>I imagine (and this is an uneducated opinion) all the junk coal and oil plants pump out under regular circumstances is probably going to kill more people than the japan nuclear crisis over the long run
Not only that, but the junk that coal and oil plants pump out has killed/will kill more people in the short run too.
Coal power kills a million people per year worldwide. That's ~2700 a day. Half of those are in China, and ~9/10 of the rest are elsewhere outside the western world. Still, ~45 people will die in the USA today because of coal power.
The coal power plants in the Soviet Union killed more people than the Chernobyl disaster every day while it was happening. The coal power plants in the United States kill more people every day than all the civilian nuclear accidents in the USA combined. The coal power plants in Japan currently kill (much) more people than the Daichi plant every day.
The most casualties indirectly caused by Daichi will be in Germany. Because the accident caused enough political pressure to force Merkel to shut down the seven oldest nuclear plants in Germany for 3 months. During those 3 months, they would have produced 17 TWh of energy. Replacing that will kill 250 people. In those 3 months.
The entire discussion on nuclear makes me feel sick to my stomach.
When running, a cpu uses much more power. But on a mobile phone, (and on your desktop), the cpu spends most of it's time shut down and powergated waiting for an interrupt. RAM, however, has to be refreshed, and will steadily drain your batteries until they run out.
The true winner is a RAID of 2Tb WD Caviar Black & WD Raptors here. Get several of them and RAID, you get all the performance benefits of SSDs to a large degree (still failing a bit short on IOPS probably
Falling a bit short. By some 4 orders of magnitude.
The GP might have missed the point, but you certainly did. Let me put it more bluntly: Comparing the price of an ssd to a disk by $/GB is idiotic, and there is exactly as much point in it as comparing the price of your processor to the price of your ram by $/MB (looking at the size of the cache). His point wasn't that you get better $/GB in a smaller ssd -- it was that the very metric of $/GB is completely and utterly stupid when evaluating the usefulness of an ssd as an upgrade.
A SSD is not an upgrade that buys you more space. It's an upgrade that makes your computer faster. In that, practically all of them are great value; for normal desktop use I'd much rather have an Intel ssd and the crappiest still-in-production dualcore from AMD than no ssd and the most expensive available quadcore from Intel. And I have actually used both kinds of systems. That is how awesome the difference is.
(well, the high-end Intel rig was actually a mid-range i7, but it was overclocked way past any of the models they sell.)
Everybody knows that if you can design an economically viable improvement on present-day batteries, you are going to be wildly, obscenely rich. There are plenty of applications where people would be perfectly willing to pay several times more for a battery than what they are paying now if there was a significant improvement in capacity/mass. This leads to a lot of research being concentrated even on very wild potential ideas. Many are viable in the lab, but are too expensive to produce (by a margin of several orders of magnitude), too dangerous, too short-lived, or any combination thereof.
No matter how many misses there will be, this situation is more or less the ideal case for a free market to optimize for -- if it is possible to safely store more electrical energy in a smaller mass, it will be found eventually.
Btrfs. Modify-in-place filesystems work poorly with block suballocation -- reiserfs is a case study. On the other hand, Copy-on-Write filesystems lose nothing from it.
You can fix this on the filesystem level by using packed files. For the actual disk, tracking 512-byte sectors when most operating systems actually always read them in groups of 8 is just insane. (If you wish to access files by mapping them to memory, and you do, you must do so at the granularity of the virtual memory page size. Which, on all architectures worth talking about, is 4K.)
The potential energy is stored as the mass of the nucleus. If you measure all the relevant masses before and after the explosion, E=mc^2 does predict the outcome perfectly. In other words, the mass of a nucleus is the mass of all the particles in it plus the energy-equivalent mass of the energy of it's bindings. In the same way as the mass of an atom is the mass of the nucleus, the electrons, and the mass of the potential energy stored in the electrons.
No-one is saying that RDRAM is better than what we have now -- but it contained some individual innovations that have not been utilized as a part of modern memory interfaces. Perhaps finally they will. And frankly, when looking at all the interfaces of a modern computer, the memory bus is the odd one out -- It's about time we switch to a proper differential signaling packet-based point-to-point link there. Perhaps the peace in memory markets will allow the next memory standard to actually modernize the link?
Care to elaborate?
It has 95% of earth's mass, and as it orbits at 0.72 au, it suffers from roughly twice the radiation from the sun. (assuming all radiation follows 1/r^2) I assume that if it had a magnetic field, it should be able to hold on to it's hydrogen.
