Questions asked while teaching Windows Vista and Computer Hardware courses
-
@rad131304 said:
@JvdL said:
Pseudophysics
Possibly, but you have not prooven this with your above argument.
It is obvious that my post was in jest and that I'm not an expert in physics but since you have taken it seriously let me back it up:
1: Do you need to add energy to a disk to magnetize it in a particular form? Answer: yes.Does it loose energy when it demagnetizes? Yes. [Reference]
1+2=3: A magnetized disk has more mass than a non-magnetized disk.
Whether a formated disk is "more magnetized" than a non-formated one is debatable but I believe a better informed entropolgist would confirm it.
-
@JvdL said:
I believe a better informed entropolgist would confirm it.
To save everybody the trip to Google:
1. entropist
n. A human being who has no purpose in life but to consume food and energy and produce nothing but excrement and trash. A truely useless human being. A waste of space and air. One who contributes nothing to humanity.
"Some people think that the beggar on the street is the entropist. Really, it's the guy who has no job, lives by himself and spends his time keeping himself amused. He'll walk right past the beggar and into a 7-eleven. He thinks his change is better spent on a week old hotdog."
"Human beings are entropists by nature, but very few truely make life on Earth better."
-
@JvdL said:
It is obvious that my post was in jest
No, it isn't. I missed that, I'll try and remember that next time. (note: I am not being sarcastic)
@JvdL said:
1: Do you need to add energy to a disk to magnetize it in a particular form? Answer: yes.Does it loose energy when it demagnetizes? Yes. [Reference]
Huh? By this logic, moving a rock to the top of a hill will give it more mass than it had at the bottom of the hill. Also, "shorted out magnetic flux" (from the link)?? WTF? How does one "short out" the velocity* of a magnetic field?
@JvdL said:
2: Whenever any type of energy is removed from a system, the mass associated with the energy is also removed, and the system therefore loses mass and vice versa.
This source (which, I incidentally agree with 100%) contradicts your first one; note how Einstein's relativistic equation of which you speak shows no positional information contained within an object's "energy".
@JvdL said:
A magnetized disk has more mass than a non-magnetized disk.
I previously said "not necessarily" to this - after thinking some more, there's just no way this is right. From your second link:
Show me where potential energy exists in Einstein's relativistic energy equation please!
*) it's not a real velocity but there's no simple explanation that I can come up with at the moment
-
@rad131304 said:
Incorrect, you apply a magnetic field - this is not the same as directly adding energy. There is no guarantee that a one state in the disk requires more energy to occupy than a zero state.
No no no... magnets are an energy source... didn't you know that?
[url]http://www.freenergy4u.com/[/url]
(I almost feel bad for spreading that link, but if you go there, you will get whatever you deserve)
-
@rad131304 said:
Show me where potential energy exists in Einstein's relativistic energy equation please!
It doesn't sound like you've quite understood what e=mc^2 (and mass-energy equivalence) means, although I'm no expert. The whole point is that mass and energy are the same thing, so if something has more energy, it has more mass. I note that the answer to your implied question 'does a rock have more mass at the top of a hill than at the bottom?' is actually yes. Similarly, a super-conducting wire loop will have more mass when a current is passed through it than when not. Other than that, a magnetic field definitely does contain energy, so obviously magnetically charged objects have a different mass to uncharged ones.Of course, we're neglecting the contribution the mass of the extra electrons makes when you plug the disk in and turn it on :)
-
I've seen many attempts to explain the 2/3 GB "barrier" and hardly anyone really gets it right.
The real answer is just efficiency. There is no reason, at all, why the kernel could not map all 4 GB of memory into a process address space (let's leave aside dedicated memory regions like register spaces for the moment). The problem is, when a process wants to go to kernel space, the kernel needs somewhere to live inside the VM.
