My understanding is it REQUIES VERY HIGH temperatures to Dissacociate water on the order of 3500 degreesf plus (PS Dont ever try to quelch a thermite reaction with water:)
This has nothing to do with the temperature needed to dissociate water.
It has to do with the fact that aluminum will happily strip oxygen out of water (3H2O + 2Al -> 3H2 + Al2O3 + 818 kJ).
This doesn't happen at room temperature (due to the activation energy and the oxide skin on aluminum), but at thermite temperatures it will most certainly happen. Aluminum is a very reactive metal.
1.the issue with adoption of hydrogegn is the entrenched position that fossil-fuels have. it's not that hydrogen is harder to use, it's that there is billions invested in transport, wells, autos, etc, all which would have to change. not to mention the industry mogul's (and current usa administration's) vested interest. in additon, you don't need so many specialized resources to create hydrogen, eh - just some electricity and water - think of the threat that poses to the oil hegemony...
2.there are always energy costs to creating portable forms of energy, but that's the issue, not that it's more energy-expensive to create hydrogen than to use it. add up the costs in shipping oil around the planet. not cheap. the real benefit is that oil is portable once extracted.
Both of these problems are addressed if you burn CO2 in a hydrogen atmosphere to produce methanol. Methanol can be stored and transported like any other volatile liquid fuel, which means you can use the existing infrastructure, and can use it automobiles with minimal modification (though you'd want a ceramic engine block to avoid corrosion in the long term).
The article directly mentioned methanol production as an application of a hydrogen plant.
Transport and infrastructure aren't the problem.
The real reason why this won't be done any time soon is that gasoline is cheaper to produce per litre (by taking it out of the ground) than methanol (which must be made from scratch, by direct synthesis or farming and fermenting).
When/if oil and natural gas reserves are depleted, it will become cost-competitive. Before then, it won't be.
If I understand correctly, broadband service is now a "commodity" - a product sold with faily low mark-up over cost.
Given that offering broadband services requires fairly substantial infrastructure upgrades (costing a pretty penny), why would any provider in their right minds jump into the market now?
I was part of a hole-in-the-wall company that was looking at getting into the ISP market shortly after it stabilized as a commodity (back in the days of 33.6 modems). Our conclusions were that we'd make very little money from offering Internet service, and that we'd only make money at all if the service we offered was lousy. And that was with our upstream connection mostly paid for by other means.
Could it be that there is no conspiracy?
[Disclaimer: I am not intimately familiar with the economics of offering broadband. If you have more detailed information, by all means post it.]
Problem is that all the kinetic energy still ends up in our system. One big piece is bad. Split that one big piece into several smaller pieces, and it's even worse. But take things to an arbitrary limit, where you pulverize the entire asteroid down to dust.
Now all that dust impacts the atmosphere, heats to incandescence, and vaporizes. Do *you* want to be in the hemisphere where *that* happens? Imagine New York City under the glare of 70 trillion E-Z-Bake Ovens.
If you fragment the asteroid when it's far enough away from Earth (months earlier in its orbit), and give the fragments enough energy that they're not going to just drift back together, then most of the fragments would likely miss Earth.
The key is fragmenting it when it's far enough away, so that the fragments have time to spread.
We are certainly in no position to prevent an impact with a large comet, meteor or asteroid.
Sure we are, if we can see it coming from far enough away.
The greater the advance warning of an asteroid's orbit passing through Earth, the smaller the perturbation to the asteroid's course needed to prevent its orbit from passing through Earth. Hitting even something as big as a planet is quite a fluke - it doesn't take much to prevent it.
With a small enough perturbation required, and a big enough collection of fusion bombs under the surface of one side of the rock, you could certainly nudge it into a slightly different orbit that would miss the Earth.
If you're feeling more environmentally friendly and have more money, you could also send a very large ion-engine tug out to perform orbital correction. More time means both a smaller needed change and more time for the tug to make the correction - smaller tug in both cases.
Fusion bombs to kick up rocks (causing the rest of the asteroid to move by reaction) are probably the most practical course, though.
The prerequisite for either of these approaches is knowing the asteroid is going to strike many months (ideally several years) in advance. Without a very thorough cataloging of near-earth objects, this won't happen.
Air resistance effects are negligable if your rocket's cross-sectional mass is much greater than the cross-sectional mass of the atmosphere it'll be plowing through (15 tonnes per square metre), or if it does most of its acceleration outside most of the atmosphere.
Correction: This is 10 tonnes per square metre if you're going straight up (about 15 pounds per square inch).
1. it does NOT have to be vertical(watch what the shuttle does soon after launch), so the acceleration can be done over longer distance, so less G's. think rail's in hundreds of miles. Expensive, yes, but less than 100 million per launch of shuttle.
I've done the calculations. Have you?
At 0.5 km/sec, and a maximum radial acceleration of (say) 10 gravities, your minimum turning radius is 2.5km - bigger than the 1km gun!
If you're building a horizontal gun and making the end turn up, turning radius gets _worse_, because of the higher muzzle velocity. It goes up as the _square_ of the velocity! You need a tower high enough that you might as well make the whole gun a tower.
Mount Everest is 4.4 km high. If you carve a giant channel in it, so that your gun gracefully curves, you get a maximum muzzle velocity of around 0.66 km/sec. Still very, very low.
If you just run a straight gun up the side of a mountain the size of Mt. Everest, you get a straight gun around 6 km long. At 10 gravities maximum acceleration (as per previous post), this gives you 6e5 J, or a velocity of 0.77 km/sec.
Still not enough to make a worthwhile difference.
Bear in mind also that tilting the gun at an angle, like you would going up the side of a mountain, gives you much more atmosphere to go through on the way up. If you try to turn the craft in the atmosphere, you're still forced to turn slowly, and your acceleration limit will be much lower than for a turning gun barrel, making the turning radius much larger (turning radius is inversely proportional to radial force).
2. even if it did, you could just build it up the side of a tall mountain, and have it curve gently up, which is kinda the most likly solution anyway, as it put you higher in the air, so less air resistance, closer to orbit, that type of thing.