Actually, there's a time-honored tradition of spies posing as journalists. Because it gives them an excuse to be snoopy, and at least in Western Europe and the USA it's considered bad form to execute journalists even when they stumble upon something they shouldn't have.
Yes, but it doesn't really matter to spies because just being a spy is a war crime. Spies that get caught get executed anyway, so what's a little more?
No, the application needs to call madvise(2) for the memory regions it thinks it can share.
To make this work automatically for all similar pages, the kernel would have to compare every page in memory against every other page in memory, not something you want to do.
Note that something similar already works for normal applications -- shared libraries and program images are shared between applications until their memory regions are written to. If a program forks, every page it has is shared until it's written to.
A cache line on a modern Intel/AMD processor is actually 512 bits, or 64 bytes.
A memory bus 512 bits wide wouldn't really help much, though -- right now when dealing with memory, most of the time is spent in the various latencies. When you are fetching a lot of memory sequentially, you can get insane speeds even today. But that's not how you usually read memory -- instead, you read a few words from different locations, and the memory controller needs to activate the correct bank, row and column before you get what you need. On typical PC-10600 DDR3, that means at least 15 bus cycles just waiting around for the memory to adjust. Making the bus 512 bits wide would speed up the actual transfer to one bus cycle from the 4 what it takes currently, but that would only mean an improvement of about 15% -- at a huge cost for having to accommodate those 384 extra data lines on the chip, socket, motherboard and ram. It's better just to try to speed up the memory so burst transfers happen "fast enough".
I don't know about nvidia cards, but at least for ati the card doesn't actually have a 256 bit memory interface -- instead, it has 4 completely separate 64-bit memory channels connected to a fast ring bus. The interleaving of data on those separate memory channels is done very coarsely -- basically, entire textures and such are allocated on a single channel. This means that when that texture is being fetched, the 3 other channels can serve other requests.
This is the way I see cpu's evolve too -- even on current hardware, namely phenom 2, you get better performance when you ungang the memory channels, and wait 8 cycles for a single memory transfer instead of 4, because that way you get to wait on separate latencies on the separate channels at the same time. Of course, in the perverse case all the data you want to access resides on one of the channels, but the chance of that happening by accident is pretty much nil.
Linux can handle 4096 cores without trouble in the main kernel tree, with support for much larger images already existing in trees forked by people who actually need such things.
I just want to point out that the admins are not doing anything -- the virtual banks in eve are being run by regular players, and the admins only provide a playing field.
When they have virtual business cycles and virtual bank runs (these things already happen in EVE), they watch with glee -- the point of eve is a very in-depth economic simulation that mimics the real world, not the perfect economic system.
Because even on earth making a good enough mirror for astronomy is something that takes months and vast amounts of high tech. If we were to put a normal mirror on the moon, making it on earth and lifting it to moon would probably still be easier than making one on the moon.
"A June 1999 National Highway Traffic Safety Administration (NHTSA) study found that ABS increased stopping distances on loose gravel by an average of 22 percent"
Increasing stopping distances is not a good thing.
Even the earliest ABS system did a better job of modulating the brakes than any drive can by pumping them manually.
Yes. However, the early system would lose in stopping distance to simply slamming the brakes. While static friction always wins in a lab, in real life situations it's not so black and white. If the road is dirty, often even a very small amount braking will cause a slide, but after a very short slide, the tire will "dig in" and get better traction. A bad early ABS system would not let the tire to dig in, and would brake quite weakly. (the newer ones actually let the wheel lose traction for a short time before easing on the brakes.)
Threshold braking is a learned skill which too many people lack.
I'd say ABS liberates us from the need to learn it.
But note that on what counts, we agree. If you're a guy who buys a car with no abs today, you're a bloody idiot endangering himself and others.
What I disagree on is the notion that you buy ABS to reduce stopping distances. It sometimes does, sometimes doesn't, but even when it does, that's not why it's there in the first place, it's just an added benefit. The reason they were developed was to turn a proverbial sled full of rocks inevitably plunging into whatever was in front of it into something that is able to nimbly evade some situations that would otherwise likely lead to loss of life. -- And this is something that is supported by pretty much all the statistics. Even the early ABS systems that pretty much always increased the stopping distances, quite dramatically reduced accident rates.
That is demonstrably not true under many driving conditions.
