Who said anything about a high-quality vacuum? Yes, a high quality vacuum would be hard to maintain, probably impossible on such a large scale, but does it really have to be high-quality for this plan to work?
Yes. The efficiency of the system increases with the degree of vacuum inside the tubes, but it's not a linear curve. So there's basically no benefit worth talking about until you get down to lab-quality vacuum.
Lots of potential energy in a lab-quality vacuum... actually, in the interface between the vacuum and the outside atmosphere. Ever seen a bell jar fail? I hear they're go off like hand grenades.
I regret that I don't have those figures-- can't remember where I read the damned things, and google was no help this time-- but consider that moving a foot over a year is just an inch per month. I've seen the foundations on houses settle faster than that in certain circumstances. (That is, of course, disastrous for the owners of the house, but that's another story.)
ANYWAY, the upshot of all this is that if you can accelerate something to 15.972 KM/s or (57,499.2 KM/h or (x0.62) 35,649 miles per hour, it will coast its way along without needing anything under it, and without consuming further gas. This could be a really great way to deliver packages.
Assuming your packages could handle high accelerations. It would take nearly half an hour at 1 g to accelerate a body to that speed, by which time your package would already be 13,000 kilometers away.
Except they don't. Read the hundred or so comments criticizing this idea on the grounds of practicality (that much vacuum is effectively impossible) and safety (that much vacuum is effectively a giant bomb).
As for personal rapid whatever you said, it suffers from exactly the same problem as all other rail-based transportation: there will always be many more destinations than there are stations. For the majority of the population, such a system would be an inconvenience at best.
the tube, from a geometric point of view, is dead straight, but from a gravitic potential point of view, it's a slope down for half the way and a slope up for the second half
But the problem with that is that you either have to dig a tunnel through the gravitational center of the earth, or you still have to have a wheel or maglev system. You got it right when you said that it's a slope down and back up, and that means the vehicle-- capsule, car, pod, whatever-- is going to be sliding all the way. If you use wheels, you have to deal with rolling friction, which is bad because the system depends on there not being any; if you lose any energy at all, the vehicle won't make it to its destination. It'll stop short, and then fall back, eventually settling at the midpoint of the tunnel.
And if you try to use maglev as a frictionless system, the cost of digging the tunnel suddenly skyrockets beyond belief. It almost reaches the point where it'd be cheaper to use rockets and ballistics to get from point A to point B, particularly for gotta-be-there-in-15-minutes cargo.
Which brings up an interesting point. Using a ballistic path, you can get from any point on Earth to any other in less than 45 minutes. (If I'm remembering my high school physics right.) I wonder why nobody's tried to combine ballistics with smart weapons technology (GPS and onboard guidance, et cetera) to come up with a way to get from here to there, literally, as quickly as physically possible. If we built 'em in mass, Redstone-type rockets wouldn't be terribly expensive....
Actually, from reading the FAQ, it seems like the company is merely promising franchise rights to this, not any actual end-product itself. That's worse than vaporware. That's meta-vaporware. Yuck.
until about 45 mile/hr drag doesn't do much of anything.
You've obviously never ridden a racing bicycle. At the leisurely pace of 20 mph or so, you can absolutely feel the difference between being at the head of the line and being in a drafting position behind someone else. And that's being a relatively slender and slippery person on an expensive space-aged bike. I imagine the drag caused by the wall-like front of a locomotive at the same speed is considerable.
What about the 4000 mph capsule that goes trans-oceanic
Sounds kinda improbable to me. In order to get from standing start to 4,000 miles per hour, you have to accelerate at one gee for a little over three minutes. (If my math is right. t = v/a, v = 5866.66 feet per second, a = 32 feet per second per second, v/a = 183.33 seconds.) That's a long time to feel like you're lying flat on your back. And slowing down in the real killer. Accelerating at one gee for three minutes in the opposite direction? I hope the seats come equipped with four-point restraints instead of just lap belts, otherwise there are going to be a lot of bloody noses.
