Murder is just another admittedly mean) use for the computer. The fact that guns are ilt in a way that allows murder is no reason to try to control what free citizens do with their legally obtained firearms.
Your argument seems to be based on because it is possible, is should be legal - which is anarchy.
You don't need a jammer. A jammer would broadcast losts of noise in radio frequencies. There is, however, another method that could be used to block cell phones.
Build each screen inside a Faraday cage. Essentially, the idea is to encase the whole room in a sheet of metal. The will block the radio frequency waves, and thus the phones won't work. The Faraday cage operates on the principle that the metal screen need not be complete, but that there can be gaps, provided that these gaps are small (and you can calculate how small they need to be.
To be quite honest, the simplest retro fit solution is to put a layer of tin foil on the walls, under wall papere (or paint it). Stick a sheet in the door, and you've got most of the place, and you'll probably be attenuating the signal by a good proportion. It might not prevent them from working, but it'd be
a) damn cheap
b) somewhere near 70% efficent.
Of course, the best solution is to educate the audience. Preferable with a good sized LART.
I didn't think that the G4 was that crippled in the flops department...
I was, indeed, thinking about the Alitvec. The code is well suited to vectorisation, but it's all doubles. The other code I use does most (2/3) of it's work on integers, so maxes out out POWER2's nicely (old, but reliable). Looks like it's the dual Athlon's. Damn, but I wanted to play with OS X.
Noticed something in that article - they state that the hard disk read heads use GMR sensors - not quite accurate. They use a single unit spin valve. GMR devices consist of many layers stacked on top of each other, and, more pertinantly, they operate at large magnetic fields. The sensor used have a lower field for peak sensitivity, and the change in resistance in smaller. GMR is conventionally used in the literature to indicate large, multi-layer devices. [0]
One thing that the article glosses over slightly is the difficulty in construction. Well, it's not so much a dificulty, as a paradigm shift. The metal GMR structures are built vertically onto a substrate, and thus the working current flows perpindicular to the plane of the substrate. This is distinct from traditional semiconductors, where the principle direction of the working current is parrallel to the plane of the substrate.
The notable exception would be the spin FET, but they've not actually been built yet, so it's a little tricky to comment on.
One option that the article didn't mention is the possibily of generating a magnetic semi-comductor / metal by using a conventional magnetic insulator (such as NiO, MnO or Fe2O3), and dopeing, or otherwise adjusting the electrical properties [1].
My research is into combined ab initio and statistical mechanical models of ultra thin films of the magnetic insulators. Particularly interesting is what happens when a two atom thick layer of iron is put over an NiO surface - spin dependant electron transfer, which is interesting. All in all, most of my work is the blue sky / basic building block level.
My point is that the spintronic devices require a finre degree of control in construction - by thier nature, the magnetic structure is important. Oh, and as a kicker to this, the length scale for a defect in a magnetic lattice is around 20 or so times larger than it's affect on the electrical properties. Additionally, it seems likely (to me) that other routes to mass manufacture may have to be found.
In other words: These are going to cost more. Not just because they are new, but also because they are inherently more complicated devices that electronic semiconductore devices.
[0] Well, in PhysRev anyway. IEEE and similar may use a different nomenclature
[1] My calculations suggest that a layer of NiO 4 formula units thick, or thinner, will be a metal.
Well, when I say me, I mean my group. We do computational chemistry (solid state). The more work we do, the more CPU we need. All the software we use runs on UNIX, (mostyl fortran). Scince OS X, we've had the option of looking at a Mac for purchasing CPU.
Our real problem is that the way the system of grants works, is that we can afford up to 3000 pounds (about 3800 USD), from a small grant. Otherwise, because of the accountability problem, we have to apply for a large grant specifically for a computer. This makes taking on small grants, and colaborations awkward.
All we want is CPU bang per buck. Fast networking for cluster building would be useful, but not needed, as we can spread the jobs over many indepentant machines.
I'll be watching these beasts. Could be just what I;ve been waiting for, if I can get a decent F90 compiler for them..
Copyright violation
on
Mashed-Up Music
·
· Score: 4, Insightful
Leave aside wether it is theft or not, let me indicate why this activity is illegal.
