The "common toy" is not a radiometer. It's a heat engine. The bulb is only partially evacuated and the hotter, black side of the vanes heats up the gas molecules, which then bounce off it with increased vigor, compared to the white side. So the vanes spin with the white side going forward.
A true radiometer would be bouncing photons off the white side, and spinning with the black side leading.
The heat-engine version has many times the efficiency of the photon one.
Er, if you actually try to go read TFA, it seems they analyze the text by semi-numerological means. Like noticing that one particular argument is about 1/12th the length of the chapter, from that somehow drawing some far-fetched conclusion. Sounds like a particularly bizarre form of BS to me.
Way back around 1972, I worked on a CDC time-share system. They charged 4 cents per CPU second, 1 cent per PRU (640 characters) transferred to/from disk, and 0.2 cents per kiloword-second of memory used.
Except after 5PM, when the rates went down 50%.
Luckily I worked for the computer center, so the long assembly times ( 5 minutes ) were charged against a funny-money account. Still it was humbling that one missing comma and I'd wasted about 20 minutes of real time and $12, when $12 was real money.
Nobody is going to invest the needed billions of dollars in a country with no real government, no laws, no protection for private property, and every expectation of being taken over by the Taliban as soon as the US army leaves.
It would take billions in up-front investment, as Afghanistan does not have any of the needed things: water, power, roads, engineers, chemical plants, railroads, ports, diging machines, huge trucks, smelters, coal, oil, and gas. Billions, and at least ten years to build the infrastructure before a pound of ore comes out of there.
And minerals only get extracted if the cost is less there than from the developed sources. That's unlikely, due to the needed up-front investment. And one of the alleged largest supplies, Lithium, is already being mined very, very cheaply in South America, where there are huge easily-accessed deposits.
* Mining-company geologists have been scouring the globe for centuries, looking for mineral deposits that are economically recoverable.
* Minerals do not know about arbitrary political boundaries, making it highly unlikely that this "treasure-trove", if it exists, is wholly contained in Afghanistan.
* Minerals are heavy and hard to extract, which makes it paramount that there be things that Afghanistan has none of, such as rail lines, roads, ports, docks, electricity, coal, fresh water, chemicals, a stable government, a stable economy, and much more. Lacking just one of those items can make mining an impractical venture.
* No bank is going to loan the hundreds of millions to billions needed to even begin to extract these minerals. Banks do not loan money into war zones with no history of a stable government or protection of private property, and when the only source of quasi-stability, the US military, is on a countdown to leave the country.
Funny, but I was just reading an old radio magazine, circa 1938, where they were using 5 to 6 volts to the heater of a rectifier tube that usually needed 25. That's 1/4 the voltage, about 1/16th the power, and the rectifier worked BETTER at detecting radio signals than at full voltage. Some complex thing about the diode work functions one might suspect.
Engineers have explored most corners of the performance envelope, nothing all that new under the sun.
Dang, and here I'd al;ways assumed "Zero Day" meant the bug had been there since the day the software was released. Like the bug in the.BMP rasterizer, revealed in 2004, that had been there since Windows 3.0
Who manages the canonical definition of "Zero Day" ?
>Comparing something unfavorably to NTSC is going to get you marked as a nutter.
You miss the point. The point is you only need 1/3 the color resolution, compared to monochrome. NTSC is just one example where this eyeball ratio is put into practice.
>>The iPhone, and every other LCD screen, has three color >>elements per pixel, while the eye has like 1/3. That's a NINE >>TIMES difference that this guy is glossing over.
>Not really. You can see those pixels, so they're relevant.
I can't see the individual color elements on a 72dpi display. Even if I could, the 9/1 ratio still holds.
(1) The human eye has very variable resolution. Down in the fovea it may be up at this guy's numbers, but much less everywhere else.
(2) The eye's color receptors are much farther apart, and therefore of poorer resolution, that the monochrome receptors. That's why the old NTSC standard had about 1/3 the color bandwidth than the Y bandwidth.
(3) The iPhone, and every other LCD screen, has three color elements per pixel, while the eye has like 1/3. That's a NINE TIMES difference that this guy is glossing over.