Dying because you aren't getting enough oxygen in a low pressure environment and dying because you aren't getting enough oxygen in a very low pressure environment don't seem all the different to me.
Of course, the difference is that a plane can dive from 40,000 feet to 10,000 feet in a minute or so, which is quickly enough to save everybody's lives. If you're in an evacuated, subterranean tunnel and your car springs a leak... hmm. According to the FAQ, "Life-support apparatus is a well developed field." So I guess it won't be a problem.
Better idea. Try sticking your hand out of a car doing 40 mph. Note the resistance, then withdraw your hand. Now try sticking your hand out of a plane doing 400 mph. Note the resistance, then withdraw your bloody stump. Can you tell the difference?
Re:From now on, we'll all travel in TUBES!
on
Pipeline Mass Transit?
·
· Score: 1, Troll
We can barely dig a little tunnel under the English Channel, and we're seriously proposing vacuum tubes?
In all fairness, digging the English Channel tunnel wasn't really a "barely" thing. Once the right people got their heads together on it, it was a logistical effort on a scale not often seen, but technically it was kind of a breeze.
Of course, it's easy for me to say it. I didn't have to dig any of it myself.
Not because it's not interesting and efective technology, but because we Americans don't like mass transit.
Pfeh. If the population density of a place gets high enough that people start talking about mass transit, move. Cities are generally not a healthy environment for humans. Some people may prefer to live in them, but you certainly don't have to.
Flee the cities. Flee the suburbs. Move to west Texas, or Montana, or Nebraska... or Australia. Whatever floats your boat. Get yourself some land and live in a house where you can't hear or smell your neighbors.
No. Say you have a train tube that's a reasonable length-- LA to San Francisco, Dallas to Houston, New York to Washington. You have to maintain a high-quality vacuum over that entire length. It's really late, so I'm not going to do the math for fear of getting it wrong and ruining my point, but the volume of such a tube would be really, really large. The surface area would also be really, really large. The likelihood that you could maintain a vacuum in such a tube is essentially zero. This is particularly true in an environment like the central California valley, where two points of land on either side of a fault line can shift as much as a foot in either direction over the course of a year or so, and that's without an earthquake.
More dangerous than flying?
Definitely. If a plane crashes, it's obviously horrible for the passengers, but the danger to bystanders is minimal. A plane crash-- one caused by failure or error, not deliberate malice-- might kill a few people on the ground, and that would be terrible. But a catastrophic failure of an evacuated tube would have the force of a medium-sized bomb, and it would be spread out all through the city, the countryside, et cetera. Thousands could be killed in a catastrophic evacuated tube failure, unless the tubes were all buried deep underground. As has already been discussed elsewhere, that idea has survivability problems of its own.
And cars are still much more dangerous.
That's a common misconception caused by the careless application of statistics. The total number of automobile fatalities per year is umpty-thousand. That sounds like a big number, even when you compare it to the total population. But when you look at the numbers another way, calculating an individual person's likelihood of being involved in a fatal automobile accident in his or her lifetime, the percentages come out very close to zero. That's why automobile liability insurance is still available, and affordable. Automotive transport is actually quite safe from an actuarial point of view.
Lucifer's Hammer would be cool too, but it's kinda dated now, and that genre's been done to death
That's true, but I think the world needs a comet-from-space, end-of-the-world movie where the comet hits 30 minutes into the movie. That hasn't been done before, unless you could The Day After, which was too depressing to be entertaining.
I can read the part of Lucifer's Hammer about the impact and the immediate aftermath again and again.
I would disagree with that. I'm maxing out my gig of RAM right now. It's not because I'm being inefficient, it's because I'm making good use of it to speed up workflow. The more I have, the more stuff I can do.