Air resistance effects are negligable if your rocket's cross-sectional mass is much greater than the cross-sectional mass of the atmosphere it'll be plowing through (15 tonnes per square metre), or if it does most of its acceleration outside most of the atmosphere.
For a conventional heavy-payload rocket, both of these conditions are true, and atmosphere resistance doesn't matter.
You don't build a magrail to give your spacecraft orbital velocity. Of course that's silly, for the reasons given above. You use it to give you some small PART of your velocity. This is extremely beneficial.
This turns out not to be the case.
First, calculate how fast a magnetic launcher can fire a craft.
Remember, the launcher needs to be pointed upwards. You can't just turn the craft at the end - G force limits would require a very large turning radius for this.
Assume a vertical launcher length of 1 km maximum.
Assume a maximum acceleration for delicate cargo (like people or delicate equipment) of 10 gravities (100 N). I'm ignoring gravity's contribution; accelerating upwards at 10 gravities, the cargo would feel 11 gravities of force.
This gives an energy transfer over the length of the gun of 1000m * 100 N = 1e5 J, corresponding to a velocity change of about 0.45 km/sec.
To get to low orbit - not geosynch or escape - you need a delta-V of about 8 km/sec. If you're burning liquid hydrogen, with a specific impulse in the 4000 N*s/kg range, you'll need a rocket that's ( 1 - exp(-8000/4000) ) = 86% fuel.
If you get 0.45 km/sec for free, you need a rocket that's ( 1 - exp(-7750/4000) ) = 85% fuel.
Magnetic launching gained you 1% of the rocket's mass for cargo. Not much.
If you can launch from an airless body like the moon, then you can build much longer launchers tangentially to the surface, which would be extremely useful for lifting payloads. However, I've yet to see any proposal for an earth-based launching scheme that would give a substantial benefit without an astronomical cost.
Building a magnetic accelerator several tens of kilometres long might work, but that would be insanely expensive, requiring huge traffic volumes to pay itself off. Building a laser-based launcher [basically a jet with a ground-based laser as the heat source] looks attractive at first due to long path length, but has strong limits on energy density (you don't want to ionize the atmosphere the beam travels through, or you'll get a reflective plasma scattering your light). A space elevator would be even more expensive than a magnetic launcher, would require advanced materials that we presently don't have, and could cause devastating amounts of damage if sabotaged.
In short, I'm doubtful of anything better than chemical rockets for launch of delicate cargo from Earth showing up any time soon. Space, of course, has considerably more interesting possibilities.
If you can build a thousand-gravity accelerator, then you might be able to send up sturdy cargo. However, that too would require very high volumes to be economically practical.
Re:From field/alloy interaction.
on
Magnetic Fridge
·
· Score: 2
But the question that has to be asked is why spin a disc through a permanent magnet? Why not use a series of electro-magnets to produce a moving magnetic field?
Because a motor and one permanent manget are a lot simpler and cheaper than a stack of electromagnets.
Having one hot location (by the magnet) and one cold location (far from the magnet) also makes the heat exchanger rig simpler. With electromagnets, any fixed position near the magnets would alternately be hot and cold (not easy to pump heat with).
In short, it's because there's lots of extra complexity and no real benefit from using electromagnets for the fridge.
Problems of the type you descripe wouldn't benefit much from OOP. I'd solve them myself in C, but that's mainly just because I don't *know* Fortran (fortran is very good at math problems with straightforward methods of solution, and parallelizes easily).
My most recent project, on the other hand, involved simulating a computer system (I won't go into detail; wait until after I've published). This is a perfect scenario for OO - a computer system is built up out of many parts, many with similarities, but with different implementations. If you need a cycle-by-cycle or event-by-event simulation of a computer system, it makes sense to build up models of all of the parts that will be interacting and put them into an OO hierarchy. This lets you avoid copying code, and lets you change things much more easily than in a straight C environment.
In a previous project, I was building the skeleton for a program that would have to be rewritten for many different purposes. Some parts would stay the same between instances, and these were what I was building. I made the mistake of writing it in C, with the idea of upgrading it to C++ later. I ended up faking OO-isms by calling function pointers hither and yon to abstract things. When maintenance of the project was finally turned over to someone else, they started the migration to C++.
In a complex program where you have a lot of abstraction going on between different types of component, an explicitly OO language will almost always make life much easier.
Would it be possible to use ordinary parabolic satellite TV antennas for radio telescopy? Could they be combined to create a huge radio telescope?
If that's possible, maybe when people are not using their satellite antennas for TV they could be combined to create a world-wide radio telescope.
You could in principle do this, but in practice there are problems.
The main problem is that the electronics in the detector used with the dish are completely unsuited to radio astronomy. To use a radio telescope as a part of the array, you need a high-fidelity sample of the radio signal being received, timestamped to atomic-clock accuracy. A satellite TV pickup doesn't have a sub-nanosecond-accurate clock, and won't give you a digitization of the raw signal. Instead, it looks for strong signals in specific, narrow bands and blindly decodes them through combined analog and digital means (i.e. it treats everything it hears as a TV signal).
A secondary problem is that your satellite dish is pointed directly at a strong source of radio noise in the frequencies it's tuned to detect (the satellite).
The idea is a great one, but because you'd need to completely replace the electronics rig with something far more expensive, a better approach might be to sell radio telescope array "kits" built from stock parts and forget about using peoples' TV dishes.
This would probably be quite practical from an engineering standpoint, as most of the parts (including timestamping radio sampling boxes) can be bought off-the-shelf. I have no idea if enough amateur astronomers would buy these for them to be marketable (they wouldn't be cheap - tens of thousands of dollars per kit).
Could a space-based array be designed so that once it reaches its target location, it spreads itself out, gradually increasing the distance between its elements in a coherent manner, thereby increasing the effective size of the array over time? I would assume that a space-borne array would already be designed with plenty of fuel/rocketry for compensating for massive objects passing nearby and tugging on its corners... The same principle might be handy for adjusting/balacing the spacing between elements if an asteroid hits the jackpot, or a failure is detected.
