But now I just remembered-- 2% efficiency in the turbine isnt going to work too well-- the bearings in the turbine and generator probably have about that much friction. A typical ball bearing has about 0.3% friction. A high-efficiency turbine is likely to need at least one on each end, maybe one or more inbetween depending on the number of stages. The generator will need two. Five times 0.3% is already 1.5%. Not a whole lot leftover for actual power generation.
"A 2% efficiency isn't a problem. Efficiency tells you the ratio of the energy you can sell to the energy you put in. But if the energy you put in costs zero, then efficiency is an utterly unimportant number."
Um, no. 2% efficiency means you need a 100 megawatt-sized turbine (to handle the large amount of working gas) which due to the very small temperature drop across it will put out a feeble 2 megawatts. Two megawatts is about $100 an hour's worth of electricity. There's no way you can even pay the maintenance, labor, or even the interest cost on a 100 megawatt turbine with that. That's why efficiency matters.
Yes, there's cold down there. So what? You need a temperature *difference* in order to extract energy with like a turbine. And Carnot's law still applies-- a small temperature difference means a very small overall efficiency.
With that small a difference it's doubtful you can generate enough power to break-even. After all, you have to run the pumps to pump up the cold water. That's not a trivial amount of energy-- water is heavy and it's waaay down there.
I'm too lazy to do the spreadsheet math right now, but a rough estimate says you can't even break even on the energy, even with an ideal turbine using some ideal working fluid that vaporizes at just the right temperature.
And any economically viable scheme has to not only be above break-even, it has to generate enough benefits to pay for the equipment and labor. Have you priced the cost of a 5,000 foot long sewer pipe recently? How's about a turbine that can extract useful power from a 40 degree F difference? Yowsa.
Reminds me of my young tender years. One of my friends had parents that were at work during the day in the summer. His mother was a nurse. In their garage there were some (old-style) intravenous infusion jars (glass), the jar caps and plastic hoses, a 5-gallon can of gasoline for the lawnmower, and an ancient but still chugging air compressor. Put all these things together, and you have a pressurized glass jar of gasoline with a convenient squirter hose. A poor-kid's flamethrower. I'm amazed we didnt all burn up several times over.
And what's left on the loan or runs your school is the principal. A concept is a principle. A big letter is a capital. The white marble round thing is the Capitol. When you don't win you lose. Not to be tight is loose. If something isnt yours it's theirs. Where they are is where they're. let's work on this, people.
Yep, Gallium does indeed conduct heat many times better than water. Too bad that's not a relevant parameter here.
With a cooling loop, you'd like a liquid that can carry a lot of heat per trip and doesnt get too hot in doing so. Water gobbles up a whole kilocalorie per cc for each degree. gallium is dreadful by comparison-- it has a FIFTEEN times poorer specific heat, so it either goes up 15 degrees per cc as it passes the GPU, or the pump has to put out 15 times the flow rate to give the same cooling rate as plain old H20.
It's amazing how much mostly uninformed discussion there is on this issue.
A quick look at the code shows it's probably easily fixed-- just add a random XORing of the polynomial table index and voila, no more cache timing artifacts. Maybe 5 lines of code total? What's the beef?
Or even better, use Hyperthreading against itself to solve the problem-- start another thread that randomly peeks at the table.
Now was that that hard to figure out?
Re:Not too well researched, like full of errors:
on
Apple's First Flops
·
· Score: 1
> It's not Apple's fault that they used an unreliable IC?
Prolly not. This was a NEW IC design, using very low power CMOS. Apple wanted a real-time clock and was willing to take the slight extra risk of using a new IC to do so. They also may have used some low-profile sockets for the RAM chips, which WAS their fault, they should have vibration and thermal shock tested the whole computer, which they apparently didnt.
Not too well researched, like full of errors:
on
Apple's First Flops
·
· Score: 4, Informative
Some errors in the article:
It was the "Sophisticated" Operating System".