You need to understand the difference between total memory and per-process virtual memory. If your computer has two gigs of RAM in it, and you launch four programs that allocate 512 MB each, you've exhausted all of your physical memory. But each of those programs could allocate another 512 MB of memory if it needed it; your computer would just start swapping. That's total physical memory.
What we're talking about is the maximum amount of RAM available to a single process. It's very, very, very unusual for a single process to need to allocate more than 2 GB of RAM for itself. I guarantee that none of the programs on your computer need anywhere even close to that amount of RAM. If you open them all at once, you can use more than 2 GB of physical memory in your computer, but not because any one process needed more than 2 GB of virtual memory.
The 32-bit/64-bit question only applies when you're talking about the amount of virtual memory available to a single process. It has nothing to do with how much memory you can put in your computer, or how many programs you can run at one time.
It's not. Even at film resolution, 3D rendering doesn't require 2 GB of RAM. It works just fine with 32-bit addressing. The killer solution for 3D rendering is massive coarse parallelization; each machine in a big farm-- not a cluster, just a farm-- gets a frame to crunch on. Since each frame is unrelated to the one before or after it, each machine in the farm can work independently of the others. So in the time it takes to render one frame, you can render N, where N is the number of identical machines you have in your farm.
Apple like to infer the Tibooks are 64bit machines.
First things first. If they lead you to think something without saying it, they imply. If you think something without their saying it, you infer.
Now, where exactly did Apple imply that their PowerBooks are 64-bit systems? The only marketing or tech material I've seen that even mentions 64-bit computing is a little blurb on the vector units, saying that they handle "information in 128-bit chunks, compared to the 32- or 64-bit chunks in traditional chips."
I've never done any real programming on a Sun, but on an SGI you have to specifically tell the compiler to compile to a 64-bit ABI if you want to use 64-bit pointers. The syntax is "cc -64 -whatever foo.c." If you omit the -64 flag, you get a 32-bit binary. The same may be true on a Sun.
But is 32-bit always worse in practice than 64-bit?
Actually, 64-bit is usually worse in practice than 32-bit, all other things being equal. Many processors let you compile code for 32-bit pointers or 64-bit pointers; the MIPS R10000 family is the one I'm familiar with. The same code compiled for the 32-bit ABI will either run at the same speed as the 64-bit version, or it will be faster. The difference is caused by cache performance. If your pointer is twice as big, you can only squeeze half of 'em into the same caches. Thus, more cache misses, and decreased performance of the application overall.
Unless you need more than 2 GB of virtual memory for your program, you should compile it with 32-bit pointers.
wouldn't it be in their best interest to make it very easy to integrate their tool with any other editor?
Look at, let's say, the top 10 IDEs available. How many of them include support for full integration with external editors? None? Then you have your answer. No, it is not in their best interest. If it were, they'd be doing it.
Surely it wouldn't create a problem, license-wise, to have their tool integrate with any binary version of vim or emacs, right?
Can't be done. If you're using a binary version of Editor X, then you're not integrated with it. At best, you're using system() calls to fire off an instance of the editor for each source file in the project, which is not the same thing at all.
Then, their product competes on it's merits of integrating the development process
Except, as I pointed out, that you're not actually integrating anything at that point.
Try out a good IDE for a change, like Visual Studio or (much better) Project Builder. It'll change your opinions.
I don't think that word means what you think it means:
Draconian \Dra*co"ni*an\, a. Pertaining to Draco, a famous lawgiver of Athens, 621 b. c.
Draconian code, or Draconian laws, a code of laws made by
Draco. Their measures were so severe that they were said
to be written in letters of blood; hence, any laws of
excessive rigor.
Slashdot Mirror
That's by far the worst example of statistical interpretation I've seen in a long time. Kudos to you for raising that bar ever higher.
Who said anything about a high-quality vacuum? Yes, a high quality vacuum would be hard to maintain, probably impossible on such a large scale, but does it really have to be high-quality for this plan to work?
Yes. The efficiency of the system increases with the degree of vacuum inside the tubes, but it's not a linear curve. So there's basically no benefit worth talking about until you get down to lab-quality vacuum.