And on many others, it is true. Read the wikipedia article again. (Particularly, they stop you slower when you're driving in soft snow or on a dirty dusty road - which nicely accounts for about 80% of my daily driving activities.)
Actually that is EXACTLY what they are supposed to do.
Noup. Back when ABS was first introduced, they were much worse than they are now, and a car with ABS almost universally stopped slower than one without. Yet they started reducing accident rates right away. The reason is that your average driver does not know how to pump his brakes so he can maintain steering during a fast braking, and usually deciding where you are going to end up on the road is worth more than a few meters of stopping distance.
Good luck with that since they are mandated on nearly all vehicles sold today. And please don't drive anywhere near me.
Read my statement again. I don't think it means what you think it means.
Actually, abs brakes do not really help or hurt your stopping distance in most cases, in laboratory conditions keeping static friction works best but on a dirty road just burning rubber often gets better results. But reducing stopping distance is not even what they are supposed to do. They make you able to steer your car while maximum braking, and even if that costs you a few meters of stopping distance, it's well worth it - something I found out first hand when some nice old lady decided to drive onto the highway right in front of me when I was going ~100km/h.
I know they tell you that the direction of front tires have no effect on where the car is going when the brakes are locked, and I'm pretty sure I actually tested it once or twice at safe speed, but nothing prepares for the horror when you realize that there is someone right in front of you and you are closing in fast, there is a truck coming on the opposite lane, and since you forgot to not to turn the wheel while brakes were locked, you have no idea what direction the tires are facing currently, so if you release the brakes until slow enough, you risk driving off-road or even flipping the car.
There are not enough ip addresses in the world to last for a decade at current growth of consumption. That is, if we got another ~3 billion of them free now, they would still run out until 2019.
As said before, adding a few percentage points worth of supply does nothing against exponential growth of consumption. Then again, there is this pesky new tech that would increase the supply by 7922816251426433759354395033500%, which I hope will eventually also not be enough for the usage of whatever that comes after us. But it should be enough for one galaxy...
Slashdot Mirror
>I imagine (and this is an uneducated opinion) all the junk coal and oil plants pump out under regular circumstances is probably going to kill more people than the japan nuclear crisis over the long run
Not only that, but the junk that coal and oil plants pump out has killed/will kill more people in the short run too.
Coal power kills a million people per year worldwide. That's ~2700 a day. Half of those are in China, and ~9/10 of the rest are elsewhere outside the western world. Still, ~45 people will die in the USA today because of coal power.
The coal power plants in the Soviet Union killed more people than the Chernobyl disaster every day while it was happening. The coal power plants in the United States kill more people every day than all the civilian nuclear accidents in the USA combined. The coal power plants in Japan currently kill (much) more people than the Daichi plant every day.
The most casualties indirectly caused by Daichi will be in Germany. Because the accident caused enough political pressure to force Merkel to shut down the seven oldest nuclear plants in Germany for 3 months. During those 3 months, they would have produced 17 TWh of energy. Replacing that will kill 250 people. In those 3 months.
The entire discussion on nuclear makes me feel sick to my stomach.
When running, a cpu uses much more power. But on a mobile phone, (and on your desktop), the cpu spends most of it's time shut down and powergated waiting for an interrupt. RAM, however, has to be refreshed, and will steadily drain your batteries until they run out.
Falling a bit short. By some 4 orders of magnitude.
The GP might have missed the point, but you certainly did. Let me put it more bluntly: Comparing the price of an ssd to a disk by $/GB is idiotic, and there is exactly as much point in it as comparing the price of your processor to the price of your ram by $/MB (looking at the size of the cache). His point wasn't that you get better $/GB in a smaller ssd -- it was that the very metric of $/GB is completely and utterly stupid when evaluating the usefulness of an ssd as an upgrade.
A SSD is not an upgrade that buys you more space. It's an upgrade that makes your computer faster. In that, practically all of them are great value; for normal desktop use I'd much rather have an Intel ssd and the crappiest still-in-production dualcore from AMD than no ssd and the most expensive available quadcore from Intel. And I have actually used both kinds of systems. That is how awesome the difference is.
(well, the high-end Intel rig was actually a mid-range i7, but it was overclocked way past any of the models they sell.)
Neutrinos are not particularly more common than antineutrinos. Both should annihilate in pretty much identical amounts.