Copyright. Copyright is a right given to the author to allow them to control how thier work is used, with the intention that (but not restricted to) the rights granted to them will promote production of further works.
There means that, if you wish to use an authors work , then you have to get thier permission. They can say no. It's that simple. Consider the GPL, which relies on copyright. It is not acceptable for a company to take GPL code, add a few bits, and then sell it on. The same applies to musical works.
Granted, there is the clause of fair use. However, fair use is inherently limited, either in scope (to a few friends prehaps), or in extent (a 5 second sample, or a shot quote from a book). With my understanding, fair use doesn _not_ extend to the works outlined above.
(Consider also, that there is more than just the perfromer, there is also the writer to be considered, in terms of claims to copyright).
As well as the retro virus taking host DNA (see other reply), there is another factor at work.
Most anitbiotics, and antvirals, work by stopping the infectious agent from reproducing, and it is the hosts immune system that kills the remaining infectious agent. HIV decimates the immune system. This doesn't help.
Or my Athalon, or any of the Sparcs, or really anything that I have lying around generally
It means I don't have to by a new box to run it.
And I don't use anything to oppose Microsoft, I do use it becuase I prefer it, I can configure it the way I want, and I know that it lets me interface cleanly with the *NIX machines I use, ranging from OpenBSD firewall to UNICOS supercomputers.
Besides, if you'd actually read about it, you'd know that it uses Objective C, and fits in at the level of GNOME and KDE. It's not a kernel.
There is one method that might work - on sum data.
Godel encoding is an old technique for compression, with a fast decompress (P time). Unfortunatly the compression statge is NP (Maybe NP-Hard, can't remember).
The method relies on expressing the number as an algebraic product, that can be expressed in less space than the result.
For example, in ASCII, the string (in RPN) "7 7 ^ 34 * 99 ^ - 7 p" has 18 characters. It's expansion has 740 characters. That's a compression ratio of, what, 35:1. [Ok, so you'd never actually do it in ASCII, but it shows the technique]
The advantages of the technique are that it gives better compression on larger numbers, in principle. In general, however, other factos come into play, and it bottoms out. My analysis suggested it bottoms out somewhere around 120-150:1.
However, the disadvantaged of the scheme are numerous. Firstly, there is no known algorithm to encode efficently. The system can't stream, like gzip and LZW can. Thus far, it's just an interesting idea.
I mention this because the mult-dimensional mathematics that they are reffering to have a passing similarity somthing I was playing with a couple of years ago, to look for faster algorithms (or any, really, other than brute force). It was cute, but always slower than brute force, save a few best cases:(.
If I put my best guess ot max compression together with the uncanny similarity of the maths. Namely, you to a split into some expression, and then re-apply the algorithm to a sub expression. Then , throw it through a symbolylic computation routine, to optimise it a bit, and gzip the whole lot. It would only work well on some numbers, but you can pad it slightly to get a very different number, and try again until you get a good fit.
So stepwise:
ZeoSync intentionally randomizes naturally occurring patterns to form entropy-like random sequences through its patent pending technology known as Zero Space Tuner
Pad to a value that gives good compression
Once randomized, ZeoSync's BinaryAccelerator encodes these singular-bit-variance strings within complex combinatorial series to result in massively reduced BitPerfect equivalents
Godel encode.
The refference to may iterations suggest that they reapply the process to any large enough numbers left in the expression.
And that's a scary match, in my mind.
Of course, pinch of salt. There was a comment above about the odds of any compression thecnology being vaild being equal to teh claimed compression rate. I can't see how this might work. But I'm not writing this off just yet, it rings just true enough.
Most relevently is that int's only really in the month or so after a release (rather, the month starting a couple of weeks after a release) that Theo gets a holiday.
With that in mind, the Dec 1st release date was obvious.
You could fill the system with an alochol. Why? The thermal conducticity, and, more importantly for a system with this type of design, lower specific heat capacity. This would mean that the pumps would have to move more fluid to remove the same amount of heat at a specified temperature.
Cooling systems based on alcohols (generally iso-propyl alcohol) have thier place, but this is not one. Also, alcohols have less viscosity, and thus different pumps aught to be used for maximum performance.