Lots of potential energy in a lab-quality vacuum... actually, in the interface between the vacuum and the outside atmosphere. Ever seen a bell jar fail? I hear they're go off like hand grenades.
I regret that I don't have those figures-- can't remember where I read the damned things, and google was no help this time-- but consider that moving a foot over a year is just an inch per month. I've seen the foundations on houses settle faster than that in certain circumstances. (That is, of course, disastrous for the owners of the house, but that's another story.)
ANYWAY, the upshot of all this is that if you can accelerate something to 15.972 KM/s or (57,499.2 KM/h or (x0.62) 35,649 miles per hour, it will coast its way along without needing anything under it, and without consuming further gas. This could be a really great way to deliver packages.
Assuming your packages could handle high accelerations. It would take nearly half an hour at 1 g to accelerate a body to that speed, by which time your package would already be 13,000 kilometers away.
They look like they may be a good idea
Except they don't. Read the hundred or so comments criticizing this idea on the grounds of practicality (that much vacuum is effectively impossible) and safety (that much vacuum is effectively a giant bomb).
As for personal rapid whatever you said, it suffers from exactly the same problem as all other rail-based transportation: there will always be many more destinations than there are stations. For the majority of the population, such a system would be an inconvenience at best.
For crying out loud, dude. Not every Slashdot article is an opportunity for you to bash Microsoft, okay? Cut it out.
the tube, from a geometric point of view, is dead straight, but from a gravitic potential point of view, it's a slope down for half the way and a slope up for the second half
But the problem with that is that you either have to dig a tunnel through the gravitational center of the earth, or you still have to have a wheel or maglev system. You got it right when you said that it's a slope down and back up, and that means the vehicle-- capsule, car, pod, whatever-- is going to be sliding all the way. If you use wheels, you have to deal with rolling friction, which is bad because the system depends on there not being any; if you lose any energy at all, the vehicle won't make it to its destination. It'll stop short, and then fall back, eventually settling at the midpoint of the tunnel.
And if you try to use maglev as a frictionless system, the cost of digging the tunnel suddenly skyrockets beyond belief. It almost reaches the point where it'd be cheaper to use rockets and ballistics to get from point A to point B, particularly for gotta-be-there-in-15-minutes cargo.
Which brings up an interesting point. Using a ballistic path, you can get from any point on Earth to any other in less than 45 minutes. (If I'm remembering my high school physics right.) I wonder why nobody's tried to combine ballistics with smart weapons technology (GPS and onboard guidance, et cetera) to come up with a way to get from here to there, literally, as quickly as physically possible. If we built 'em in mass, Redstone-type rockets wouldn't be terribly expensive....
Except--the company actually is promising this.
Actually, from reading the FAQ, it seems like the company is merely promising franchise rights to this, not any actual end-product itself. That's worse than vaporware. That's meta-vaporware. Yuck.
until about 45 mile/hr drag doesn't do much of anything.
You've obviously never ridden a racing bicycle. At the leisurely pace of 20 mph or so, you can absolutely feel the difference between being at the head of the line and being in a drafting position behind someone else. And that's being a relatively slender and slippery person on an expensive space-aged bike. I imagine the drag caused by the wall-like front of a locomotive at the same speed is considerable.
What about the 4000 mph capsule that goes trans-oceanic
Sounds kinda improbable to me. In order to get from standing start to 4,000 miles per hour, you have to accelerate at one gee for a little over three minutes. (If my math is right. t = v/a, v = 5866.66 feet per second, a = 32 feet per second per second, v/a = 183.33 seconds.) That's a long time to feel like you're lying flat on your back. And slowing down in the real killer. Accelerating at one gee for three minutes in the opposite direction? I hope the seats come equipped with four-point restraints instead of just lap belts, otherwise there are going to be a lot of bloody noses.