Yes, but you'd be trading off angular resolution against aliasing artifacts (the less of your aperture is filled, the worse aliasing artifacts will be, even when you assume constant sources and integrate over time). IMO, you'd be better off just adding more satellites:).
By coincidence, I recently did the calculations for the size of a metre-band radio telescope array needed to resolve features 100 km in size at a distance of 10 light-years (enough to resolve the aurorae of earth-sized planets, show thunderstorms on gas giants, and so forth). You'd need thousands of radio telescopes in solar orbits out to a radius of about 4 AU, but you could do it. Put them in eccentric polar orbits (i.e. away from most of the junk in the ecliptic), add excellent GPS-style beacons in precisely known orbits (constantly observed from Earth) to let the satellites track themselves, and you could get a very nice radio telescope for a surprisingly modest price (cheap satellites, well-known technologies and electronics, and the benefits of mass production, since you'll be making a thousand or more of them).
Such a telescope would be able to see aurorae and civilization-induced radio junk from Earth-sized planets out to around 10 light-years, map the magnetospheres of Earth-like planets and see detailed magnetic features in gas giants out to about 100 light-years, and get very detailed pictures of the outer envelopes of stars out to about 1000 light-years. It would be a very useful project.
From field/alloy interaction.
on
Magnetic Fridge
·
· Score: 5, Interesting
Does the energy that does the work to remove the heat from the refridgerated side of the setup come from the work being done to insert the Gadolinium alloy into the magnetic field(and to remove it) or is the energy input done simply by the circulation of coolant fluid through the heated/cooled alloy.
If the alloy changes temperature based on its magnetic field environment, then the work that goes into the heating and cooling is from the force needed to move the alloy through the magnetic field in the rig they're using.
I agree, sorry if my comment was mis-understood. The idea is to use a gigabit switch for apt-to-apt connections and a linux router for outbound traffic going to the ISP. I would not dream of using a PC as a gigabit switch, just think how many PCI slots you'd need for an apt complex!!
Also, I really don't think the IP's will be a problem, it really just depends on the ISP. For example, my ISP sells a high speed DSL package with 30 IPs. The apartment complex I live in has only 20 apts.
Neat. Sounds like the project is well in-hand, then.
this will work fine since hes not routing internal traffic. the only thing being routed/firewalled is internal to external and ext to int traffic.
He was talking about routing streaming video within the building. And having gigabit connections to *each* apartment. This suggests he's planning to set up a system with more than 100 Mbit/apartment load internally (I agree that external load would likely be low enough to route with a PC).
He's planning on running gigabit ethernet to every apt for apt-to-apt networking and use highspeed ADSL with several static IPs (one for each apt) for outbound internet access. [...] Of course he'll be using Linux and FreeBSD for just about everything from the router to the "apt game servers" and video on demand servers.
It sounds like either your friend doesn't have a good handle on the technologies involved with this, or there was some miscommunication between the two of you.
It sounded good up until "use Linux and FreeBSD for the router".
You need something better than a PC to route many apartments' worth of gigabit ethernet to each other. A PC doesn't have the internal bandwidth for more than one gigabit connection. If you're using an off-the-shelf gigabit ethernet hub or router, it'll be running its own embedded OS from the vendor (if it's complex enough to run anything at all). If you're using a souped-up non-PC workstation as the router... you're spending far more than you have to for a simple router.
In a similar vein, you'll have an interesting time getting enough static IPs for a medium-sized apartment building without a fight. Maybe when IP6 finally takes over.
This sounds like a really cool project, and your friend deserves praise for trying to pull it off, but he'd better take a close look at the tools he's planning to use for it, and make sure that he's using the right tools for the right parts of it.
A few people have been speculating about 3D processors based on this or similar technology. While this is a neat way of building memory, don't get your hopes up for 3D processors any time soon.
What they've done, according to the article, is deposit several layers of thin-film transistors on top of a more or less standard chip.
These transistors will be *slow*. That's fine for something you're using to replace flash, but not fine for a processor. The hard problem of building high-quality transistors in a multi-level structure has not been solved.
The other problem is heat. With a hierarchically-designed memory array, you can make a larger array without power per access going up very much (at the cost of a very small amount of extra delay). This means that packing ten times as much memory into the same chip area won't cause much of a heating problem.
The core of a microprocessor, on the other hand, is pretty much all active at once (or mostly active). You have calculation results flying to reservation stations everywhere, you have a lot of fully-associative arrays being indexed, and you have a lot of logic churning away. Packing this into a tenth the area would make the used area much, much hotter (remember Newton's law of heating and cooling - you need the same heat flow from a tenth the area, so ten times the heat difference between chip and environment).
The good news is that you might be able to put the L2 cache in higher layers with technology similar to this and save space that way, but this is a one-time saving, with a performance penalty (until the holy grail of stacked transistor technology is found).
The problem is hubs. I have yet to see a good gigabit hub for under $2k or so. Most of the gigabit-compatible hubs offered use gigabit for uplink, and a handful of 100-base-T links for the rest of the ports.
What sort of crack are you on? Check out the Linux [linuxrouter.org] Router Project [sourceforge.net], freesco, etc. I own an ISP [stormforge.net] that is Cisco free using entirely LRP based routers and firewalls.
And your internal routers are either not routing (saturated) gigabit traffic through multiple cards, or not running on commodity hardware.
Even a hub - not a router, which is more expensive - that can handle more than one gigabit connection at full data rate costs about $2000 US. Find me a better price, with a link. My own numbers are from dlink (http://www.dlink.com). The links you posted describe _software_. The problem is expensive _hardware_. The machines you describe do not have the internal data transfer bandwidth to handle multiple saturated gigabit interfaces, so I'm puzzled as to why you even mention them as examples.
I think for a router you will find that in practice it will be pretty difficult to saturate the 32-bit 33MHz PCI bus.
I am a-priori assuming an application that will saturate gigabit ethernet channels. Otherwise there's no reason to use gigabit at all, as you point out.
