Thats why I stay with Intel graphics, they may not be the fastest graphics card in the world but their Linux drivers never made any problems. Be it mode switching or dual-head, everything *just works*.
And that by using existing Xorg standards (XRandR, etc.), e.g. no nightmare "nvidia-settings" program as in the case for "the other" manufacturer.
And yes, for most models the drivers are open source, IIRC.
A not totally off-topic question: Can anyone recommend a free data visualization and analysis/plotting package? Something a bit more powerful than gnuplot:)
Personally (I am a lay person astronomically), I think Dark Matter raises more questions than it answers. While I acknowledge the effort, time and rigor that many astrophysicists have put into studying these phenomena, I still feel that dark matter, a substance which is invisible, intangible, and undetectable expect through its gravitational effects is too far of a step for physics to take without more evidence. I feel as a theory, dark matter is only a stepping stone on the way to a better explanation for what we are observing.
That certainly is the nightmare version of Dark Matter. However, most (if not all?) dark matter models do not in fact propose (other than gravitationally) noninteracting particles. They certainly must interact very weak with each other and ordinary matter but not per se not at all.
Dark Matter models make in fact verifiable predictions, such as annihilation products and rates (positrons in this case). They are valid science!
I'm a smart software developer, so I'm pretty sure my computer is not affected (secured hardware firewall, etc). But how can I be sure? I firmly believe that you can never be sure. It all comes down to trust: Do you trust - morally and technicaly - the people who wrote the programs you are running and the people who compiled them and those who packaged them onto a CD or a webserver... and so on.
As it is nowadays impossible to have complete insight into all your running softwere let alone your hardware, you will never be sure. But you can have confidence:)
IIRC they keep the fuel in a hot liquid state and basically keep it covered in molten sodium. You do realize that there are accidents other than uncontrolled nuclear chain reactions? A simple _chemical_ sodium fire for example (http://yro.slashdot.org/article.pl?sid=08/01/27/180239&from=rss). Hot sodium is probably one of the most aggressive chemical you can come across.
I personally do not believe they are safe but I have noticed that whenever this is brought up on Slashdot dozens of posts are sent in reply claiming that nuclear has "Come so far" since Chernobyl and 3 Mile Island and "nothing like that could ever happen again" It always amazes me how people stick to that line of reasoning. I hope they realize that there are _still_ RBMK reactors (Chernobyl-type) operating today in Russia. Some of them had accidents with partial core meltdown in the past (The "Leningrad Nuclear Power Plant").
[quote]They didn't, they've created a detector which works out whether the chip was hit by a cosmic ray or not.[/quote]
As the GP said, there is no way of knowing wheter a cosmic ray passed through you or not. The cosmic ray could easily just smash your bit to a new, random state and pass happily unhindered through the actual detector thingy. Only way to improve the situation would be to build a large detector volume (at least a couple cm^3).
The companies are defending themselves by saying "you would need a high level of expertise to break the system". But all you need is a basic understanding of soldering, an Internet connection (to Google some tutorials) and unsupervised access to a terminal (something that any store employee being paid minimum wage would have on occasion in theory). The problem is, they are getting away with it! At least here in Europe, jurisdiction usually assumes that the system is secure and that the fact that you lost money in a card fraud must therefore be obviously some negligence on your part.
Studies like this one are therefore very important, if only to educate judges...
Well, you have to admit that that in this case the paper clip is quite important. For those of you who haven't actually read the article (it is not unheard of!): They use it to peel through a hole in the back of a owner-accessible compartment for some rarely used extra modules to insert it into an open via in the pcb which just happens to carry a serial data line transmitting PIN and card details...
You could even nicely mount your eavesdropper circuit in that compartment.
This is quite startling IMO, as the designers of the module have gone to great lengths to hide most signals under layers of a dense sensor-maze to prevent access by drilling your way into the lower layers of the circuit board.
Before I answer every reply I'll just reply to myself:
Perhaps I should have made myself clearer: I am not for cellar nuclear power generation. I know that nuclear reactors have to be large to be efficient. That's just the point, that makes them less useful.