Dying because you aren't getting enough oxygen in a low pressure environment and dying because you aren't getting enough oxygen in a very low pressure environment don't seem all the different to me.
Of course, the difference is that a plane can dive from 40,000 feet to 10,000 feet in a minute or so, which is quickly enough to save everybody's lives. If you're in an evacuated, subterranean tunnel and your car springs a leak... hmm. According to the FAQ, "Life-support apparatus is a well developed field." So I guess it won't be a problem.
Better idea. Try sticking your hand out of a car doing 40 mph. Note the resistance, then withdraw your hand. Now try sticking your hand out of a plane doing 400 mph. Note the resistance, then withdraw your bloody stump. Can you tell the difference?
We can barely dig a little tunnel under the English Channel, and we're seriously proposing vacuum tubes?
In all fairness, digging the English Channel tunnel wasn't really a "barely" thing. Once the right people got their heads together on it, it was a logistical effort on a scale not often seen, but technically it was kind of a breeze.
Of course, it's easy for me to say it. I didn't have to dig any of it myself.
Not because it's not interesting and efective technology, but because we Americans don't like mass transit.
Pfeh. If the population density of a place gets high enough that people start talking about mass transit, move. Cities are generally not a healthy environment for humans. Some people may prefer to live in them, but you certainly don't have to.
Flee the cities. Flee the suburbs. Move to west Texas, or Montana, or Nebraska... or Australia. Whatever floats your boat. Get yourself some land and live in a house where you can't hear or smell your neighbors.
Don't fence me in, baby.
Less than it is to keep a plane in the air?
No. Say you have a train tube that's a reasonable length-- LA to San Francisco, Dallas to Houston, New York to Washington. You have to maintain a high-quality vacuum over that entire length. It's really late, so I'm not going to do the math for fear of getting it wrong and ruining my point, but the volume of such a tube would be really, really large. The surface area would also be really, really large. The likelihood that you could maintain a vacuum in such a tube is essentially zero. This is particularly true in an environment like the central California valley, where two points of land on either side of a fault line can shift as much as a foot in either direction over the course of a year or so, and that's without an earthquake.
More dangerous than flying?
Definitely. If a plane crashes, it's obviously horrible for the passengers, but the danger to bystanders is minimal. A plane crash-- one caused by failure or error, not deliberate malice-- might kill a few people on the ground, and that would be terrible. But a catastrophic failure of an evacuated tube would have the force of a medium-sized bomb, and it would be spread out all through the city, the countryside, et cetera. Thousands could be killed in a catastrophic evacuated tube failure, unless the tubes were all buried deep underground. As has already been discussed elsewhere, that idea has survivability problems of its own.
And cars are still much more dangerous.
That's a common misconception caused by the careless application of statistics. The total number of automobile fatalities per year is umpty-thousand. That sounds like a big number, even when you compare it to the total population. But when you look at the numbers another way, calculating an individual person's likelihood of being involved in a fatal automobile accident in his or her lifetime, the percentages come out very close to zero. That's why automobile liability insurance is still available, and affordable. Automotive transport is actually quite safe from an actuarial point of view.
Lucifer's Hammer would be cool too, but it's kinda dated now, and that genre's been done to death
That's true, but I think the world needs a comet-from-space, end-of-the-world movie where the comet hits 30 minutes into the movie. That hasn't been done before, unless you could The Day After, which was too depressing to be entertaining.
I can read the part of Lucifer's Hammer about the impact and the immediate aftermath again and again.
Dune was special. Dune tied the mythic dream-time world into the world of day. Dune dealt with the essence out of which religions are created.
Fweet! Illegal use of purple prose. Fifteen yard penalty, Vikings have the ball.
(Yeah, yeah. Old joke.)
I would disagree with that. I'm maxing out my gig of RAM right now. It's not because I'm being inefficient, it's because I'm making good use of it to speed up workflow. The more I have, the more stuff I can do.