More than one saturated gigabit ethernet interface would certainly swamp a 32-bit 33 MHz PCI bus.
Joe user's LAN and cable modem won't do this. A fileserver with a single IDE drive won't do this.
Something like, say, a communications-limited distributed computing project sure would, and most interesting computing problems are communications-limited when scaled up past one box.
Any application that requires streaming uncompressed (lossless) video streams between machines at better-than-NTSC resolution would also require better bandwidth than 100-base-T provides. Why someone would use multiple machines for something like this is left as an exercise for the reader.
Either way, both I and the post I was replying to were assuming the existence of some need for gigabit ethernet.
Hi definition uncompressed video is more than 100Mb/sec - but 1Gb/sec over copper is on the horizon.
Gigabit ethernet, at least, is fairly cheap right now. Check dlink's site for a fair idea of what pricing is like.
Gigabit will run over standard cat-5 cable (it actually just runs slower signals in parallel over multiple pairs), so you won't even have to rewire for it.
The problem is hubs. I have yet to see a good gigabit hub for under $2k or so. Most of the gigabit-compatible hubs offered use gigabit for uplink, and a handful of 100-base-T links for the rest of the ports.
Has this changed in the past 6 months or so?
Using a PC as a router in place of a hub isn't an option, as one gigabit ethernet card will come very close to saturating a 32-bit 33 MHz PCI bus. Start streaming large amounts of data through the house and the router will fail to handle the traffic.
I expect that eventually we'll see a chemically-pumped laser rifle or pistol, about the same size as a normal rifle or pistol, with an optical cavity where the barrel would be, powered by cartridges of solid fuel that are fed by a mechanism similar to the one that feeds cartridges consisting of case/primer/powder/bullet.
While this would probably work, I'm having trouble seeing why someone would use this instead of a gun.
The big problem with lasers is that you can do a lot more damage to a target by hitting it with a fast-moving slug than by heating it. Lasers are useful when you have a target that's far away and moving erratically enough that you can't fire slugs at it, but personal weapons are usually used at much shorter range. This suggests a chemically-pumped kinetic slug weapon might better fit infantry's needs... which is what they currently have.
As for shooting down missiles, one of the more interesting ideas in the Star Wars project was guided bullets. If you can guide a bullet - or even a countermissile, as is currently being tried - accurately enough to shoot down an incoming rocket, this method will likely be cheaper than using a laser to destroy the same rocket.
There might still be scenarios where you'd rather use a laser than a slug-based weapon. I'd certainly be interested in hearing suggestions (I certainly don't claim to have thought of everything; it's just that lasers aren't the best solution for the scenarios that I _have_ thought of).
As for the Dreamcast, don't forget the additional cost of the modem (if you like crawling) or the ethernet adaptor (if you can find one).
The Dreamcast ships with a modem. You don't need to buy one.
Turn off images in your browser, and bandwidth is not a concern (this was an email machine, remember?). I do this regularly when the university's internet pipe gets swamped.
Do you have any idea what it takes to get the surfadces flat enough? How long it takes to design a coating, and what sort of processes it takes to apply it?
Sure. And it's about an order of magnitude simpler than the processes involved in building the integrated circuit chips used in RAM. If you can find a magical method of producing RAM as simply and cheaply (per unit storage capacity) as hard drive platters, go and patent it and make a bundle.
Besides, people don't want insanely powerful computers. The computer manufacturers want people to buy them, so they make sure there's no decent alternative.
And I'm sure the conspiracy's mind-control satellites force you to upgrade yearly, too.
I got by with a 286 well into the days of Pentium machines. When I upgrade, I buy middle-end systems. If I want a system to act as a router or fileserver or other low-load system, I use a half-decade-old hand-me-down.
If I wanted a low-heat x86 box (at the expense of computing power), I'd buy a socket-7 motherboard and put a Cyrix chip in it (or a chip from whoever bought them; I don't remember if they're still around under the Cyrix name).
If I wanted a lower-power system that was still reasonably powerful, I'd follow another poster's suggestion and by an iMac.
If you can't find alternatives, you're not looking hard enough.
The point is that we desperately need processors that produce less heat and use less energy. If you take a moment to think about it, it's totally ridiculous that we need so many noisy fans inside a computer that someone's using to compose an email.
If you're using a high-end computer solely to compose email, I'd argue that the problem isn't the hardware.
Heck, if power is a concern, buy a Dreamcast and use the web client to access Hotmail. $50, and you get a low-power embedded box that you can read and write email and even play games on.
Desktop systems are overpowered because people want to be able to run insanely high-powered applications on them, no matter how much of a waste this is when they're not playing Quake XIV.
It's even more ridiculous when you consider that some graphics processors require a fan as well, and so does the power supply.
Same thing. A real-time realistically rendered 3D environment requires one hell of a lot of computing power to generate. This means heat. If you're just answering email, buy a PCI Rage XL card and save on the fan and heatsink.
Now if only they'd come up with a breakthrough that will make fast, long lasting, solid-state hard drives a reality.
They're called "flash cards".
If you want to store gigabytes of images or gigabytes of game install files, however, they won't be sufficient.
RAM is harder to make per unit storage space than a magnetic platter. This is just the nature of the universe - RAM is intrinsically more complex. A magnetic platter is just a flat surface with the right kind of coating; it doesn't get much simpler than that. You can buy a solid-state drive off the shelf right now, but the the cost will reflect the fact that it's harder to build, and this will continue to be the case for quite a while.
In summary, the problem isn't the technology, it's the fact that people *want* insanely powerful computers, with large amounts of storage, for the lowest price that still gives them the power and space they crave.
Slashdot Mirror
My understanding is it REQUIES VERY HIGH temperatures to Dissacociate water on the order of 3500 degreesf plus (PS Dont ever try to quelch a thermite reaction with water :)
This has nothing to do with the temperature needed to dissociate water.
It has to do with the fact that aluminum will happily strip oxygen out of water (3H2O + 2Al -> 3H2 + Al2O3 + 818 kJ).