The clock chip in the Apple III was unreliable, but that wasnt Apple's fault, it was an intrinsic problem with the IC.
It's doubtful that the Apple III got hot enough to unseat the chips.
It would be nearly impossible to add a fan to the Apple III without a major hack job on the case, power supply, and the large rear heatsink. It's not something that could be just tacked on.
Ok, YOU try writing a column each and every week. Sometimes you may have very little hard info to draw from. Sometimes your speculation may be a bit off kilter. Give the poor guy a break.
A little calculation suggests this is a really screwy idea. One Curie is 3.7 x 10^10 nuclear particles/sec. One coulomb is 6 x 10^23 electrons.
Assuming each particle is one electron at a (guess) 600 volts, to get 10 watts, you need 10^22 electons/sec. That's 2.7x10^12 Curies, considerably more than the total Tritium ever made. Does someone have a decimal point off by say 10 places?
I count about 15 seconds on the Apollo 11 blastoff from ignition to tower clear. The first stage burns for about 150 seconds total. That would make the fuel burnoff about 10%, not 30%.
A video projector uses a lens to focus the pixels onto the screen at a certain fixed distance. Any objects nearer or farther are going to be probed by out-of-focus probe pixels, resulting in poorer resolution. You could use RGB lasers I guess, but then you're back to mostly single probe beams, which undoes some of their somewhat clever multiple scanning beams.
Any parts of the scene that are hidden from the camera but visible to the light source will only be "visible" if there is some other object positioned just right to catch and reflect some of the backscatter. Note in the "amazing" card trick there just happens to be a book in exactly the right position to do just this.
Impressive, but mostly crap. They don't really have a diffuse light source, they're projecting a focused probe beam. So obviously, yes, you can figure out the color of what the beam is hitting, and from the beam's point of view. Big whoop. What you can't do is do this to any scene with any depth to it. And you don't get very good resolution. Not until someone comes out with a white-light laser beam. What we used to call a Polish invention.
Hmmm, well if the space shuttle has, as a worst-case estimate, one chance in 200 of blowing up, ten this new thingy will have one chance in 2000. To verify that within a reasonable error bound, they'll need to run at least ten times that many launches.
2,000 launches at 1/10th the cost means the equivalent of 200 launches using the current equipment.
Image search will kinda work for airplanes in this database,as there are a very limited set of airplane model numbers, which are going to be attached to each photo.
But if the database didnt have these text clues the image search is going to be unlikely to see the similarity between an 747 in the air, as seen from the ground, with a head-on view of a 747, or one at the gate, or one in a hangar, or one in twilight, or one of a different color.
Maybe it could be done by outsourcing the task to India?
There's probably TWO cameras, one for visible light, one for infrared. Not two cameras for binocular vision. The two "eyes" would be too close together for any usable stereoscopic effect.
If you take the worst-case possibilities, then your car car would need to somehow ensure X car-lengths of clearance on front and back, Y car-lengths to each possible angle of trajectory. Which will lead to highways carrying about 20% of current traffic, and all of it going about 30 MPH.
it's not very accurate. I spotted three factual errors in the first few pages. (1) The Edison Effect is a diode, not a "tri-valve" which isnt a word. (2) IBM had electronic adders and multipliers going before WWII (3) Von Neumann did not design the ENIAC, and the major need was for artillery tables for North Africa sand, not Navy guns.
>That's why it's a 'phased array'.
Phasing is post-processing-- it does nothing to filter out noise that saturates, de-sensitizes, or cross-modulates in the receiver.
>Fact is that this 'idea' is up and running, and it works!
Well, swell, but one has to wonder how well it works compared to, say, the same amount of money spent on better antennas in a less noisy environment. There's an awful lot of land out in the boonies, several horizons away from all that noise.
> Why would you cool a computer with a conductive liquid?