Putting aluminium dust in the water tank is a stupid idea. Firstly, it'll sink, and not be of much use unless the resoviour is desingned to have turbulent flow (unlikely). Even with most of it sinking, you'd get enough going through the pumps to gently abrade away all of the delicate seals, making the system fairly useless.
If you wanted to improve the cooling over water, the best bet is to go for oil cooling, but that requires totaly differnt styles of pumps for high performance.
The best thing you could do would be to use a saturated sugar solution, as it has a slightly higher specific heat capacity, for not significantly increased viscosity.. However, that would also require you to prevent bacterial growth, and so on.
This comes, by the way, from someone in a chemistry deparment that involves a signifcant number of pumps, moving various liquids, and, indeed, a number of water cooled systems, some of which are rated at about 100kW. Water is a damn good coolant.
The way the wireless battery packs work are actually pretty cool.
Firstly, the big problem with any implanted device is the power requirement, and that it's a major health risk to leave things poking through the skin.
So, to power implanted devices (there are a number of newer pacemakers that also work in this manner), there is a coil of wires implanted into the skin (it's actually wrapped in plastic, and place in the subcutaneus layer). Then this is wired up to a rechargeable battery implanted elsewhere (generally tucked between the base of the lung and the stomach). Over the skin where the wire is implanted, a complementary coil is placed, and the current is transferred by magnetic induction.
There is one major problem of this system, and that's the limit to current transfer that's possible. However, it turns out that the maximum current before unaccepable heating effects isn't that much greater.
So this external battery pack will be in something like a bum bag, and a patch over the coils (they stick by magenetism).
In reality, the issue of power, and heat dissipation is the major barrier to implanted components, along with lack of space.
It's no more illegal than a pen to modify a book. Or a paintbrush to modify a painting.
They are selling technology that _allows_ the user (who already has the mmovies) to make small (?) modifications to the film. It's nothing that I caouldn't do with the mute button and fast forward, just a lot more convient.
If they were reselling films they've edited, that would be. But that's not what they are doing.
Doesn't matter what, but your speakers and sound card are well able to reporduce a signal to within 0.5 V.
There are operational amplifiers working at radio frequencies, that have plenty of resolution and precision.
Besides, if you were looking to do a ternery computer, you'd use -5V, 0V and +5V as your markers, drawing on decades of knowledge from analouge tecniqies, and then not have to worry.
For those that don't know, the standard RS-232C serial bus uses a tri-state signal. Has done scince it's introduction.
Um, not meaning to tell you how to do your job, but if your admining Linux machines with new kernels, and _not_ following the lists, or Kernel Traffic at least, then I don't think you can really complain about not hearing about it. It's been well known scince 2.4.[012] that there was a VM issue at heavy load.
Um, I think all the replies here I've read have missed the point. I don't think the poster was asking to be able to switch between the two VM's at complie time, but rather having one VM that was configurable.
That would allow the system to be tuned at compile time for the large servers, and for the small desktops, without haveing to have a 'one size fit's all' solution.
I've always felt that that would be the best answear. The reality is, however, that Andrea's VM would not allow for such a range of configurability, being a very simple, and thus easy to balance, system. That's not to put it down, often the simplest solution is best.
However, Rick's VM is more complex, and can, in principle be made more configurable at compile (or even run) time. It would be a lot of work, but I think that that's the best way to get good performance across the wide range of platforms.
For example, If I knew that my system would have to work with millions of very small files, and only read them once, then I would configure the VM to forget about caching the files, and keep anything that is used more than once in RAM. Or, of dealing with a computation, have large pages RAM to be swapped in or out that match with the arrays the computation uses, so that everything is pre-fetched. Yes, there are other ways of accomplishing these goals, but I think that that would be a good way to go.
If nothing else, it acknowledges that a system with 32 Meg of ram and one processor has a very different VM needs from an Octuple processor system with 32 Gig of ram.
Fusion is not a funding machine for science. Did you know that most of the fusion reactors output more energy than it takes to initiate + run? The only problem they're working on is stability. I'm sure your well aware of a time when computers worked, just were not very reliable, only of use to a few people, and very expensive.
I'm surprised that you think that research money funds an extravagant lifestyle for academics. As someone who is in reasearch, I'd like to point out that I've had job offers in industry, paying double what I currently get. My friends in accounting and managment couldn't belive how little I was offered, never mind how little I get at the moment. If you want to look how the money is spent, don't claim it goes to all to the academics.