Everybody knows that if you can design an economically viable improvement on present-day batteries, you are going to be wildly, obscenely rich. There are plenty of applications where people would be perfectly willing to pay several times more for a battery than what they are paying now if there was a significant improvement in capacity/mass. This leads to a lot of research being concentrated even on very wild potential ideas. Many are viable in the lab, but are too expensive to produce (by a margin of several orders of magnitude), too dangerous, too short-lived, or any combination thereof.
No matter how many misses there will be, this situation is more or less the ideal case for a free market to optimize for -- if it is possible to safely store more electrical energy in a smaller mass, it will be found eventually.
Btrfs. Modify-in-place filesystems work poorly with block suballocation -- reiserfs is a case study. On the other hand, Copy-on-Write filesystems lose nothing from it.
You can fix this on the filesystem level by using packed files. For the actual disk, tracking 512-byte sectors when most operating systems actually always read them in groups of 8 is just insane. (If you wish to access files by mapping them to memory, and you do, you must do so at the granularity of the virtual memory page size. Which, on all architectures worth talking about, is 4K.)
The potential energy is stored as the mass of the nucleus. If you measure all the relevant masses before and after the explosion, E=mc^2 does predict the outcome perfectly. In other words, the mass of a nucleus is the mass of all the particles in it plus the energy-equivalent mass of the energy of it's bindings. In the same way as the mass of an atom is the mass of the nucleus, the electrons, and the mass of the potential energy stored in the electrons.
http://en.wikipedia.org/wiki/Mount_everest#Using_bottled_oxygen
Didn't bother to use half a minute to check wikipedia?
No-one is saying that RDRAM is better than what we have now -- but it contained some individual innovations that have not been utilized as a part of modern memory interfaces. Perhaps finally they will. And frankly, when looking at all the interfaces of a modern computer, the memory bus is the odd one out -- It's about time we switch to a proper differential signaling packet-based point-to-point link there. Perhaps the peace in memory markets will allow the next memory standard to actually modernize the link?
Care to elaborate? It has 95% of earth's mass, and as it orbits at 0.72 au, it suffers from roughly twice the radiation from the sun. (assuming all radiation follows 1/r^2) I assume that if it had a magnetic field, it should be able to hold on to it's hydrogen.
Actually, there's a time-honored tradition of spies posing as journalists. Because it gives them an excuse to be snoopy, and at least in Western Europe and the USA it's considered bad form to execute journalists even when they stumble upon something they shouldn't have.
Yes, but it doesn't really matter to spies because just being a spy is a war crime. Spies that get caught get executed anyway, so what's a little more?
No, the application needs to call madvise(2) for the memory regions it thinks it can share.
To make this work automatically for all similar pages, the kernel would have to compare every page in memory against every other page in memory, not something you want to do.
Note that something similar already works for normal applications -- shared libraries and program images are shared between applications until their memory regions are written to. If a program forks, every page it has is shared until it's written to.
A cache line on a modern Intel/AMD processor is actually 512 bits, or 64 bytes.
A memory bus 512 bits wide wouldn't really help much, though -- right now when dealing with memory, most of the time is spent in the various latencies. When you are fetching a lot of memory sequentially, you can get insane speeds even today. But that's not how you usually read memory -- instead, you read a few words from different locations, and the memory controller needs to activate the correct bank, row and column before you get what you need. On typical PC-10600 DDR3, that means at least 15 bus cycles just waiting around for the memory to adjust. Making the bus 512 bits wide would speed up the actual transfer to one bus cycle from the 4 what it takes currently, but that would only mean an improvement of about 15% -- at a huge cost for having to accommodate those 384 extra data lines on the chip, socket, motherboard and ram. It's better just to try to speed up the memory so burst transfers happen "fast enough".
I don't know about nvidia cards, but at least for ati the card doesn't actually have a 256 bit memory interface -- instead, it has 4 completely separate 64-bit memory channels connected to a fast ring bus. The interleaving of data on those separate memory channels is done very coarsely -- basically, entire textures and such are allocated on a single channel. This means that when that texture is being fetched, the 3 other channels can serve other requests.
This is the way I see cpu's evolve too -- even on current hardware, namely phenom 2, you get better performance when you ungang the memory channels, and wait 8 cycles for a single memory transfer instead of 4, because that way you get to wait on separate latencies on the separate channels at the same time. Of course, in the perverse case all the data you want to access resides on one of the channels, but the chance of that happening by accident is pretty much nil.