Now, we could dynamically map the kernel in and out every time a process transitioned to/from kernel space. This could be done by twiddling the page tables. But it sucks to do that on every context switch, and who cares if some part of the VM space is reserved for the kernel, because the kernel does indeed EXIST in memory and it has to be mapped SOMEWHERE, eventually. (Another important point is that the context switches are switching to supervisor code which HAS TO BE MAPPED somewhere in the first place, otherwise how would the system trap get a handle on where it is? Some portion of the kernel must always be mapped for this reason)
So, purely for efficiency, the kernel is permanently mapped into every process's address space. This takes usable pages away from the process, no way around it. So the next question is, why on earth does the kernel reserve a huge slice (1 or 2 GB) for itself? Surely the kernel does not need that much.
Well, it used to be for convenience. Machines simply didn't have 4 GB, much less 1 GB, of RAM. So the kernel likes to play a trick where it maps all of physical RAM, in a linear fashion, into its own VM space. This is convenient because now the kernel can access pages through virtual addresses or physical addresses, at will, with no page table trickery.
These days, it's common for the machine to actually have 4 GB of RAM installed in it, so this trick doesn't work -- there aren't enough virtual addresses to linearly map all of physical RAM. And yet kernels still do this.mod: re-added your linebreaks. -dh
-
@alegr said:
All modern processors support PAE, which with introduction of x64 architecture was extended to support even larger memory.
Nearly all - Pentiums M (ok, not modern anymore) and certain Atoms don't support PAE.@alegr said:32-bit Windows supports PAE and can support memory over 4GB. But client SKUs are artificially limited to support only 4GB of physicall address. Thanks to lazy programmers and half-assed devices that can't or don't suport proper PHYSICAL_ADDRESS type. Server SKU (Windows Server 2003) x86 flavor doesn't have that limitation.
IIRC, this restriction was only added with XP SP1 - before that, you could get it to access full 4GB of RAM with /PAE switch (but with SP1 DEP/NX support was added, which requires PAE, and too many drivers didn't work properly, so Microsoft simply removed support for memory mapped above the 4GB barrier).@mallard said:It is not mapped in it's entirety into the system address space, instead, a movable "aperture" of graphics memory (usually only about 128MB) of graphics memory is mapped into the I/O region of the system address space.
AFAIK, the aperture was only used with AGP cards, not with PCIe.@smxlong said:The real answer is just efficiency. There is no reason, at all, why the kernel could not map all 4 GB of memory into a process address space (let's leave aside dedicated memory regions like register spaces for the moment).
ISTR reading that OS X does it this way. Also, on x64 Windows, 32-bit programs actually get access to nearly full 4GB of their virtual address space (assuming they're "large address aware").
-
@ender said:
@mallard said:
It is not mapped in it's entirety into the system address space, instead, a movable "aperture" of graphics memory (usually only about 128MB) of graphics memory is mapped into the I/O region of the system address space.
AFAIK, the aperture was only used with AGP cards, not with PCIeErr yes Graphics memory is mapped to system address space. The aperture is an additional window of system memory allocated to the graphics card.
-
@blakeyrat said:
@JvdL said:
I believe a better informed entropologist would confirm it.
To save everybody the trip to Google:Entropologist
In human thermodynamics, an entropologist is an anthropologist whose studies and theories rest on the prediction of the ultimate thermodynamic leveling of all culture. The term is a derivative of French anthropologist and philosopher Claude Lévi-Strauss' 1960s concept of entropology. In some scenarios, an entropologist is considered as any literary writer who uses the verbal ideas of entropy and disintegration.FTFY
Also, you got it all wrong. Since the earth contains a vast amount of metal, the disk will either be attracted or repelled from the earth, thus making it weigh more or less without changing its mass. This means you should all carry your newly formatted notebooks upside down to lighten the load.
-
@Faxmachinen said:
This means you should all carry your newly formatted notebooks upside down to lighten the load.
Unless you're Australian, obviously.
-
@smxlong said:
I've seen many attempts to explain the 2/3 GB "barrier" and hardly anyone really gets it right.
@smxlong said:So the next question is, why on earth does the kernel reserve a huge slice (1 or 2 GB) for itself? Surely the kernel does not need that much.