Oh, and how much interest do you think there is in sodium ion desnsity in the upper atmosphere? Or electron interaction with air? Pretty useless, right?
Excpet that the first led to radar, and the second was the work the was pivotal in electron microscopes.
Sorry, 2GB of memory is nothing. I'm working with calculations that use about 4GB of disk space to hold interim values. If I could fit them all in ram, I might actually get some performence out of the x86 boxes. And those are small calculations.
More directly, a 64 bit processor moves twice as much data around as a 32 bit processor, assuming all bits are significant. It's worth noting that an IEEE double precision number is 8 bytes (that's 64 bits) of strorage, and so a 64 bit procesor can load, add etc a double precision FP number in one operation. That helps a lot.
It's not going to be moticable for most people, but if your doing serious calculations, then it wil show.
...reasons for mistrusting information supposedly produced by the scientific method...
...hybrid crops are often credited with producing huge agricultural benefits but there isn't a single scientific paper which can be used to verify this.
Read that again. What your desciribng here is the problem, but your dressing it up as the reason.
It looks like the GRAPE boards should be re-configurable. Why'd they do it with custom chips?
A whack of FPGA's should be pretty decent, but you can configure it for more than just as a N-body gravitational problem.
Because in silicon it's pretty much twice as fast as an FPGA gets? (EE's rule of thumb, admitingly reffering to microwave app's).
Not only that, but why would they want to do something other than N-body gravitational problems? _You_ might, but there are a lot of such problems to do, and that's what this is designed for. --
Slashdot Mirror
Apply your argument to fire arms.
Murder is just another admittedly mean) use for the computer. The fact that guns are ilt in a way that allows murder is no reason to try to control what free citizens do with their legally obtained firearms.
Your argument seems to be based on because it is possible, is should be legal - which is anarchy.
You don't need a jammer. A jammer would broadcast losts of noise in radio frequencies. There is, however, another method that could be used to block cell phones.
Build each screen inside a Faraday cage. Essentially, the idea is to encase the whole room in a sheet of metal. The will block the radio frequency waves, and thus the phones won't work. The Faraday cage operates on the principle that the metal screen need not be complete, but that there can be gaps, provided that these gaps are small (and you can calculate how small they need to be.
To be quite honest, the simplest retro fit solution is to put a layer of tin foil on the walls, under wall papere (or paint it). Stick a sheet in the door, and you've got most of the place, and you'll probably be attenuating the signal by a good proportion. It might not prevent them from working, but it'd be
a) damn cheap
b) somewhere near 70% efficent.
Of course, the best solution is to educate the audience. Preferable with a good sized LART.
I didn't think that the G4 was that crippled in the flops department...
I was, indeed, thinking about the Alitvec. The code is well suited to vectorisation, but it's all doubles. The other code I use does most (2/3) of it's work on integers, so maxes out out POWER2's nicely (old, but reliable). Looks like it's the dual Athlon's. Damn, but I wanted to play with OS X.
Oh, well.
Pun not intended.
Noticed something in that article - they state that the hard disk read heads use GMR sensors - not quite accurate. They use a single unit spin valve. GMR devices consist of many layers stacked on top of each other, and, more pertinantly, they operate at large magnetic fields. The sensor used have a lower field for peak sensitivity, and the change in resistance in smaller. GMR is conventionally used in the literature to indicate large, multi-layer devices. [0]
One thing that the article glosses over slightly is the difficulty in construction. Well, it's not so much a dificulty, as a paradigm shift. The metal GMR structures are built vertically onto a substrate, and thus the working current flows perpindicular to the plane of the substrate. This is distinct from traditional semiconductors, where the principle direction of the working current is parrallel to the plane of the substrate.
The notable exception would be the spin FET, but they've not actually been built yet, so it's a little tricky to comment on.
One option that the article didn't mention is the possibily of generating a magnetic semi-comductor / metal by using a conventional magnetic insulator (such as NiO, MnO or Fe2O3), and dopeing, or otherwise adjusting the electrical properties [1].