Linux can handle 4096 cores without trouble in the main kernel tree, with support for much larger images already existing in trees forked by people who actually need such things.
You do realize that this won't work in the long run, right?
If I was comcast's rival, I'd want them to buy me out. Then I'd party for a month, and then start a new company to challenge them.
I just want to point out that the admins are not doing anything -- the virtual banks in eve are being run by regular players, and the admins only provide a playing field.
When they have virtual business cycles and virtual bank runs (these things already happen in EVE), they watch with glee -- the point of eve is a very in-depth economic simulation that mimics the real world, not the perfect economic system.
Because even on earth making a good enough mirror for astronomy is something that takes months and vast amounts of high tech. If we were to put a normal mirror on the moon, making it on earth and lifting it to moon would probably still be easier than making one on the moon.
Sometimes I read too fast for my own good.
Still not over ;)
"A June 1999 National Highway Traffic Safety Administration (NHTSA) study found that ABS increased stopping distances on loose gravel by an average of 22 percent"
Increasing stopping distances is not a good thing.
Even the earliest ABS system did a better job of modulating the brakes than any drive can by pumping them manually.
Yes. However, the early system would lose in stopping distance to simply slamming the brakes. While static friction always wins in a lab, in real life situations it's not so black and white. If the road is dirty, often even a very small amount braking will cause a slide, but after a very short slide, the tire will "dig in" and get better traction. A bad early ABS system would not let the tire to dig in, and would brake quite weakly. (the newer ones actually let the wheel lose traction for a short time before easing on the brakes.)
Threshold braking is a learned skill which too many people lack.
I'd say ABS liberates us from the need to learn it.
But note that on what counts, we agree. If you're a guy who buys a car with no abs today, you're a bloody idiot endangering himself and others. What I disagree on is the notion that you buy ABS to reduce stopping distances. It sometimes does, sometimes doesn't, but even when it does, that's not why it's there in the first place, it's just an added benefit. The reason they were developed was to turn a proverbial sled full of rocks inevitably plunging into whatever was in front of it into something that is able to nimbly evade some situations that would otherwise likely lead to loss of life. -- And this is something that is supported by pretty much all the statistics. Even the early ABS systems that pretty much always increased the stopping distances, quite dramatically reduced accident rates.
That is demonstrably not true under many driving conditions.
And on many others, it is true. Read the wikipedia article again. (Particularly, they stop you slower when you're driving in soft snow or on a dirty dusty road - which nicely accounts for about 80% of my daily driving activities.)
Actually that is EXACTLY what they are supposed to do.
Noup. Back when ABS was first introduced, they were much worse than they are now, and a car with ABS almost universally stopped slower than one without. Yet they started reducing accident rates right away. The reason is that your average driver does not know how to pump his brakes so he can maintain steering during a fast braking, and usually deciding where you are going to end up on the road is worth more than a few meters of stopping distance.
Good luck with that since they are mandated on nearly all vehicles sold today. And please don't drive anywhere near me.
Read my statement again. I don't think it means what you think it means.
Actually, abs brakes do not really help or hurt your stopping distance in most cases, in laboratory conditions keeping static friction works best but on a dirty road just burning rubber often gets better results. But reducing stopping distance is not even what they are supposed to do. They make you able to steer your car while maximum braking, and even if that costs you a few meters of stopping distance, it's well worth it - something I found out first hand when some nice old lady decided to drive onto the highway right in front of me when I was going ~100km/h.
I know they tell you that the direction of front tires have no effect on where the car is going when the brakes are locked, and I'm pretty sure I actually tested it once or twice at safe speed, but nothing prepares for the horror when you realize that there is someone right in front of you and you are closing in fast, there is a truck coming on the opposite lane, and since you forgot to not to turn the wheel while brakes were locked, you have no idea what direction the tires are facing currently, so if you release the brakes until slow enough, you risk driving off-road or even flipping the car.
Would not buy a car without ABS again.
There are not enough ip addresses in the world to last for a decade at current growth of consumption. That is, if we got another ~3 billion of them free now, they would still run out until 2019.
As said before, adding a few percentage points worth of supply does nothing against exponential growth of consumption. Then again, there is this pesky new tech that would increase the supply by 7922816251426433759354395033500%, which I hope will eventually also not be enough for the usage of whatever that comes after us. But it should be enough for one galaxy...
In my experience, at least 90% of Vista problems are driver-related. You have a good set of hardware.