You didn't get it right, either. The kernel doesn't reserve "a huge slice" of RAM for itself. The kernel needs address space for the drivers, their memory, to map the PCI devices BARs, etc. The actual necessary size can't be determined upfront.
Linux x86 used to map all physical RAM to kernel space, for convenience. This stopped to work when it needed to support memory sizes over 1 GB.
x64 Windows now maps all physical RAM to its enormous kernel space, just because it can. This is convenient for short-circuiting MmGetSystemAddressForMdl in case of sub-page buffers.
-
@davedavenotdavemaybedave said:
I note that the answer to your implied question 'does a rock have more mass at the top of a hill than at the bottom?' is actually yes.
No, because the increase in energy is only potential, and it's not actually contained within the rock itself. It's like asking if a charged battery weighs more than the discharged one. A rock at the top of a hill actually weighs less than the same rock at the bottom of the hill, and it has the same mass all the time. The energy comes from gravity, and it's not contained within the rock itself. So, you didn't add any mass/energy to the rock... that is what trips people up about magnets. Magnets don't supply energy for the same reason why the power company isn't rolling rocks up hills all day.
-
@smxlong said:
I've seen many attempts to explain the 2/3 GB "barrier" and hardly anyone really gets it right.
http://blogs.msdn.com/b/oldnewthing/archive/2004/08/22/218527.aspx is a pretty good series of articles describing PAE, /3gb and sundry related things.
-
@jasmine2501 said:
No, because the increase in energy is only potential, and it's not actually contained within the rock itself.
No, yes, that's a meaningless concept in this instance.@jasmine2501 said:
It's like asking if a charged battery weighs more than the discharged one.
It does. Counter-intuitive, perhaps, but absolutely true. It really is very simple: E=mc^2, where E is the total energy contained in a system, implies that unless the speed of light varies - it doesn't - then any variation in the total energy contained in a system must result in a variation in its mass.@jasmine2501 said:
A rock at the top of a hill actually weighs less than the same rock at the bottom of the hill, and it has the same mass all the time.
We're talking about mass, not weight. And, once again, no, the mass changes. Not measurably, but it changes.@jasmine2501 said:
The energy comes from gravity, and it's not contained within the rock itself. So, you didn't add any mass/energy to the rock
No again. The energy (work done) comes from you moving the rock to the top of the hill. Hmmn. Thinking about it a bit more, I'm not sure if it would be the earth or the rock which gains the energy, or even if you can meaningfully separate them in this context. The rock-planet system has gained energy (and therefore mass), assuming the work done came from outside.@jasmine2501 said:
Magnets don't supply energy for the same reason why the power company isn't rolling rocks up hills all day.
I don't know quite what that's supposed to mean. Magnets don't supply energy any more than lead-acid cells do, but a magnetic field contains energy in just the same way as an electric field. When you bring two opposed magnets together, you put energy into their magnetic fields as you do work, and that energy then comes out again as the magnets do the work of pushing each other apart when you let go.
-
@jasmine2501 said:
@davedavenotdavemaybedave said:
In order to lift the rock, you have to push against the earth. By conservation of momentum, the earth will be pushed away from the rock as much as the rock is pushed away from the earth (not the same distance, rather the rock will go up a lot and the earth will go down a tiny bit due to the difference in masses). The inertial reference frame of the earth-rock system will gain mass. The non-inertial reference frame of the ground will see the rock as gaining no mass.I note that the answer to your implied question 'does a rock have more mass at the top of a hill than at the bottom?' is actually yes.
No, because the increase in energy is only potential, and it's not actually contained within the rock itself. It's like asking if a charged battery weighs more than the discharged one. A rock at the top of a hill actually weighs less than the same rock at the bottom of the hill, and it has the same mass all the time. The energy comes from gravity, and it's not contained within the rock itself. So, you didn't add any mass/energy to the rock... that is what trips people up about magnets. Magnets don't supply energy for the same reason why the power company isn't rolling rocks up hills all day.The answer to the original question of whether the hard drive gains mass is "it depends on the observer's reference frame". If the observer is only considering the hard drive, than it will experience no change in relativistic mass when formatted. However, if the hard drive is considered along with the device that formats it, the system will likely gain mass, but the mass change due to magnetizing the platter will be insignificant compared to the mass gained from heated up the drive by operating it.