(4) It really doesn't matter. We don't spend our lives inspecting individual pixels-- we let our brain process the images into coherent high-level objects, such as "letters" and "faces".
So pathetic. This is not a way to generate electricity, but to destroy it.
You see those "p-n" junctions did not appear out of thin air-- they're the result of using scads of electricity to heat silicon to the melting point, extract it into perfect crystals, then slice it, anneal it again in an electric furnace, then more hours at 1200C to diffuse in minute amounts of p and n dopants, then more electricity to slice, dice, solder, and cement these into usable devices.
And in the end you have some very expensive, in both dollars and energy used, heat to microamps of electricity converters. And you can easily compute exactly how much electricity you get back given say a 10 degree temperature difference between the warm and slightly less warm sides. It's miniscule. Microwatts per square centimeter. Even if you wore these for 10,000 hours, you're nowhere near making back the amount of electricity, not to mention the $$$, it cost to make these things.
The problems with anti-missile defense are more basic than that:
(1) Basic geometry -- you have to station a slew of defensive missiles every 20 miles along your borders. That's because you are not going to hit anything going Mach 12 across your path-- you need a close to head-on intercept angle.
(2) Cheap and easy countermeasures. Even if you bankrupt your country setting up (1), the bad guys just switch to using sub or boat launched cruise missiles. Or low-trajectory ICBM's. Or put the bomb on a freight or passenger plane. It's mighty foolish to spend a trillion $ and have all that effort counteracted by a visit to UPS and $187.54.
JR Oppenheimer did this math in his head in 1952 as he was testifying to a govt comittee. Nothing has changed since then.
>But don't you think that they would use space-certified RAM chips for such a project?
They did, but cosmic rays come in a wide range of intensities, from feeble all the way up to having enough energy in one photon to make (baseball analogy ahead) a baseball jump.
Nice Try, MIT, but it's all paper airplane stuff if the design cannot be certified for commercial flight.
Aerodynamicists have been trying different shapes, configurations, and designs for about 120 years now.
There are low-drag designs, but most of them are impractical in some way-- like no space for landing gear, or customer's won't ride in them ( Flying Wings, scimitar props ), or the design won't pass FAA rules. Lots of very picky FAA rules that make lots of designs impossible to certify for flight. The rules are mostly there to assure safety and conservative margins. For instance, the plane has to be able to climb with all engines on one side out. That rules out lots of designs with wingtip-mounted engines and designs with insufficient power. It has to be able to land with a certain crosswind component. It has to have a certain maximum airspeed where one engine is out and the plane is still controllable. That rules out a lot of designs with small tails. It has to be able to sustain a catastrophic failure of one engine without it affecting other engines. That rules out a lot of designs with multiple engines close together.
In addition the plane has to be manufacturable and inspectable and serviceable. That tosses out designs with complex wings that can't be easily assembled in pieces or later on opened up for inspection and service.
Plus the plane has to have space for all the accessories-- not just landing gear, but fuel tanks, hydraulic tanks, air-conditioning packs, batteries, surge tanks, hydraulic pumps, slats, speed-brakes. That rules out a lot of designs with very thin wings.
One may hope this MIT design kept all these real-world constraints in mind.
While you might be able to make ten gazillion AND gates, you still have the minor problem of HOOKING THEM UP into some useful logical building blocks, like adders, buffers, and memory. And the bigger problem of amplifying the results to a level acceptable to the following inputs. And figuring out how to distribute power (ATP) to each amplifier.
And the signal levels are so low, thermal noise is going to induce a lot more errors than you'd like.
And the speed is not likely to be very exciting.
I would not start short-selling Intel stock based on this technology.
Somebody got the idea that you could use this stuff to steer light onto solar cells. Reasonable mistake.
You have to steer a solar cell to follow the sun so it's collecting the most light.
Steering the light once it's hit the panel is mostly useless-- you're too late -- you're just not intercepting the sun.
For example if the sun is 45 degrees to the side, you're only getting cos(45) or 70.7% of the rays. Nothing you do at the panel can change that.
And there are already special reflector shapes that have the amazing property of steering light from many angles to one destination. And they're just plain metal surfaces, no nanotech required.
The rate of evaporation from the oceans is about 400,000 cubic kilometers per year.