The coolant ran through pipes, not freely over the components. One might suspect mercury has some property that's lacking in the other obvious choices. Maybe they couldnt stand to have the coolant boiling and generating bubbles, which would make the center of gravity unpredictable.
Yes, it's obvious to anybody with half a brain. But from the looks of several apps out there, there's a LOT of coding being done by the lower half of the bell curve:
One very old 200,000 line app, written in assembler, that could only ever run on one particular CPU, had a #define for the number of bits in a packed character field (6 bits BTW).
One really losing Java app does about a bazillion (200+) separate SQL queries to ask for things that have not changed in 50 years. Funny, the app runs slowly, even on a rather hefty server cluster. It runs much slower than the old CICS mainframe app it replaced, which ran in one 30 MHz CPU, 4MB of RAM.
Many apps do SQL queries to get the names of the days of the week. And the names of the months. And the abbreviations for same.
What's the point of having a wireless link-- eventually you have to go to it to get the drink. What it needs is a radio direction-finder and a throwing arm. Caps for the drinks too.
Slashdot Mirror
But now I just remembered-- 2% efficiency in the turbine isnt going to work too well-- the bearings in the turbine and generator probably have about that much friction. A typical ball bearing has about 0.3% friction. A high-efficiency turbine is likely to need at least one on each end, maybe one or more inbetween depending on the number of stages. The generator will need two. Five times 0.3% is already 1.5%. Not a whole lot leftover for actual power generation.
"A 2% efficiency isn't a problem. Efficiency tells you the ratio of the energy you can sell to the energy you put in. But if the energy you put in costs zero, then efficiency is an utterly unimportant number." Um, no. 2% efficiency means you need a 100 megawatt-sized turbine (to handle the large amount of working gas) which due to the very small temperature drop across it will put out a feeble 2 megawatts. Two megawatts is about $100 an hour's worth of electricity. There's no way you can even pay the maintenance, labor, or even the interest cost on a 100 megawatt turbine with that. That's why efficiency matters.
With that small a difference it's doubtful you can generate enough power to break-even. After all, you have to run the pumps to pump up the cold water. That's not a trivial amount of energy-- water is heavy and it's waaay down there.
I'm too lazy to do the spreadsheet math right now, but a rough estimate says you can't even break even on the energy, even with an ideal turbine using some ideal working fluid that vaporizes at just the right temperature.
And any economically viable scheme has to not only be above break-even, it has to generate enough benefits to pay for the equipment and labor. Have you priced the cost of a 5,000 foot long sewer pipe recently? How's about a turbine that can extract useful power from a 40 degree F difference? Yowsa.
Reminds me of my young tender years. One of my friends had parents that were at work during the day in the summer. His mother was a nurse. In their garage there were some (old-style) intravenous infusion jars (glass), the jar caps and plastic hoses, a 5-gallon can of gasoline for the lawnmower, and an ancient but still chugging air compressor. Put all these things together, and you have a pressurized glass jar of gasoline with a convenient squirter hose. A poor-kid's flamethrower. I'm amazed we didnt all burn up several times over.
And what's left on the loan or runs your school is the principal. A concept is a principle. A big letter is a capital. The white marble round thing is the Capitol. When you don't win you lose. Not to be tight is loose. If something isnt yours it's theirs. Where they are is where they're. let's work on this, people.
With a cooling loop, you'd like a liquid that can carry a lot of heat per trip and doesnt get too hot in doing so. Water gobbles up a whole kilocalorie per cc for each degree. gallium is dreadful by comparison-- it has a FIFTEEN times poorer specific heat, so it either goes up 15 degrees per cc as it passes the GPU, or the pump has to put out 15 times the flow rate to give the same cooling rate as plain old H20.
Good old H20.
Or even better, use Hyperthreading against itself to solve the problem-- start another thread that randomly peeks at the table.
Now was that that hard to figure out?
