software for computing devices...
No, no innovation or revolution there.
Sorry if I'm sounding a bit harsh. It's just that I've seen the "we've been using [foo] forever, we're shackled to it and should switch to something new and revolutionary" argument a few times now, and it almost always a) ignores a lot of fundamental change that's gone on over the years both in [foo] and in the design and use of [foo], and b) fails to propose an alternative.
What would be the "revolution" that you hope for?
Optical computers? The computer hardware layer would change utterly, but the design concepts used wouldn't, and the software would show almost no change at all. You'd also be stuck with a feature resolution no smaller than one wavelength of light (about half a micron to a micron).
Nanomechanical computers? Same deal (though without the feature size limit). You'd still be implementing digital logic, so all of the upper layers of the design and use of computers would stay the same.
The only thing that's fundamentally different both in design and use is quantum computing, and I'll bet that even that would have strong similarities on several levels.
Technological progress rarely happens by revolution - it's by evolution of existing ideas and devices, sometimes put together in new ways. Over time, the result of these changes can be profound, but the idea that you *must* turn the world on its head to move forward is a misconception.
Sorry if I'm being impolite; this is just something that's bothered me for a while
Me, I'm looking forward to being able to change eye or hair color with a swallowed capsule!
You'll be waiting a bit longer, then.
This experiment involved producing mice with what amounts to an on/off switch built in to their pigment genes. The dietary supplements just toggled the switch.
The research project demonstrated one possible switch mechanism, and showed that such switches could fairly easily be spliced into arbitrary genes, if you're in a position to modify an organism's DNA (easiest before conception).
Your pill would have to rewrite all of your DNA to accomplish the same effect. If you can do that, you don't need to bother with the switch.
Still a neat experiment, though. This will make it easier to study the function of genes, and easier to study certain types of genetic disease.
The real issue is that you can't just throw away half of the data. If all you care about is magnitudes, then yes, you can take the magnitude of the energies for a given frequency. However, if you want to reproduce the original signal, you can't.
:). My library just does 1d DFT and 1d FFT, though the n-dimensional versions are numerically equivalent (you just have to chop up and reassemble the 1d spectrum).
This would be the case if I was throwing away all of the real or all of the imaginary data, but I'm keeping half of each.
The spectrum really is perfectly symmetric for the real component and antisymetric for the imaginary component. Deviations from this symmetry would cause an imaginary component to exist in the input signal, which can a priori be assumed not to be the case. Phase in the real signal is encoded in the ratio of real and imaginary spectrum components for a given frequency.
The handwaving argument from information theory is that because I'm only encoding N values for N samples (the real components of the samples), I only need N components out to retrieve all of the information (the real and imaginary components of half of the spectrum). The full spectrum has a factor of two redundancy.
Again, the DCT makes the same assumptions I do in throwing away this data (it just does some additional tweaking on top of it). I can show the derivation if you like, but you probably already have it on file.
Your signal processing library is more complete than mine
BTW - I'm looking for a way of doing the equivalent of a FFT on a non-power-of-2 number of samples. Zero-padding produces an interpolated spectrum containing more information than I need. Any other scheme I can think of takes as much work as a DFT of the samples would. Any thoughts on this?
BTW, I just wanted to say that its been a pleasure discussing things with a person who actually knows what they're talking about and doesn't dodge questions :)
Likewise.
I should probably email you with my email address, as these articles will be archived soon, preventing further comments (especially the antimatter propulsion one). Is your listed email address accurate (modulo spam-removal)?
I've never heard of lunar mass drivers before, but I'm quite interested :) Care to elaborate?
A mass driver is a ground-based device that accelerates cargo to escape speeds. Usually they're based on electrical, magnetic, or electromagentic principles (configured as giant railguns, coilguns, or other such devices).
The advantage to this is that you don't need to carry any fuel at all with your cargo, so the only energy consumed is that imparted to the useful cargo. This would be atou 60 MJ/kg launching from Earth, or about 3 MJ/kg launching from the moon.