To increase that by just one percent would mean pumping 4,000 km^3 of water.
Just raising that much water to 3,000 feet would take approximately, oh let's see, carry the 0x100, about 1,651,445,966.51 horsepower. One Point Six BILLION horsepower.
Get the following volumes too:
on
Hacking Vim 7.2
·
· Score: 5, Funny
Get the followup volumes too:
Noise cancelling algorithm design using sh. ( Shhhhhh... )
Real-time traffic control with bash.
Time-domain-reflectometry made easy, with sed.
GPS satellite tracking with tr.
Build a species database with Python.... and many more...
>The point is that they are cheaper compared to the alternatives.
That's like saying you should stab rather than shoot yourself in the foot, as it's a whole lot cheaper.
It does not matter how cheap something is if it's still below the break-even point. In fact it's usually better to pay more as there are usually economies of size and scale.
I wish these folks well, but there's no indication from TFA that this concept is any better than the alternatives.
>It's at this point that you should realize that your understanding of Physics is a caricature of the real world.
Likewise, for you.
The total energy does not change when you narrow the river. You have X amount of water dropping by distance D. That's the energy. It's exactly the same amount of energy, whether it's flowing at 1 meter per second with a cross section of 100m^2, or flowing at 100 m/s with a cross section of 1m^2. It sure looks more impressive, but it's exactly the same amount of energy (actually, less, as the drag goes up as the cube of the speed).
These Swedes are speeding up the rate of traverse through the water, but that's just in order to have enough speed to match the intake speed needed for a tiny turbine. And it HAS to be a tiny turbine as the weight of a turbine goes up as the CUBE of its linear dimensions, while its power only goes up as the square. So a floating turbine has to live on the low end of the weight/power curve.
Your vituperation would be more effective and justified if you'd include just a smidgen of a hint that you know anything about energy conservation.
Perhaps I've been unclear. I will repeat,using different words: It does not matter exactly how the kites work-- whether it's linear motion downstream, or tacking across the flow like a sailboat. There's only a FIXED AMOUNT of energy there. It's very dilute. I used the downstream example as a best-case example where we're capturing all the energy from a sail of the posited size. It matters not one whit whether they're tacking sideways at an angle that gives them a 10/1 glide ratio. THE AMOUNT OF ENERGY IS THE SAME. The high speed they're using just makes it easier to match the speed to the optimum intake speed for the turbine. That's mildly good engineering, but I hope they're correcting for the increased parasitic drag, which goes up as the cube of the speed.
Thanks David, for giving us a living example of the lack of knowledge of Physics in the general populance. One learns in first-semester Physics about the equivalence of physical motion and conservation laws. Speeding up the water flow for the turbine is just basic impedance-matching, it's not creating any energy or tapping any hidden font of power.
The basic issues are that there is very limited and diffuse power in tidal flow, and the significant cost and short life of the equipment to capture that energy. For sturdier equipment, like coffer dams, you also have to consider the cost of money. I wish these folks every bit of luck, but they're working in a very difficult and cost-sensitive area.
Sorry for my oversimplifying things. I just assumed everybody knew basic physics and knew that it matters not one whit whether the kite is slewing, sliding, swooping or gliding, and it matters even less what speed it's achieving, or whether the water is moving horizontally or vertically, or whether the kite is moving and the water is behind it, or the kite is standing still and the water is moving past it or through it. All those fancy scenarios are just different angular projections of the same basic kinematics. You can't make any more energy than is available by the basic fact that water is dropping in a gravitational field.
Even if my assumptions are off by a factor of ten, we are still a very long way from even paying the interest on the capital investment, much less paying off the investment, which is in effect saying that we're going to be going to a lot of trouble to lose energy.
Slashdot Mirror
The "common toy" is not a radiometer. It's a heat engine. The bulb is only partially evacuated and the hotter, black side of the vanes heats up the gas molecules, which then bounce off it with increased vigor, compared to the white side. So the vanes spin with the white side going forward.
A true radiometer would be bouncing photons off the white side, and spinning with the black side leading.
The heat-engine version has many times the efficiency of the photon one.
Er, if you actually try to go read TFA, it seems they analyze the text by semi-numerological means.