I actually agree with you. My points was that because of this fact, nuclear power plants are less suited _generally_ to serve a large fraction of the generated power, from a net reliability point of view.
You really think a tiny little turbine is going to be as efficient as a huge one in a powerplant? Yes, I do. Of course the great turbine in the power plant is more efficient as my tiny little local one, but the power from the large, centralised and thereby far-off power plant has to come to me first. The biggest consumer on the net is the net itself. Most of the power is just lost traveling to my home!
But that's not even the worst part: what about all the heat? In a big power plant it is usually just blown in the air (or at most used locally). With village-sized plant most of it could be harnessed.
Ok, in Florida you probably don't have to heat that much during the year(?) but its rumored that there are unfriendlier places.
Generally bigger things are more efficient. (Excluding future techs and unobtainium). Like a centrally planned economy? Or perhaps like a mainframe? We should ask the dinosaurs!
One does *not* currently try to break up generation into smaller parts for nuclear reactors...
For nuclear, the economics of initial construction and design requirements make much more sense to do huge reactors. A reactor has to have huge amounts of shielding for protection in case of mishap (it's mostly not for the regular reaction from the core). We're talking shells of concrete several feet thick. And steel too. It's cheaper the larger your volume/power ratio and such is. You are arguing from the assumption that power has to be generated nuclear and that by their nature they work better centralised. Granted! But my argument was that because of this very fact, power generation would be better off with a smaller fraction of nuclear powered energy, because of the inherent drawbacks of centralization.
The circuit being broken, the reactors had no place to dump their power output, so they automatically shut off. That's what is supposed to happen. Nothing nuclear to see here, move along. Well, that's not entirely true and you probably know that. Because of their - well - nuclear nature you cannot arbitrarily switch a nuclear power plant on and off, reactor start up can in fact easily take a couple of days. So from a power grid point of view the nuclear power plants are _part_ of this problem.
Of course, one can have various definitions of "huge" (insert Viagra jokes here), but the US Navy might not agree with you.
But I really don't think it's a good idea for everyone to have a nuclear reactor in their cellar. Most folks don't have the technologic where-with-all to keep their PC's or cars running correctly. Until and unless you can get any power generation technology simple enough that it rivals a toaster in complexity, I will take centralized facilities any day. Yes, it's mostly because of security concerns! But that's just the point, you can build small nuclear reactors - but build securely (that is with multiple layers of containment, emergency automation, a couple of engineers, etc. pp.) they just aren't profitable. That is, if you not happen to be the military - they have quite different views on cost-benefit:)
Well, the problem is that huge, bulky plants are much more fragile - in terms of network disruptions - than a more distributed net of many smaller plants.
Nuclear plants however are only available in the huge, bulky variation. In fact they come from some technological stone-age where the idea of giant-gigawatt-city-plants was considered the best solution imaginable.
Nowadays one tries to break power generation up into much smaller parts - perhaps as far as to your own cellar. This would have in fact many advantages besides reliability, "combined heat and power" comes to mind.
Oh, and what i forgot to say: You can't demagnetize a permanent magnet, that's why they are they are called _permanent_. You can disorder his large scale orientation (by heating the thing up for example). But each individual atom is still a permanent magnet, albeit not oriented the same way as every other atom.
If this is the case then expect the 'permanent' magnets to lose their magnetism over time, and if this magnetism was imparted to them from an industrial process (ie. they are not naturally magnetic) then the extra energy would be coming from the magnet factory's machinery.
I don't know, there isn't that much energy in a magnet. Let's see:
The energy density of a magnetic field is: B^2 / (2 mu0). Assume a very strong rare earth magnet with a field of 1 Tesla, which gives a magnetic energy density of 0,4 Joule/cm^3
Comparison: a 1000mAh 1,5V AA battery has an energy density of roughly 700 J / cm^3
Since they are accelerated, I'm guessing they penetrate further, but they will be stopped quicker too (charge, mass, volume, all these will make them easier to stop than high energy photon radiation). Best of all, it's the stopping/slowing of the protons that kills the cells (they hit stuff, break stuff, and stop/slow down), so less energy will be needed since the majority of the high-energy photons would just pass through. The trickiest part would be to determine how many protons and with how much energy.