> It's not Apple's fault that they used an unreliable IC? Prolly not. This was a NEW IC design, using very low power CMOS. Apple wanted a real-time clock and was willing to take the slight extra risk of using a new IC to do so. They also may have used some low-profile sockets for the RAM chips, which WAS their fault, they should have vibration and thermal shock tested the whole computer, which they apparently didnt.
Ok, YOU try writing a column each and every week. Sometimes you may have very little hard info to draw from. Sometimes your speculation may be a bit off kilter. Give the poor guy a break.
A little calculation suggests this is a really screwy idea. One Curie is 3.7 x 10^10 nuclear particles/sec. One coulomb is 6 x 10^23 electrons. Assuming each particle is one electron at a (guess) 600 volts, to get 10 watts, you need 10^22 electons/sec. That's 2.7x10^12 Curies, considerably more than the total Tritium ever made. Does someone have a decimal point off by say 10 places?
I count about 15 seconds on the Apollo 11 blastoff from ignition to tower clear. The first stage burns for about 150 seconds total. That would make the fuel burnoff about 10%, not 30%.
Any parts of the scene that are hidden from the camera but visible to the light source will only be "visible" if there is some other object positioned just right to catch and reflect some of the backscatter. Note in the "amazing" card trick there just happens to be a book in exactly the right position to do just this.
"last years theft" : A theft, in the last years of Cisco "last year's theft": A theft, in the previous year. Apostrophes do make a difference.
Impressive, but mostly crap. They don't really have a diffuse light source, they're projecting a focused probe beam. So obviously, yes, you can figure out the color of what the beam is hitting, and from the beam's point of view. Big whoop. What you can't do is do this to any scene with any depth to it. And you don't get very good resolution. Not until someone comes out with a white-light laser beam. What we used to call a Polish invention.
2,000 launches at 1/10th the cost means the equivalent of 200 launches using the current equipment.
Hmmm.....
Image search will kinda work for airplanes in this database,as there are a very limited set of airplane model numbers, which are going to be attached to each photo.
But if the database didnt have these text clues the image search is going to be unlikely to see the similarity between an 747 in the air, as seen from the ground, with a head-on view of a 747, or one at the gate, or one in a hangar, or one in twilight, or one of a different color.
Maybe it could be done by outsourcing the task to India?
There's probably TWO cameras, one for visible light, one for infrared. Not two cameras for binocular vision. The two "eyes" would be too close together for any usable stereoscopic effect.
If you take the worst-case possibilities, then your car car would need to somehow ensure X car-lengths of clearance on front and back, Y car-lengths to each possible angle of trajectory. Which will lead to highways carrying about 20% of current traffic, and all of it going about 30 MPH.
And that's in just the first few pages!
>That's why it's a 'phased array'. Phasing is post-processing-- it does nothing to filter out noise that saturates, de-sensitizes, or cross-modulates in the receiver. >Fact is that this 'idea' is up and running, and it works! Well, swell, but one has to wonder how well it works compared to, say, the same amount of money spent on better antennas in a less noisy environment. There's an awful lot of land out in the boonies, several horizons away from all that noise.
> Why would you cool a computer with a conductive liquid? The coolant ran through pipes, not freely over the components. One might suspect mercury has some property that's lacking in the other obvious choices. Maybe they couldnt stand to have the coolant boiling and generating bubbles, which would make the center of gravity unpredictable.
- A reasonable sized antenna for this frequency range is not very directional.
- You'd really like to have a directional antenna to block out signals that are not coming from the sky.
- You'd really like to have the antennas in a quiet locale, far from civilization.
- There's an awful lot of man-made noise and very strong signals in this frequency range.
- One lousy piece of rusty fence wire can intermix all that crud and rebroadcast all kinds of sum and difference gobs of spurious signals.
- All the supercomputers in the world are unlikely to be able to undo all the losing parameters listed above.
Sounds like a really bad idea to me.What's the point of having a wireless link-- eventually you have to go to it to get the drink. What it needs is a radio direction-finder and a throwing arm. Caps for the drinks too.