This doesn't happen at room temperature (due to the activation energy and the oxide skin on aluminum), but at thermite temperatures it will most certainly happen. Aluminum is a very reactive metal.
1.the issue with adoption of hydrogegn is the entrenched position that fossil-fuels have. it's not that hydrogen is harder to use, it's that there is billions invested in transport, wells, autos, etc, all which would have to change. not to mention the industry mogul's (and current usa administration's) vested interest. in additon, you don't need so many specialized resources to create hydrogen, eh - just some electricity and water - think of the threat that poses to the oil hegemony...
2.there are always energy costs to creating portable forms of energy, but that's the issue, not that it's more energy-expensive to create hydrogen than to use it. add up the costs in shipping oil around the planet. not cheap. the real benefit is that oil is portable once extracted.
Both of these problems are addressed if you burn CO2 in a hydrogen atmosphere to produce methanol. Methanol can be stored and transported like any other volatile liquid fuel, which means you can use the existing infrastructure, and can use it automobiles with minimal modification (though you'd want a ceramic engine block to avoid corrosion in the long term).
The article directly mentioned methanol production as an application of a hydrogen plant.
Transport and infrastructure aren't the problem.
The real reason why this won't be done any time soon is that gasoline is cheaper to produce per litre (by taking it out of the ground) than methanol (which must be made from scratch, by direct synthesis or farming and fermenting).
When/if oil and natural gas reserves are depleted, it will become cost-competitive. Before then, it won't be.
If I understand correctly, broadband service is now a "commodity" - a product sold with faily low mark-up over cost.
Given that offering broadband services requires fairly substantial infrastructure upgrades (costing a pretty penny), why would any provider in their right minds jump into the market now?
I was part of a hole-in-the-wall company that was looking at getting into the ISP market shortly after it stabilized as a commodity (back in the days of 33.6 modems). Our conclusions were that we'd make very little money from offering Internet service, and that we'd only make money at all if the service we offered was lousy. And that was with our upstream connection mostly paid for by other means.
Could it be that there is no conspiracy?
[Disclaimer: I am not intimately familiar with the economics of offering broadband. If you have more detailed information, by all means post it.]
Problem is that all the kinetic energy still ends up in our system. One big piece is bad. Split that one big piece into several smaller pieces, and it's even worse. But take things to an arbitrary limit, where you pulverize the entire asteroid down to dust.
Now all that dust impacts the atmosphere, heats to incandescence, and vaporizes. Do *you* want to be in the hemisphere where *that* happens? Imagine New York City under the glare of 70 trillion E-Z-Bake Ovens.
If you fragment the asteroid when it's far enough away from Earth (months earlier in its orbit), and give the fragments enough energy that they're not going to just drift back together, then most of the fragments would likely miss Earth.
The key is fragmenting it when it's far enough away, so that the fragments have time to spread.
We are certainly in no position to prevent an impact with a large comet, meteor or asteroid.
Sure we are, if we can see it coming from far enough away.
The greater the advance warning of an asteroid's orbit passing through Earth, the smaller the perturbation to the asteroid's course needed to prevent its orbit from passing through Earth. Hitting even something as big as a planet is quite a fluke - it doesn't take much to prevent it.
With a small enough perturbation required, and a big enough collection of fusion bombs under the surface of one side of the rock, you could certainly nudge it into a slightly different orbit that would miss the Earth.
If you're feeling more environmentally friendly and have more money, you could also send a very large ion-engine tug out to perform orbital correction. More time means both a smaller needed change and more time for the tug to make the correction - smaller tug in both cases.
Fusion bombs to kick up rocks (causing the rest of the asteroid to move by reaction) are probably the most practical course, though.
The prerequisite for either of these approaches is knowing the asteroid is going to strike many months (ideally several years) in advance. Without a very thorough cataloging of near-earth objects, this won't happen.
Air resistance effects are negligable if your rocket's cross-sectional mass is much greater than the cross-sectional mass of the atmosphere it'll be plowing through (15 tonnes per square metre), or if it does most of its acceleration outside most of the atmosphere.
Correction: This is 10 tonnes per square metre if you're going straight up (about 15 pounds per square inch).
1. it does NOT have to be vertical(watch what the shuttle does soon after launch), so the acceleration can be done over longer distance, so less G's. think rail's in hundreds of miles. Expensive, yes, but less than 100 million per launch of shuttle.
I've done the calculations. Have you?
At 0.5 km/sec, and a maximum radial acceleration of (say) 10 gravities, your minimum turning radius is 2.5km - bigger than the 1km gun!
If you're building a horizontal gun and making the end turn up, turning radius gets _worse_, because of the higher muzzle velocity. It goes up as the _square_ of the velocity! You need a tower high enough that you might as well make the whole gun a tower.
Mount Everest is 4.4 km high. If you carve a giant channel in it, so that your gun gracefully curves, you get a maximum muzzle velocity of around 0.66 km/sec. Still very, very low.
If you just run a straight gun up the side of a mountain the size of Mt. Everest, you get a straight gun around 6 km long. At 10 gravities maximum acceleration (as per previous post), this gives you 6e5 J, or a velocity of 0.77 km/sec.
Still not enough to make a worthwhile difference.
Bear in mind also that tilting the gun at an angle, like you would going up the side of a mountain, gives you much more atmosphere to go through on the way up. If you try to turn the craft in the atmosphere, you're still forced to turn slowly, and your acceleration limit will be much lower than for a turning gun barrel, making the turning radius much larger (turning radius is inversely proportional to radial force).
2. even if it did, you could just build it up the side of a tall mountain, and have it curve gently up, which is kinda the most likly solution anyway, as it put you higher in the air, so less air resistance, closer to orbit, that type of thing.
Air resistance effects are negligable if your rocket's cross-sectional mass is much greater than the cross-sectional mass of the atmosphere it'll be plowing through (15 tonnes per square metre), or if it does most of its acceleration outside most of the atmosphere.