For a nice picture of energy deposition vs. depth see e.g. http://www.gsi.de/forschung/bio/energy_e.html
One can adjust the peak energy deposition's depth by varying the proton's energy. The surrounding tissue gets a much lower dose than in X-Ray irradiations.
Combine the particle accelerator with a PET (http://en.wikipedia.org/wiki/Positron_emission_tomography) and you can irradiate a cancer with cubic millimeter resolution.
This is in fact the only way to increase your potential payout in a (fair) lottery. I read an article about it once, which analyzed thousands of bets from people for the lottery of Switzerland, IIRC. There were all sorts of accumulations of specific sequences, 1 2 3 4 5, prime numbers, Fibonacci's numbers, dates, geometric arrangements on the rectangular number grid of the Swiss lottery ticket etc. Conclusion was that to increase the (still negative, of course) expectation value of your win
one should pick at random from the set of commonly unused sequences.
Of course this would only increase your expectation value up to the theoretic negative expection value if you were the only player...
Slashdot Mirror
Thats why I stay with Intel graphics, they may not be the fastest graphics card in the world
but their Linux drivers never made any problems. Be it mode switching or dual-head, everything *just works*.
And that by using existing Xorg standards (XRandR, etc.), e.g. no nightmare "nvidia-settings" program as in the case for "the other" manufacturer.
And yes, for most models the drivers are open source, IIRC.
A not totally off-topic question: Can anyone recommend a free data visualization and analysis/plotting package? Something a bit more powerful than gnuplot :)
Personally (I am a lay person astronomically), I think Dark Matter raises more questions than it answers. While I acknowledge the effort, time and rigor that many astrophysicists have put into studying these phenomena, I still feel that dark matter, a substance which is invisible, intangible, and undetectable expect through its gravitational effects is too far of a step for physics to take without more evidence. I feel as a theory, dark matter is only a stepping stone on the way to a better explanation for what we are observing.
That certainly is the nightmare version of Dark Matter. However, most (if not all?) dark matter models do not in fact propose (other than gravitationally) noninteracting particles. They certainly must interact very weak with each other and ordinary matter but not per se not at all.
Dark Matter models make in fact verifiable predictions, such as annihilation products and rates (positrons in this case). They are valid science!
Price?
As it is nowadays impossible to have complete insight into all your running softwere let alone your hardware, you will never be sure. But you can have confidence
You are joking, are you?
[quote]They didn't, they've created a detector which works out whether the chip was hit by a cosmic ray or not.[/quote]
As the GP said, there is no way of knowing wheter a cosmic ray passed through you or not. The cosmic ray could easily just smash your bit to a new, random state and pass happily unhindered through the actual detector thingy. Only way to improve the situation would be to build a large detector volume (at least a couple cm^3).
Studies like this one are therefore very important, if only to educate judges...
Well, you have to admit that that in this case the paper clip is quite important.
For those of you who haven't actually read the article (it is not unheard of!):
They use it to peel through a hole in the back of a owner-accessible compartment for some rarely used extra modules to insert it into an open via in the pcb which just happens to carry a serial data line transmitting PIN and card details...
You could even nicely mount your eavesdropper circuit in that compartment.
This is quite startling IMO, as the designers of the module have gone to great lengths to hide most signals under layers of a dense sensor-maze to prevent access by drilling your way into the lower layers of the circuit board.
Before I answer every reply I'll just reply to myself:
Perhaps I should have made myself clearer: I am not for cellar nuclear power generation. I know that nuclear reactors have to be large to be efficient. That's just the point, that makes them less useful.
I said "one tries" not "one does" or "you do".
I actually agree with you.
My points was that because of this fact, nuclear power plants are less suited _generally_ to serve a large fraction of the generated power, from a net reliability point of view.
Of course the great turbine in the power plant is more efficient as my tiny little local one, but the power from the large, centralised and thereby far-off power plant has to come to me first. The biggest consumer on the net is the net itself. Most of the power is just lost traveling to my home!