Like noticing that one particular argument is about 1/12th the length of the chapter, from that somehow drawing some far-fetched conclusion.
Sounds like a particularly bizarre form of BS to me.
Way back around 1972, I worked on a CDC time-share system. They charged 4 cents per CPU second, 1 cent per PRU (640 characters) transferred to/from disk, and 0.2 cents per kiloword-second of memory used.
Except after 5PM, when the rates went down 50%.
Luckily I worked for the computer center, so the long assembly times ( 5 minutes ) were charged against a funny-money account. Still it was humbling that one missing comma and I'd wasted about 20 minutes of real time and $12, when $12 was real money.
Are you sure you haven't reinvented the wheel?
Mapmakers and Mathematicians have been working in this area for like, centuries.
If you're talking straight-line, great-circle routes, that was reduced to simple formulas a very long time ago.
If you're talking about contour-following, or minimum-energy paths, or road-following, that was worked out before we were born.
If you're talking about efficient algorithms for searching geo databases, that's been well plumbed too.
If you're talking about an efficient algorithm or implementation on a particular platform, that's not so much science, as a blurb in Dr Dobbs.
What a ridiculous story.
Nobody is going to invest the needed billions of dollars in a country with no real government, no laws, no protection for private property, and every expectation of being taken over by the Taliban as soon as the US army leaves.
It would take billions in up-front investment, as Afghanistan does not have any of the needed things: water, power, roads, engineers, chemical plants, railroads, ports, diging machines, huge trucks, smelters, coal, oil, and gas. Billions, and at least ten years to build the infrastructure before a pound of ore comes out of there.
And minerals only get extracted if the cost is less there than from the developed sources. That's unlikely, due to the needed up-front investment. And one of the alleged largest supplies, Lithium, is already being mined very, very cheaply in South America, where there are huge easily-accessed deposits.
Ridiculous.
Someone needs to inform whomver wrote this story:
* Mining-company geologists have been scouring the globe for centuries, looking for mineral deposits that are economically recoverable.
* Minerals do not know about arbitrary political boundaries, making it highly unlikely that this "treasure-trove", if it exists, is wholly contained in Afghanistan.
* Minerals are heavy and hard to extract, which makes it paramount that there be things that Afghanistan has none of, such as rail lines, roads, ports, docks, electricity, coal, fresh water, chemicals, a stable government, a stable economy, and much more. Lacking just one of those items can make mining an impractical venture.
* No bank is going to loan the hundreds of millions to billions needed to even begin to extract these minerals. Banks do not loan money into war zones with no history of a stable government or protection of private property, and when the only source of quasi-stability, the US military, is on a countdown to leave the country.
Funny, but I was just reading an old radio magazine, circa 1938, where they were using 5 to 6 volts to the heater of a rectifier tube that usually needed 25. That's 1/4 the voltage, about 1/16th the power, and the rectifier worked BETTER at detecting radio signals than at full voltage. Some complex thing about the diode work functions one might suspect.
Engineers have explored most corners of the performance envelope, nothing all that new under the sun.
Dang, and here I'd al;ways assumed "Zero Day" meant the bug had been there since the day the software was released. Like the bug in the .BMP rasterizer, revealed in 2004, that had been there since Windows 3.0
Who manages the canonical definition of "Zero Day" ?
>Comparing something unfavorably to NTSC is going to get you marked as a nutter.
You miss the point. The point is you only need 1/3 the color resolution, compared to monochrome. NTSC is just one example where this eyeball ratio is put into practice.
>>The iPhone, and every other LCD screen, has three color >>elements per pixel, while the eye has like 1/3. That's a NINE >>TIMES difference that this guy is glossing over.
>Not really. You can see those pixels, so they're relevant.
I can't see the individual color elements on a 72dpi display. Even if I could, the 9/1 ratio still holds.
balderdash and poppycock, on so many levels:
(1) The human eye has very variable resolution. Down in the fovea it may be up at this guy's numbers, but much less everywhere else.
(2) The eye's color receptors are much farther apart, and therefore of poorer resolution, that the monochrome receptors. That's why the old NTSC standard had about 1/3 the color bandwidth than the Y bandwidth.