For a conventional heavy-payload rocket, both of these conditions are true, and atmosphere resistance doesn't matter.
You don't build a magrail to give your spacecraft orbital velocity. Of course that's silly, for the reasons given above. You use it to give you some small PART of your velocity. This is extremely beneficial.
This turns out not to be the case.
First, calculate how fast a magnetic launcher can fire a craft.
Remember, the launcher needs to be pointed upwards. You can't just turn the craft at the end - G force limits would require a very large turning radius for this.
Assume a vertical launcher length of 1 km maximum.
Assume a maximum acceleration for delicate cargo (like people or delicate equipment) of 10 gravities (100 N). I'm ignoring gravity's contribution; accelerating upwards at 10 gravities, the cargo would feel 11 gravities of force.
This gives an energy transfer over the length of the gun of 1000m * 100 N = 1e5 J, corresponding to a velocity change of about 0.45 km/sec.
To get to low orbit - not geosynch or escape - you need a delta-V of about 8 km/sec. If you're burning liquid hydrogen, with a specific impulse in the 4000 N*s/kg range, you'll need a rocket that's ( 1 - exp(-8000/4000) ) = 86% fuel.
If you get 0.45 km/sec for free, you need a rocket that's ( 1 - exp(-7750/4000) ) = 85% fuel.
Magnetic launching gained you 1% of the rocket's mass for cargo. Not much.
If you can launch from an airless body like the moon, then you can build much longer launchers tangentially to the surface, which would be extremely useful for lifting payloads. However, I've yet to see any proposal for an earth-based launching scheme that would give a substantial benefit without an astronomical cost.
Building a magnetic accelerator several tens of kilometres long might work, but that would be insanely expensive, requiring huge traffic volumes to pay itself off. Building a laser-based launcher [basically a jet with a ground-based laser as the heat source] looks attractive at first due to long path length, but has strong limits on energy density (you don't want to ionize the atmosphere the beam travels through, or you'll get a reflective plasma scattering your light). A space elevator would be even more expensive than a magnetic launcher, would require advanced materials that we presently don't have, and could cause devastating amounts of damage if sabotaged.
In short, I'm doubtful of anything better than chemical rockets for launch of delicate cargo from Earth showing up any time soon. Space, of course, has considerably more interesting possibilities.
If you can build a thousand-gravity accelerator, then you might be able to send up sturdy cargo. However, that too would require very high volumes to be economically practical.
But the question that has to be asked is why spin a disc through a permanent magnet? Why not use a series of electro-magnets to produce a moving magnetic field?
Because a motor and one permanent manget are a lot simpler and cheaper than a stack of electromagnets.
Having one hot location (by the magnet) and one cold location (far from the magnet) also makes the heat exchanger rig simpler. With electromagnets, any fixed position near the magnets would alternately be hot and cold (not easy to pump heat with).
In short, it's because there's lots of extra complexity and no real benefit from using electromagnets for the fridge.
Problems of the type you descripe wouldn't benefit much from OOP. I'd solve them myself in C, but that's mainly just because I don't *know* Fortran (fortran is very good at math problems with straightforward methods of solution, and parallelizes easily).
My most recent project, on the other hand, involved simulating a computer system (I won't go into detail; wait until after I've published). This is a perfect scenario for OO - a computer system is built up out of many parts, many with similarities, but with different implementations. If you need a cycle-by-cycle or event-by-event simulation of a computer system, it makes sense to build up models of all of the parts that will be interacting and put them into an OO hierarchy. This lets you avoid copying code, and lets you change things much more easily than in a straight C environment.
In a previous project, I was building the skeleton for a program that would have to be rewritten for many different purposes. Some parts would stay the same between instances, and these were what I was building. I made the mistake of writing it in C, with the idea of upgrading it to C++ later. I ended up faking OO-isms by calling function pointers hither and yon to abstract things. When maintenance of the project was finally turned over to someone else, they started the migration to C++.
In a complex program where you have a lot of abstraction going on between different types of component, an explicitly OO language will almost always make life much easier.
Would it be possible to use ordinary parabolic satellite TV antennas for radio telescopy? Could they be combined to create a huge radio telescope?
If that's possible, maybe when people are not using their satellite antennas for TV they could be combined to create a world-wide radio telescope.
You could in principle do this, but in practice there are problems.
The main problem is that the electronics in the detector used with the dish are completely unsuited to radio astronomy. To use a radio telescope as a part of the array, you need a high-fidelity sample of the radio signal being received, timestamped to atomic-clock accuracy. A satellite TV pickup doesn't have a sub-nanosecond-accurate clock, and won't give you a digitization of the raw signal. Instead, it looks for strong signals in specific, narrow bands and blindly decodes them through combined analog and digital means (i.e. it treats everything it hears as a TV signal).
A secondary problem is that your satellite dish is pointed directly at a strong source of radio noise in the frequencies it's tuned to detect (the satellite).
The idea is a great one, but because you'd need to completely replace the electronics rig with something far more expensive, a better approach might be to sell radio telescope array "kits" built from stock parts and forget about using peoples' TV dishes.
This would probably be quite practical from an engineering standpoint, as most of the parts (including timestamping radio sampling boxes) can be bought off-the-shelf. I have no idea if enough amateur astronomers would buy these for them to be marketable (they wouldn't be cheap - tens of thousands of dollars per kit).
Could a space-based array be designed so that once it reaches its target location, it spreads itself out, gradually increasing the distance between its elements in a coherent manner, thereby increasing the effective size of the array over time? I would assume that a space-borne array would already be designed with plenty of fuel/rocketry for compensating for massive objects passing nearby and tugging on its corners... The same principle might be handy for adjusting/balacing the spacing between elements if an asteroid hits the jackpot, or a failure is detected.
:).