-
@Jaime said:
However, if the hard drive is considered along with the device that formats it, the system will likely gain mass
From an information theory point of view, it might in fact lose mass. Indeed, a formatted disk contains almost no information (only a few bits), and so it will have higher entropy, therefore lower internal energy, therefore lower mass.
-
@vic said:
@Jaime said:
However, if the hard drive is considered along with the device that formats it, the system will likely gain mass
From an information theory point of view, it might in fact lose mass. Indeed, a formatted disk contains almost no information (only a few bits), and so it will have higher entropy, therefore lower internal energy, therefore lower mass.
You overlooked the minus sign. A string of identical bits (for example, a disk with all bits zero) has the lowest possible information entropy. A completely random sequence of bits has the highest possible information entropy. Reference
This is similar to physical entropy. Other than that and some mathematical analogies, information entropy cannot be used to do energy book keeping, so your argument is not applicable to the mass discussion anyway.
-
@JvdL said:
This is similar to physical entropy. Other than that and some mathematical analogies, information entropy cannot be used to do energy book keeping, so your argument is not applicable to the mass discussion anyway.
Yah, I don't think any of this is applicable to the discussion.
Now let's spend 26 posts debating over the guy who thought the CD tray was a cup holder! Har har!
-
Think about what that would mean though... since the universe has the potential to collapse on itself, and as it expands, the potential energy of such a collapse is greater - which would mean the universe is constantly gaining mass, and that is impossible.
I don't buy the rock has more mass... ever. You can push it around all day and it's still the same rock, same mass. The hard drive doesn't change mass either, and you can move the little charges around all you want - they are still there. Same as a battery - you're just moving things around.
Still, I think it is an intelligent question - since the hard drive starts with "nothing" on it, and then you put "something" on it, it's reasonable to think it might increase in mass. Here we are actually discussing it - proving that it's a smart question which is not easy to answer.
-
@jasmine2501 said:
Here we are actually discussing it - proving that it's a smart question which is not easy to answer.
No it proves you have way too much free time. Get a hobby.
-
@jasmine2501 said:
Think about what that would mean though... since the universe has the potential to collapse on itself, and as it expands, the potential energy of such a collapse is greater - which would mean the universe is constantly gaining mass, and that is impossible.
No, it's slowing down - that is, losing kinetic energy - as it does so.@jasmine2501 said:
I don't buy the rock has more mass... ever. You can push it around all day and it's still the same rock, same mass.
You can 'not buy' it all you like, it doesn't suddenly disprove Einstein's theorems. Your argument basically amounts to 'It's ridiculous to suggest that mass can be gained and lost as energy, so I don't believe it'.
-
You didn't get it right, either. The kernel doesn't reserve "a huge slice" of RAM for itself.
I meant a slice of address space, not RAM. I didn't say RAM anywhere.
The approach of, "We don't know how much we'll need so let's just reserve a shitload of it", regardless of what "it" is, is a bad strategy in almost all cases and yet most kernels do it this way. I think it's weird, that's all.
-
@jasmine2501 said:
I don't buy the rock has more mass... ever.
Relativity doesn't require your buy-in. Here's why things with less energy have to decrease mass:A device is moving in one direction rapidly and emits an equal number of photon both in front of and behind it. Those photons in front of it blue-shift and those behind it red-shift. If all of the photons split into electron/positron pairs, there will be more mass in front of it than behind it. Mass has magically been transmitted forwards. However, if the mass loss of the device due to the energy emitted as photons is taken into account, the system balances out.
-
-
@davedavenotdavemaybedave said:
It really is very simple: E=mc^2, where E is the total energy contained in a system, implies that unless the speed of light in a vacuum varies - it doesn't
For counter-examples of the uncorrected version, consider a straw in a glass of water, or a rainbow.