(3) The iPhone, and every other LCD screen, has three color elements per pixel, while the eye has like 1/3. That's a NINE TIMES difference that this guy is glossing over.
(4) It really doesn't matter. We don't spend our lives inspecting individual pixels-- we let our brain process the images into coherent high-level objects, such as "letters" and "faces".
Otherwise okay.
The problem is not in separating oil from water-- gravity already does that quite well, without the intervention of some special cloth.
The problem is the dilution-- the stuff is spread over thousands of square miles.
So pathetic. This is not a way to generate electricity, but to destroy it.
You see those "p-n" junctions did not appear out of thin air-- they're the result of using scads of electricity to heat silicon to the melting point, extract it into perfect crystals, then slice it, anneal it again in an electric furnace, then more hours at 1200C to diffuse in minute amounts of p and n dopants, then more electricity to slice, dice, solder, and cement these into usable devices.
And in the end you have some very expensive, in both dollars and energy used, heat to microamps of electricity converters. And you can easily compute exactly how much electricity you get back given say a 10 degree temperature difference between the warm and slightly less warm sides. It's miniscule. Microwatts per square centimeter. Even if you wore these for 10,000 hours, you're nowhere near making back the amount of electricity, not to mention the $$$, it cost to make these things.
The problems with anti-missile defense are more basic than that:
(1) Basic geometry -- you have to station a slew of defensive missiles every 20 miles along your borders. That's because you are not going to hit anything going Mach 12 across your path-- you need a close to head-on intercept angle.
(2) Cheap and easy countermeasures. Even if you bankrupt your country setting up (1), the bad guys just switch to using sub or boat launched cruise missiles. Or low-trajectory ICBM's. Or put the bomb on a freight or passenger plane. It's mighty foolish to spend a trillion $ and have all that effort counteracted by a visit to UPS and $187.54.
JR Oppenheimer did this math in his head in 1952 as he was testifying to a govt comittee. Nothing has changed since then.
>But don't you think that they would use space-certified RAM chips for such a project?
They did, but cosmic rays come in a wide range of intensities, from feeble all the way up to having enough energy in one photon to make (baseball analogy ahead) a baseball jump.
Nice Try, MIT, but it's all paper airplane stuff if the design cannot be certified for commercial flight.
Aerodynamicists have been trying different shapes, configurations, and designs for about 120 years now.
There are low-drag designs, but most of them are impractical in some way-- like no space for landing gear, or
customer's won't ride in them ( Flying Wings, scimitar props ), or the design won't pass FAA rules. Lots of
very picky FAA rules that make lots of designs impossible to certify for flight. The rules are mostly there to
assure safety and conservative margins. For instance, the plane has to be able to climb with all engines on one side out.
That rules out lots of designs with wingtip-mounted engines and designs with insufficient power. It has to be able to land with a certain crosswind component.
It has to have a certain maximum airspeed where one engine is out and the plane is still controllable. That rules out a lot of designs with small tails.
It has to be able to sustain a catastrophic failure of one engine without it affecting other engines. That rules out a lot of designs with multiple engines close together.
In addition the plane has to be manufacturable and inspectable and serviceable. That tosses out designs with complex wings that can't be easily assembled in pieces or later on opened up for inspection and service.
Plus the plane has to have space for all the accessories-- not just landing gear, but fuel tanks, hydraulic tanks, air-conditioning packs, batteries, surge tanks, hydraulic pumps, slats, speed-brakes. That rules out a lot of designs with very thin wings.
One may hope this MIT design kept all these real-world constraints in mind.
As Turing might say: What a load of cobblers
While you might be able to make ten gazillion AND gates, you still have the minor problem of HOOKING THEM UP into some useful logical building blocks, like adders, buffers, and memory. And the bigger problem of amplifying the results to a level acceptable to the following inputs. And figuring out how to distribute power (ATP) to each amplifier.
And the signal levels are so low, thermal noise is going to induce a lot more errors than you'd like.
And the speed is not likely to be very exciting.
I would not start short-selling Intel stock based on this technology.
Nothing to see here.
Somebody got the idea that you could use this stuff to steer light onto solar cells. Reasonable mistake.
You have to steer a solar cell to follow the sun so it's collecting the most light.