On Earth, you'd have to worry about heat shielding on the cargo, keeping a barrel around the launch path so you can take out the air or otherwise reduce turbulence, powering the device, and just finding somewhere to put it. On the moon, you have no atmosphere to interfere with things, lots of space to build, and lots of space for solar power generators of various types. Sending material from the lunar surface to lunar orbit or interplanetary space is beautifully easy. This is why the moon is often proposed as a source of raw materials for space-based construction.
Moving material in from the asteroid belt would be expensive, because there's a great difference in gravitational potential energy between the belt and earth's orbit. You could use a near-earth asteroid to reduce this problem, but the moon's in a very convenient location and facilities there would be useful for many purposes, so IMO it's probably the best choice to supply any construction near Earth.
Re. mining, it would actually be pretty easy to attach a mining facility to an asteroid, either on the surface or inside the asteroid itself.
If the asteroid is in danger of crumbling, you can always turn some of its material into fiberglass rope and wrap the asteroid in webbing to keep material from drifting.
The weak gravity of the asteroid itself is still enough to cause most crumbled material to return eventually (a few hours for a medium-sized asteroid).
Dear god, a tonne of anti-matter???? We're trying to get to stars, not take them out!. All kidding aside, the proposed mission to alpha centauri mentioned in the article - the only mission for which they proposed using pure antimatter/matter annihilation - was to take 1 gram of antimatter, to push a several hundred ton spacecraft at 0.4c to alpha centauri.
As you noted, they're building an antimatter-triggered fusion drive; different beast. Most of the energy comes from fusion, and most of the craft mass is reaction mass. They're still being very optimistic about their numbers.
If my old "interstellar drives" post is still on file, it gives the numbers for fusion drives with varying degrees of optimism. The most optimistic would get you a delta-V of 0.14C at 50% fuel.
That sounds like it's in line with the numbers you're quoting. If you can perfect fusion technology, you won't even need the antimatter.
The problem is that they're assuming that all of the fusion energy can go to kinetic energy of the reaction products. For most fusion reactions, you lose most of it to gamma rays. Even if you use one of the more difficult reactions and get all of the kinetic energy in the plasma, the plasma will cool *extremely* quickly due to radiative heat emission (rate of radiative heat loss is proportional to the fourth power of the temperature). You'll have to divert the exhaust to its final direction before substantial energy is lost. Good luck.
Thus, I question how realistic their assumptions are.
A more realistic fusion drive would just set off fusion bombs and count on most of the energy being emitted as light and gamma rays. Put your ship on one side of the blast, and photon pressure pushes it in the desired direction (at photon drives' horrible energy efficiency).
This is probably how a realistic pure-antimatter drive would work, too.
For all practical purposes, without a space elevator, a solar sail is economicly impossible (if it is even feasable, given stability issues), if the issue of lifting material from the earth is required (just think of how much a day's worth of water costs for an astronaut, even including water recycling). The only way to build such a thing would be to do it in space, from asteroids and the like.
The sailcraft itself would be light enough to build with material from Earth, though you'd still be much better off using a lunar mass-driver to supply it.
The laser array would have to be built using space-derived mass (though a lunar mass driver would probably be a cheap enough solution). However, if you can afford to build the laser array at all, you can afford to put the required facilities up there.
it requires a one-time investment in focusable mirrors orbiting near the sun. You take an asteroid and then spin it incredibly quickly (how quickly depending on the dimensions of what you want). You then keep light constantly focused on it at high temperatures. It'll steadily melt inwards, and flatten into a disc (which can then be cut up into sheet metal) after being cooled
It turns out that space mirrors are quite cheap, actually. You don't need any support structure; just lots of aluminized mylar and lots of spray-foam for structural ribbing.
Trying to melt and form the entire asteroid at once would probably result it being spun apart into globules (take a hose and shake spinning water blobs into the air for an idea of how this behaves). You could certainly use solar furnaces to aid in melting smaller quantities to smelt or to form by other means, though. This would be by far the cheapest way to melt large amounts of material.
Larry Niven proposed an interesting variant of your idea in a few of his sci-fi novels - putting water in the core of an asteroid, and then heating it to get it to expand into a bubble of rock to build a colony in. While a neat concept, it would probably also fail due to stability problems.
well for realtime video at least --- it it sloooooooow.
Actually, the FFT is quite fast. It works in near-linear time on the data (O(n log n)). It's the DFT that's the slow version.