My research is into combined ab initio and statistical mechanical models of ultra thin films of the magnetic insulators. Particularly interesting is what happens when a two atom thick layer of iron is put over an NiO surface - spin dependant electron transfer, which is interesting. All in all, most of my work is the blue sky / basic building block level.
My point is that the spintronic devices require a finre degree of control in construction - by thier nature, the magnetic structure is important. Oh, and as a kicker to this, the length scale for a defect in a magnetic lattice is around 20 or so times larger than it's affect on the electrical properties. Additionally, it seems likely (to me) that other routes to mass manufacture may have to be found.
In other words: These are going to cost more. Not just because they are new, but also because they are inherently more complicated devices that electronic semiconductore devices.
[0] Well, in PhysRev anyway. IEEE and similar may use a different nomenclature
[1] My calculations suggest that a layer of NiO 4 formula units thick, or thinner, will be a metal.
Well, when I say me, I mean my group. We do computational chemistry (solid state). The more work we do, the more CPU we need. All the software we use runs on UNIX, (mostyl fortran). Scince OS X, we've had the option of looking at a Mac for purchasing CPU.
Our real problem is that the way the system of grants works, is that we can afford up to 3000 pounds (about 3800 USD), from a small grant. Otherwise, because of the accountability problem, we have to apply for a large grant specifically for a computer. This makes taking on small grants, and colaborations awkward.
All we want is CPU bang per buck. Fast networking for cluster building would be useful, but not needed, as we can spread the jobs over many indepentant machines.
I'll be watching these beasts. Could be just what I;ve been waiting for, if I can get a decent F90 compiler for them..
Leave aside wether it is theft or not, let me indicate why this activity is illegal.
Copyright. Copyright is a right given to the author to allow them to control how thier work is used, with the intention that (but not restricted to) the rights granted to them will promote production of further works.
There means that, if you wish to use an authors work , then you have to get thier permission. They can say no. It's that simple. Consider the GPL, which relies on copyright. It is not acceptable for a company to take GPL code, add a few bits, and then sell it on. The same applies to musical works.
Granted, there is the clause of fair use. However, fair use is inherently limited, either in scope (to a few friends prehaps), or in extent (a 5 second sample, or a shot quote from a book). With my understanding, fair use doesn _not_ extend to the works outlined above.
(Consider also, that there is more than just the perfromer, there is also the writer to be considered, in terms of claims to copyright).
As well as the retro virus taking host DNA (see other reply), there is another factor at work.
Most anitbiotics, and antvirals, work by stopping the infectious agent from reproducing, and it is the hosts immune system that kills the remaining infectious agent. HIV decimates the immune system. This doesn't help.
Take an AK-47
Take a magazine of 7.62 mm S ammo, and load.
Walk into computer room.
Aim at [Your Dell, The IBM]. Empty magazine at said server.
Which one still works?
More usefully, which one can you fix up back to full condition, without losing a single email?
That's reliability.
Sounds like you've not unmuted the audio channels.
Muteing is separate from the volume, in alsamixer I think the key is M to mute/unmute.
Do excuse if you already tried that, but you haven't mentioned it, and it is one of the gotcha's with ALSA (Anyone know why they mute by default?)
Or my Athalon, or any of the Sparcs, or really anything that I have lying around generally
It means I don't have to by a new box to run it.
And I don't use anything to oppose Microsoft, I do use it becuase I prefer it, I can configure it the way I want, and I know that it lets me interface cleanly with the *NIX machines I use, ranging from OpenBSD firewall to UNICOS supercomputers.
Besides, if you'd actually read about it, you'd know that it uses Objective C, and fits in at the level of GNOME and KDE. It's not a kernel.
There is one method that might work - on sum data.
:(.
Godel encoding is an old technique for compression, with a fast decompress (P time). Unfortunatly the compression statge is NP (Maybe NP-Hard, can't remember).
The method relies on expressing the number as an algebraic product, that can be expressed in less space than the result.
For example, in ASCII, the string (in RPN) "7 7 ^ 34 * 99 ^ - 7 p" has 18 characters. It's expansion has 740 characters. That's a compression ratio of, what, 35:1. [Ok, so you'd never actually do it in ASCII, but it shows the technique]
The advantages of the technique are that it gives better compression on larger numbers, in principle. In general, however, other factos come into play, and it bottoms out. My analysis suggested it bottoms out somewhere around 120-150:1.