Yes, but you'd be trading off angular resolution against aliasing artifacts (the less of your aperture is filled, the worse aliasing artifacts will be, even when you assume constant sources and integrate over time). IMO, you'd be better off just adding more satellites
By coincidence, I recently did the calculations for the size of a metre-band radio telescope array needed to resolve features 100 km in size at a distance of 10 light-years (enough to resolve the aurorae of earth-sized planets, show thunderstorms on gas giants, and so forth). You'd need thousands of radio telescopes in solar orbits out to a radius of about 4 AU, but you could do it. Put them in eccentric polar orbits (i.e. away from most of the junk in the ecliptic), add excellent GPS-style beacons in precisely known orbits (constantly observed from Earth) to let the satellites track themselves, and you could get a very nice radio telescope for a surprisingly modest price (cheap satellites, well-known technologies and electronics, and the benefits of mass production, since you'll be making a thousand or more of them).
Such a telescope would be able to see aurorae and civilization-induced radio junk from Earth-sized planets out to around 10 light-years, map the magnetospheres of Earth-like planets and see detailed magnetic features in gas giants out to about 100 light-years, and get very detailed pictures of the outer envelopes of stars out to about 1000 light-years. It would be a very useful project.
Does the energy that does the work to remove the heat from the refridgerated side of the setup come from the work being done to insert the Gadolinium alloy into the magnetic field(and to remove it) or is the energy input done simply by the circulation of coolant fluid through the heated/cooled alloy.
If the alloy changes temperature based on its magnetic field environment, then the work that goes into the heating and cooling is from the force needed to move the alloy through the magnetic field in the rig they're using.
not every packet needs to be "routed" through a router. some are switched or broadcast over a shared segment, no gateway necessary.
Two possibilities:
* You didn't know this
* You were dishonestly ignoring the possibility
Or 3), I discussed this in my original post. RTFM.
This issue is closed, as the original poster made it clear in a reply that he did indeed intend to use switches internally.
I agree, sorry if my comment was mis-understood. The idea is to use a gigabit switch for apt-to-apt connections and a linux router for outbound traffic going to the ISP. I would not dream of using a PC as a gigabit switch, just think how many PCI slots you'd need for an apt complex!!
Also, I really don't think the IP's will be a problem, it really just depends on the ISP. For example, my ISP sells a high speed DSL package with 30 IPs. The apartment complex I live in has only 20 apts.
Neat. Sounds like the project is well in-hand, then.
this will work fine since hes not routing internal traffic. the only thing being routed/firewalled is internal to external and ext to int traffic.
He was talking about routing streaming video within the building. And having gigabit connections to *each* apartment. This suggests he's planning to set up a system with more than 100 Mbit/apartment load internally (I agree that external load would likely be low enough to route with a PC).
He's planning on running gigabit ethernet to every apt for apt-to-apt networking and use highspeed ADSL with several static IPs (one for each apt) for outbound internet access. [...] Of course he'll be using Linux and FreeBSD for just about everything from the router to the "apt game servers" and video on demand servers.
It sounds like either your friend doesn't have a good handle on the technologies involved with this, or there was some miscommunication between the two of you.
It sounded good up until "use Linux and FreeBSD for the router".
You need something better than a PC to route many apartments' worth of gigabit ethernet to each other. A PC doesn't have the internal bandwidth for more than one gigabit connection. If you're using an off-the-shelf gigabit ethernet hub or router, it'll be running its own embedded OS from the vendor (if it's complex enough to run anything at all). If you're using a souped-up non-PC workstation as the router... you're spending far more than you have to for a simple router.
In a similar vein, you'll have an interesting time getting enough static IPs for a medium-sized apartment building without a fight. Maybe when IP6 finally takes over.
This sounds like a really cool project, and your friend deserves praise for trying to pull it off, but he'd better take a close look at the tools he's planning to use for it, and make sure that he's using the right tools for the right parts of it.
A few people have been speculating about 3D processors based on this or similar technology. While this is a neat way of building memory, don't get your hopes up for 3D processors any time soon.
What they've done, according to the article, is deposit several layers of thin-film transistors on top of a more or less standard chip.
These transistors will be *slow*. That's fine for something you're using to replace flash, but not fine for a processor. The hard problem of building high-quality transistors in a multi-level structure has not been solved.
The other problem is heat. With a hierarchically-designed memory array, you can make a larger array without power per access going up very much (at the cost of a very small amount of extra delay). This means that packing ten times as much memory into the same chip area won't cause much of a heating problem.
The core of a microprocessor, on the other hand, is pretty much all active at once (or mostly active). You have calculation results flying to reservation stations everywhere, you have a lot of fully-associative arrays being indexed, and you have a lot of logic churning away. Packing this into a tenth the area would make the used area much, much hotter (remember Newton's law of heating and cooling - you need the same heat flow from a tenth the area, so ten times the heat difference between chip and environment).
The good news is that you might be able to put the L2 cache in higher layers with technology similar to this and save space that way, but this is a one-time saving, with a performance penalty (until the holy grail of stacked transistor technology is found).
Still an interesting accomplishment, of course.
The problem is hubs. I have yet to see a good gigabit hub for under $2k or so. Most of the gigabit-compatible hubs offered use gigabit for uplink, and a handful of 100-base-T links for the rest of the ports.
What sort of crack are you on? Check out the Linux [linuxrouter.org] Router Project [sourceforge.net], freesco, etc. I own an ISP [stormforge.net] that is Cisco free using entirely LRP based routers and firewalls.
And your internal routers are either not routing (saturated) gigabit traffic through multiple cards, or not running on commodity hardware.
Even a hub - not a router, which is more expensive - that can handle more than one gigabit connection at full data rate costs about $2000 US. Find me a better price, with a link. My own numbers are from dlink (http://www.dlink.com). The links you posted describe _software_. The problem is expensive _hardware_. The machines you describe do not have the internal data transfer bandwidth to handle multiple saturated gigabit interfaces, so I'm puzzled as to why you even mention them as examples.
I think for a router you will find that in practice it will be pretty difficult to saturate the 32-bit 33MHz PCI bus.
I am a-priori assuming an application that will saturate gigabit ethernet channels. Otherwise there's no reason to use gigabit at all, as you point out.
More than one saturated gigabit ethernet interface would certainly swamp a 32-bit 33 MHz PCI bus.