You need to understand the difference between total memory and per-process virtual memory. If your computer has two gigs of RAM in it, and you launch four programs that allocate 512 MB each, you've exhausted all of your physical memory. But each of those programs could allocate another 512 MB of memory if it needed it; your computer would just start swapping. That's total physical memory.
What we're talking about is the maximum amount of RAM available to a single process. It's very, very, very unusual for a single process to need to allocate more than 2 GB of RAM for itself. I guarantee that none of the programs on your computer need anywhere even close to that amount of RAM. If you open them all at once, you can use more than 2 GB of physical memory in your computer, but not because any one process needed more than 2 GB of virtual memory.
The 32-bit/64-bit question only applies when you're talking about the amount of virtual memory available to a single process. It has nothing to do with how much memory you can put in your computer, or how many programs you can run at one time.
It's not. Even at film resolution, 3D rendering doesn't require 2 GB of RAM. It works just fine with 32-bit addressing. The killer solution for 3D rendering is massive coarse parallelization; each machine in a big farm-- not a cluster, just a farm-- gets a frame to crunch on. Since each frame is unrelated to the one before or after it, each machine in the farm can work independently of the others. So in the time it takes to render one frame, you can render N, where N is the number of identical machines you have in your farm.
Apple like to infer the Tibooks are 64bit machines.
First things first. If they lead you to think something without saying it, they imply. If you think something without their saying it, you infer.
Now, where exactly did Apple imply that their PowerBooks are 64-bit systems? The only marketing or tech material I've seen that even mentions 64-bit computing is a little blurb on the vector units, saying that they handle "information in 128-bit chunks, compared to the 32- or 64-bit chunks in traditional chips."
I've never done any real programming on a Sun, but on an SGI you have to specifically tell the compiler to compile to a 64-bit ABI if you want to use 64-bit pointers. The syntax is "cc -64 -whatever foo.c." If you omit the -64 flag, you get a 32-bit binary. The same may be true on a Sun.
But is 32-bit always worse in practice than 64-bit?
Actually, 64-bit is usually worse in practice than 32-bit, all other things being equal. Many processors let you compile code for 32-bit pointers or 64-bit pointers; the MIPS R10000 family is the one I'm familiar with. The same code compiled for the 32-bit ABI will either run at the same speed as the 64-bit version, or it will be faster. The difference is caused by cache performance. If your pointer is twice as big, you can only squeeze half of 'em into the same caches. Thus, more cache misses, and decreased performance of the application overall.
Unless you need more than 2 GB of virtual memory for your program, you should compile it with 32-bit pointers.
He is part of team that write Perl to download driver for IBM server.
Willy Wang: What meaning of this?
Lionel Twain: Is the. Is the. What IS THE meaning of this! Use your god-damned articles and prepositions!
(Sorry, no offense. Your post made me think of that great line from Murder By Death, and I just cracked myself up.)
wouldn't it be in their best interest to make it very easy to integrate their tool with any other editor?
Look at, let's say, the top 10 IDEs available. How many of them include support for full integration with external editors? None? Then you have your answer. No, it is not in their best interest. If it were, they'd be doing it.
Surely it wouldn't create a problem, license-wise, to have their tool integrate with any binary version of vim or emacs, right?
Can't be done. If you're using a binary version of Editor X, then you're not integrated with it. At best, you're using system() calls to fire off an instance of the editor for each source file in the project, which is not the same thing at all.
Then, their product competes on it's merits of integrating the development process
Except, as I pointed out, that you're not actually integrating anything at that point.
Try out a good IDE for a change, like Visual Studio or (much better) Project Builder. It'll change your opinions.
I don't think that word means what you think it means:
Draconian \Dra*co"ni*an\, a. Pertaining to Draco, a famous lawgiver of Athens, 621 b. c.
Draconian code, or Draconian laws, a code of laws made by
Draco. Their measures were so severe that they were said
to be written in letters of blood; hence, any laws of
excessive rigor.