However, the disadvantaged of the scheme are numerous. Firstly, there is no known algorithm to encode efficently. The system can't stream, like gzip and LZW can. Thus far, it's just an interesting idea.
I mention this because the mult-dimensional mathematics that they are reffering to have a passing similarity somthing I was playing with a couple of years ago, to look for faster algorithms (or any, really, other than brute force). It was cute, but always slower than brute force, save a few best cases
If I put my best guess ot max compression together with the uncanny similarity of the maths. Namely, you to a split into some expression, and then re-apply the algorithm to a sub expression. Then , throw it through a symbolylic computation routine, to optimise it a bit, and gzip the whole lot. It would only work well on some numbers, but you can pad it slightly to get a very different number, and try again until you get a good fit.
So stepwise:
ZeoSync intentionally randomizes naturally occurring patterns to form entropy-like random sequences through its patent pending technology known as Zero Space Tuner
Pad to a value that gives good compression
Once randomized, ZeoSync's BinaryAccelerator encodes these singular-bit-variance strings within complex combinatorial series to result in massively reduced BitPerfect equivalents
Godel encode.
The refference to may iterations suggest that they reapply the process to any large enough numbers left in the expression.
And that's a scary match, in my mind.
Of course, pinch of salt. There was a comment above about the odds of any compression thecnology being vaild being equal to teh claimed compression rate. I can't see how this might work. But I'm not writing this off just yet, it rings just true enough.
Most relevently is that int's only really in the month or so after a release (rather, the month starting a couple of weeks after a release) that Theo gets a holiday.
With that in mind, the Dec 1st release date was obvious.
You could fill the system with an alochol. Why? The thermal conducticity, and, more importantly for a system with this type of design, lower specific heat capacity. This would mean that the pumps would have to move more fluid to remove the same amount of heat at a specified temperature.
Cooling systems based on alcohols (generally iso-propyl alcohol) have thier place, but this is not one. Also, alcohols have less viscosity, and thus different pumps aught to be used for maximum performance.
Putting aluminium dust in the water tank is a stupid idea. Firstly, it'll sink, and not be of much use unless the resoviour is desingned to have turbulent flow (unlikely). Even with most of it sinking, you'd get enough going through the pumps to gently abrade away all of the delicate seals, making the system fairly useless.
If you wanted to improve the cooling over water, the best bet is to go for oil cooling, but that requires totaly differnt styles of pumps for high performance.
The best thing you could do would be to use a saturated sugar solution, as it has a slightly higher specific heat capacity, for not significantly increased viscosity.. However, that would also require you to prevent bacterial growth, and so on.
This comes, by the way, from someone in a chemistry deparment that involves a signifcant number of pumps, moving various liquids, and, indeed, a number of water cooled systems, some of which are rated at about 100kW. Water is a damn good coolant.
The way the wireless battery packs work are actually pretty cool.
Firstly, the big problem with any implanted device is the power requirement, and that it's a major health risk to leave things poking through the skin.
So, to power implanted devices (there are a number of newer pacemakers that also work in this manner), there is a coil of wires implanted into the skin (it's actually wrapped in plastic, and place in the subcutaneus layer). Then this is wired up to a rechargeable battery implanted elsewhere (generally tucked between the base of the lung and the stomach). Over the skin where the wire is implanted, a complementary coil is placed, and the current is transferred by magnetic induction.
There is one major problem of this system, and that's the limit to current transfer that's possible. However, it turns out that the maximum current before unaccepable heating effects isn't that much greater.
So this external battery pack will be in something like a bum bag, and a patch over the coils (they stick by magenetism).
In reality, the issue of power, and heat dissipation is the major barrier to implanted components, along with lack of space.
It's no more illegal than a pen to modify a book. Or a paintbrush to modify a painting.
They are selling technology that _allows_ the user (who already has the mmovies) to make small (?) modifications to the film. It's nothing that I caouldn't do with the mute button and fast forward, just a lot more convient.
If they were reselling films they've edited, that would be. But that's not what they are doing.
Um, sorry. Listen to something.
Doesn't matter what, but your speakers and sound card are well able to reporduce a signal to within 0.5 V.