Steering the light once it's hit the panel is mostly useless-- you're too late -- you're just not intercepting the sun.
For example if the sun is 45 degrees to the side, you're only getting cos(45) or 70.7% of the rays. Nothing you do at the panel can change that.
And there are already special reflector shapes that have the amazing property of steering light from many angles to one destination. And they're just plain metal surfaces, no nanotech required.
Is this a joke?
The rate of evaporation from the oceans is about 400,000 cubic kilometers per year.
To increase that by just one percent would mean pumping 4,000 km^3 of water.
Just raising that much water to 3,000 feet would take approximately, oh let's see, carry the 0x100,
about 1,651,445,966.51 horsepower. One Point Six BILLION horsepower.
Get the followup volumes too:
Noise cancelling algorithm design using sh. ( Shhhhhh... )
Real-time traffic control with bash.
Time-domain-reflectometry made easy, with sed.
GPS satellite tracking with tr.
Build a species database with Python. ... and many more ...
Sorry, not enough coffee. I meant "Law of Conservation of Energy", not "energy conservation".
>The point is that they are cheaper compared to the alternatives.
That's like saying you should stab rather than shoot yourself in the foot, as it's a whole lot cheaper.
It does not matter how cheap something is if it's still below the break-even point. In fact it's usually better to pay more as there are usually economies of size and scale.
I wish these folks well, but there's no indication from TFA that this concept is any better than the alternatives.
>It's at this point that you should realize that your understanding of Physics is a caricature of the real world.
Likewise, for you.
The total energy does not change when you narrow the river. You have X amount of water dropping by distance D. That's the energy. It's exactly the same amount of energy, whether it's flowing at 1 meter per second with a cross section of 100m^2, or flowing at 100 m/s with a cross section of 1m^2. It sure looks more impressive, but it's exactly the same amount of energy (actually, less, as the drag goes up as the cube of the speed).
These Swedes are speeding up the rate of traverse through the water, but that's just in order to have enough speed to match the intake speed needed for a tiny turbine. And it HAS to be a tiny turbine as the weight of a turbine goes up as the CUBE of its linear dimensions, while its power only goes up as the square. So a floating turbine has to live on the low end of the weight/power curve.
Your vituperation would be more effective and justified if you'd include just a smidgen of a hint that you know anything about energy conservation.
Perhaps I've been unclear. I will repeat,using different words: It does not matter exactly how the kites work-- whether it's linear motion downstream, or tacking across the flow like a sailboat. There's only a FIXED AMOUNT of energy there. It's very dilute. I used the downstream example as a best-case example where we're capturing all the energy from a sail of the posited size. It matters not one whit whether they're tacking sideways at an angle that gives them a 10/1 glide ratio. THE AMOUNT OF ENERGY IS THE SAME. The high speed they're using just makes it easier to match the speed to the optimum intake speed for the turbine. That's mildly good engineering, but I hope they're correcting for the increased parasitic drag, which goes up as the cube of the speed.
Thanks David, for giving us a living example of the lack of knowledge of Physics in the general populance. One learns in first-semester Physics about the equivalence of physical motion and conservation laws. Speeding up the water flow for the turbine is just basic impedance-matching, it's not creating any energy or tapping any hidden font of power.
The basic issues are that there is very limited and diffuse power in tidal flow, and the significant cost and short life of the equipment to capture that energy. For sturdier equipment, like coffer dams, you also have to consider the cost of money. I wish these folks every bit of luck, but they're working in a very difficult and cost-sensitive area.
Sorry for my oversimplifying things. I just assumed everybody knew basic physics and knew that it matters not one whit whether the kite is slewing, sliding, swooping or gliding, and it matters even less what speed it's achieving, or whether the water is moving horizontally or vertically, or whether the kite is moving and the water is behind it, or the kite is standing still and the water is moving past it or through it. All those fancy scenarios are just different angular projections of the same basic kinematics. You can't make any more energy than is available by the basic fact that water is dropping in a gravitational field.
Even if my assumptions are off by a factor of ten, we are still a very long way from even paying the interest on the capital investment, much less paying off the investment, which is in effect saying that we're going to be going to a lot of trouble to lose energy.