Joe user's LAN and cable modem won't do this. A fileserver with a single IDE drive won't do this.
Something like, say, a communications-limited distributed computing project sure would, and most interesting computing problems are communications-limited when scaled up past one box.
Any application that requires streaming uncompressed (lossless) video streams between machines at better-than-NTSC resolution would also require better bandwidth than 100-base-T provides. Why someone would use multiple machines for something like this is left as an exercise for the reader.
Either way, both I and the post I was replying to were assuming the existence of some need for gigabit ethernet.
Hi definition uncompressed video is more than 100Mb/sec - but 1Gb/sec over copper is on the horizon.
Gigabit ethernet, at least, is fairly cheap right now. Check dlink's site for a fair idea of what pricing is like.
Gigabit will run over standard cat-5 cable (it actually just runs slower signals in parallel over multiple pairs), so you won't even have to rewire for it.
The problem is hubs. I have yet to see a good gigabit hub for under $2k or so. Most of the gigabit-compatible hubs offered use gigabit for uplink, and a handful of 100-base-T links for the rest of the ports.
Has this changed in the past 6 months or so?
Using a PC as a router in place of a hub isn't an option, as one gigabit ethernet card will come very close to saturating a 32-bit 33 MHz PCI bus. Start streaming large amounts of data through the house and the router will fail to handle the traffic.
I expect that eventually we'll see a chemically-pumped laser rifle or pistol, about the same size as a normal rifle or pistol, with an optical cavity where the barrel would be, powered by cartridges of solid fuel that are fed by a mechanism similar to the one that feeds cartridges consisting of case/primer/powder/bullet.
While this would probably work, I'm having trouble seeing why someone would use this instead of a gun.
The big problem with lasers is that you can do a lot more damage to a target by hitting it with a fast-moving slug than by heating it. Lasers are useful when you have a target that's far away and moving erratically enough that you can't fire slugs at it, but personal weapons are usually used at much shorter range. This suggests a chemically-pumped kinetic slug weapon might better fit infantry's needs... which is what they currently have.
As for shooting down missiles, one of the more interesting ideas in the Star Wars project was guided bullets. If you can guide a bullet - or even a countermissile, as is currently being tried - accurately enough to shoot down an incoming rocket, this method will likely be cheaper than using a laser to destroy the same rocket.
There might still be scenarios where you'd rather use a laser than a slug-based weapon. I'd certainly be interested in hearing suggestions (I certainly don't claim to have thought of everything; it's just that lasers aren't the best solution for the scenarios that I _have_ thought of).
Uh, when I set up X to refresh my Viewsonic P225F [viewsonic.com] at 20hz it still flickers...
I read the original post as meaning LCD (flat-panel) monitors.
Serves me right for reading at 2am. My bad.
and absolutely useless gimmicks like cube-shaped computers and flat-screen monitors
Flat-screen monitors don't have flicker.
This is one *hell* of a useful improvement, especially for those of us who *have* to sit in front of a monitor all day.
Saving the desk real-estate of a CRT would be nice too.
*Sigh*.
As for the Dreamcast, don't forget the additional cost of the modem (if you like crawling) or the ethernet adaptor (if you can find one).
The Dreamcast ships with a modem. You don't need to buy one.
Turn off images in your browser, and bandwidth is not a concern (this was an email machine, remember?). I do this regularly when the university's internet pipe gets swamped.
Do you have any idea what it takes to get the surfadces flat enough? How long it takes to design a coating, and what sort of processes it takes to apply it?
Sure. And it's about an order of magnitude simpler than the processes involved in building the integrated circuit chips used in RAM. If you can find a magical method of producing RAM as simply and cheaply (per unit storage capacity) as hard drive platters, go and patent it and make a bundle.
Besides, people don't want insanely powerful computers. The computer manufacturers want people to buy them, so they make sure there's no decent alternative.
And I'm sure the conspiracy's mind-control satellites force you to upgrade yearly, too.
I got by with a 286 well into the days of Pentium machines. When I upgrade, I buy middle-end systems. If I want a system to act as a router or fileserver or other low-load system, I use a half-decade-old hand-me-down.
If I wanted a low-heat x86 box (at the expense of computing power), I'd buy a socket-7 motherboard and put a Cyrix chip in it (or a chip from whoever bought them; I don't remember if they're still around under the Cyrix name).
If I wanted a lower-power system that was still reasonably powerful, I'd follow another poster's suggestion and by an iMac.
If you can't find alternatives, you're not looking hard enough.
The point is that we desperately need processors that produce less heat and use less energy. If you take a moment to think about it, it's totally ridiculous that we need so many noisy fans inside a computer that someone's using to compose an email.
If you're using a high-end computer solely to compose email, I'd argue that the problem isn't the hardware.
Heck, if power is a concern, buy a Dreamcast and use the web client to access Hotmail. $50, and you get a low-power embedded box that you can read and write email and even play games on.
Desktop systems are overpowered because people want to be able to run insanely high-powered applications on them, no matter how much of a waste this is when they're not playing Quake XIV.
It's even more ridiculous when you consider that some graphics processors require a fan as well, and so does the power supply.
Same thing. A real-time realistically rendered 3D environment requires one hell of a lot of computing power to generate. This means heat. If you're just answering email, buy a PCI Rage XL card and save on the fan and heatsink.
Now if only they'd come up with a breakthrough that will make fast, long lasting, solid-state hard drives a reality.
They're called "flash cards".
If you want to store gigabytes of images or gigabytes of game install files, however, they won't be sufficient.
RAM is harder to make per unit storage space than a magnetic platter. This is just the nature of the universe - RAM is intrinsically more complex. A magnetic platter is just a flat surface with the right kind of coating; it doesn't get much simpler than that. You can buy a solid-state drive off the shelf right now, but the the cost will reflect the fact that it's harder to build, and this will continue to be the case for quite a while.
In summary, the problem isn't the technology, it's the fact that people *want* insanely powerful computers, with large amounts of storage, for the lowest price that still gives them the power and space they crave.