There are operational amplifiers working at radio frequencies, that have plenty of resolution and precision.
Besides, if you were looking to do a ternery computer, you'd use -5V, 0V and +5V as your markers, drawing on decades of knowledge from analouge tecniqies, and then not have to worry.
For those that don't know, the standard RS-232C serial bus uses a tri-state signal. Has done scince it's introduction.
Um, not meaning to tell you how to do your job, but if your admining Linux machines with new kernels, and _not_ following the lists, or Kernel Traffic at least, then I don't think you can really complain about not hearing about it. It's been well known scince 2.4.[012] that there was a VM issue at heavy load.
Um, I think all the replies here I've read have missed the point. I don't think the poster was asking to be able to switch between the two VM's at complie time, but rather having one VM that was configurable.
That would allow the system to be tuned at compile time for the large servers, and for the small desktops, without haveing to have a 'one size fit's all' solution.
I've always felt that that would be the best answear. The reality is, however, that Andrea's VM would not allow for such a range of configurability, being a very simple, and thus easy to balance, system. That's not to put it down, often the simplest solution is best.
However, Rick's VM is more complex, and can, in principle be made more configurable at compile (or even run) time. It would be a lot of work, but I think that that's the best way to get good performance across the wide range of platforms.
For example, If I knew that my system would have to work with millions of very small files, and only read them once, then I would configure the VM to forget about caching the files, and keep anything that is used more than once in RAM. Or, of dealing with a computation, have large pages RAM to be swapped in or out that match with the arrays the computation uses, so that everything is pre-fetched. Yes, there are other ways of accomplishing these goals, but I think that that would be a good way to go.
If nothing else, it acknowledges that a system with 32 Meg of ram and one processor has a very different VM needs from an Octuple processor system with 32 Gig of ram.
Fusion is not a funding machine for science. Did you know that most of the fusion reactors output more energy than it takes to initiate + run? The only problem they're working on is stability. I'm sure your well aware of a time when computers worked, just were not very reliable, only of use to a few people, and very expensive.
I'm surprised that you think that research money funds an extravagant lifestyle for academics. As someone who is in reasearch, I'd like to point out that I've had job offers in industry, paying double what I currently get. My friends in accounting and managment couldn't belive how little I was offered, never mind how little I get at the moment. If you want to look how the money is spent, don't claim it goes to all to the academics.
Oh, and how much interest do you think there is in sodium ion desnsity in the upper atmosphere? Or electron interaction with air? Pretty useless, right?
Excpet that the first led to radar, and the second was the work the was pivotal in electron microscopes.
Sorry, 2GB of memory is nothing. I'm working with calculations that use about 4GB of disk space to hold interim values. If I could fit them all in ram, I might actually get some performence out of the x86 boxes. And those are small calculations.
More directly, a 64 bit processor moves twice as much data around as a 32 bit processor, assuming all bits are significant. It's worth noting that an IEEE double precision number is 8 bytes (that's 64 bits) of strorage, and so a 64 bit procesor can load, add etc a double precision FP number in one operation. That helps a lot.
It's not going to be moticable for most people, but if your doing serious calculations, then it wil show.
Your close.
Your calculating for 9 9's. They count the two before the decimal place too.
If you allow for 100x as much downtime, you;d get 3.15 seconds, for 7 9's. Which is what they quote.
...reasons for mistrusting information supposedly produced by the scientific method...
...hybrid crops are often credited with producing huge agricultural benefits but there isn't a single scientific paper which can be used to verify this.
Read that again. What your desciribng here is the problem, but your dressing it up as the reason.
One at work, that I have sole control over (along with a 100 Mbps net connect), my personal burner at home, and my parents one.
My interpetation of what he means is that:
The code was a cut and paste from the ROT-13 code, with a few lines changed, so that it used a different fixed position cypher.
In otherwords, it didn't do ROT-13. That comment is about how much time they actually spent writing the plugin (virtually nil)
--
Because in silicon it's pretty much twice as fast as an FPGA gets? (EE's rule of thumb, admitingly reffering to microwave app's).
Not only that, but why would they want to do something other than N-body gravitational problems? _You_ might, but there are a lot of such problems to do, and that's what this is designed for.
--