But that's not even the worst part: what about all the heat? In a big power plant it is usually just blown in the air (or at most used locally). With village-sized plant most of it could be harnessed.
Ok, in Florida you probably don't have to heat that much during the year(?) but its rumored that there are unfriendlier places. Generally bigger things are more efficient. (Excluding future techs and unobtainium). Like a centrally planned economy? Or perhaps like a mainframe? We should ask the dinosaurs!
For nuclear, the economics of initial construction and design requirements make much more sense to do huge reactors. A reactor has to have huge amounts of shielding for protection in case of mishap (it's mostly not for the regular reaction from the core). We're talking shells of concrete several feet thick. And steel too. It's cheaper the larger your volume/power ratio and such is. You are arguing from the assumption that power has to be generated nuclear and that by their nature they work better centralised. Granted! But my argument was that because of this very fact, power generation would be better off with a smaller fraction of nuclear powered energy, because of the inherent drawbacks of centralization.
But I really don't think it's a good idea for everyone to have a nuclear reactor in their cellar. Most folks don't have the technologic where-with-all to keep their PC's or cars running correctly. Until and unless you can get any power generation technology simple enough that it rivals a toaster in complexity, I will take centralized facilities any day. Yes, it's mostly because of security concerns! But that's just the point, you can build small nuclear reactors - but build securely (that is with multiple layers of containment, emergency automation, a couple of engineers, etc. pp.) they just aren't profitable. That is, if you not happen to be the military - they have quite different views on cost-benefit
Well, the problem is that huge, bulky plants are much more fragile - in terms of network disruptions - than a more distributed net of many smaller plants.
Nuclear plants however are only available in the huge, bulky variation. In fact they come from some technological stone-age where the idea of giant-gigawatt-city-plants was considered the best solution imaginable.
Nowadays one tries to break power generation up into much smaller parts - perhaps as far as to your own cellar. This would have in fact many advantages besides reliability, "combined heat and power" comes to mind.
Oh, and what i forgot to say: You can't demagnetize a permanent magnet, that's why they are they are called _permanent_. You can disorder his large scale orientation (by heating the thing up for example). But each individual atom is still a permanent magnet, albeit not oriented the same way as every other atom.
I don't know, there isn't that much energy in a magnet. Let's see:
The energy density of a magnetic field is: B^2 / (2 mu0). Assume a very strong rare earth magnet with a field of 1 Tesla, which gives a magnetic energy density of
0,4 Joule/cm^3
Comparison: a 1000mAh 1,5V AA battery has an energy density of roughly 700 J / cm^3
One thing: http://www.gsi.de/forschung/bio/energy_e.html is actually about heavy ions (carbon). The curve is not _too_ different for a proton, though.
For a nice picture of energy deposition vs. depth see e.g. http://www.gsi.de/forschung/bio/energy_e.html
One can adjust the peak energy deposition's depth by varying the proton's energy. The surrounding tissue gets a much lower dose than in X-Ray irradiations.
Combine the particle accelerator with a PET (http://en.wikipedia.org/wiki/Positron_emission_tomography) and you can irradiate a cancer with cubic millimeter resolution.
This is actually not a new, purely academic technique, it is already commercially available, see http://en.wikipedia.org/wiki/Proton_therapy
Attention: I'm not a doctor but a physics student :)
For everone else in the world:
1 / 30 mpg = 7.8 litre / 100 km
1 / 27.5 mpg = 8.6 litre / 100 km
Assuming 1 gallon to be 231 in^3 (you have to love those units!)
Very good point.
:)
This is in fact the only way to increase your potential payout in a (fair) lottery. I read an article about it once, which analyzed thousands of bets from people for the lottery of Switzerland, IIRC. There were all sorts of accumulations of specific sequences, 1 2 3 4 5, prime numbers, Fibonacci's numbers, dates, geometric arrangements on the rectangular number grid of the Swiss lottery ticket etc. Conclusion was that to increase the (still negative, of course) expectation value of your win one should pick at random from the set of commonly unused sequences.
Of course this would only increase your expectation value up to the theoretic negative expection value if you were the only player...
But by telling the public, it is now spoiled