it strikes me that if you have to ask, there probably isn't a compelling story to tell. why do you feel a need to tell one?
the best IT is an IT that no one is aware of. IT *should* be an utility. if it's not, you're not doing your job.
typical density of seawater, from wikipedia, is rho_s ~ 1025 kg/m^3. the density of fresh water is rho_w = 1000 kg/m^3. the density of frozen fresh water (ice) is rho_i = 917 kg/m^3 at 0C.
take any typical floating iceberg made out of fresh water, say it's volume is V_i. so, the iceberg's mass is rho_i V_i and it's weight is g rho_i V_i, where g is the acceleration from gravity (9.8m/s^2).
iceberg floats because it's weight can be offset by displacing a less volume of seawater. in fact, the volume of seawater displaced, V_s, is given by g rho_i V_i = g rho_s V_s, or simply V_s = rho_i V_i / rho_s = (917 / 1025) V_i.
when the iceberg completely melts, its mass (not it's volume) is conserved, so the volume of the melted iceberg, V_w, is given by rho_w V_w = rho_i V_i, or simply V_w = rho_i V_i / rho_w = (917 / 1000) V_i.
we see that while the volume of the melted iceberg is less than the volume of the iceberg, it's more than the originally displaced seawater needed to float it. in fact, the extra volume (which is directly proportional to the change of sea level) is given by rho_i / rho_w - rho_i / rho_s = rho_i ( rho_s - rho_w) / (rho_s rho_w) per unit of volume iceberg, or simply 0.0224 V_i.
we also note that if the iceberg is actually frozen seawater, then regardless of the density of the frozen seawater, rho_w is replaced by the rho_s and hence is no change in sealevel.
ice resting on land (not floating) an additional volume to the sea is given directly by (rho_i / rho_w) V_i = (917 / 1000) V_i = 0.917 V_i, where V_i is, again, the volume of ice melted (though not an iceberg this time)
assuming you do have an apple, final cut & dvd studio- it's pretty straight forward. final cut will import directly, compressor will encode and dvd studio will author the dvd. i had no idea what i was doing and was able to do it in a day.
do as few encoding/decodings as possible and burn the disk slow. make sure your (total) bit rate is below the dvd spec for stand-alone players. at least that's what i figured out...
when i got my phd in physics 12 years ago (sigh)- i felt the same way. perhaps being a physicist, attaching 'applied' in front of it rang as a 'sell-out'. so, off i went to a post-doc at LLNL. that environment was very similar to the university other than once every year, people ran around, formed new groups and tried to get funding. no one really planned for longer than a year, particularly when elections were close. but, really, it was easy to find outstanding people (a big plus in my book) and as a post-doc 15% of your time you're allowed to do whatever you damn well pleased. NIST is similar, better if you have problems w/ making weapons.
after some consulting jobs for private industries, universities and various government labs, i ended up at an industrial research lab. while the work was applied, the focus was 6-10 years to product. in some sense, a dream job. i hated it. some because i found the work boring and mostly because my peers, in general, sucked. if i'm the brightest one in the room, there is a problem. management had no vision nor could they understand the implications of the research. but, the research was quite basic.
for the past couple years i've been w/ GE global research (upstate NY). yes, it's applied work and yes you are expected to deliver- on time and accurate results. there are projects here which are 'basic research' since it appears the business knows that if they miss out, it could hurt them. most of the work is directly product related and are typically year long duration. at the same time perhaps only 20% of the work done here get into a product. management isn't concerned- they are more concerned that the science is done correctly so that the correct business decision can be derived. i get to work w/ a variety of people (other physicists, chemists, material scientists, various engineers,...) and, in general, good highly-motivated people. another plus, for the 2.5 years here, i've worked on 5 different project- i get to constantly learn. not just new physics, but how it fits into the real world. this year one of my projects is product driven, but the research is so bloody basic that there is nothing in the literature to help. if this project works (it will go on for at least 2 more years), i'll be helping to change the world for the better, slightly (yeah, sure, GE gets to make money too).
really, 'applied' and 'pure' doesn't mean much. it's the people you work with (at least in my case). don't go to a place where they only do one type of research (regardless of applied or pure)- you may end up hating it and then what?
under 10.4 (and even 10.3), i can run large multiple fp calculations (~3.5 GB data size) on a dual g5 2.0GHz pegging both cpus at 100% and still do anything else w/o seeing any performance hit for interactive use. it's much better than the equivalent situation on w2k...
i may be a perfect example of something, but not what you claim.
which part of of the natural numbers don't i know? it's clear i know every part i care to think about, therefore i must know it all. remarkable- a finite being can understand something infinite.
the 'largest number' is a non-question since i can prove by induction no such number exists: say A is the largest number. A+1 is larger than A, therefore A can not exist. are you claiming i don't understand that proof?
i can prove many things about infinity. i also know there are different types of infinity. E.g. the number of real numbers and the number of natural numbers are both infinite, but there are more real numbers than natural numbers. would you like a proof?
just because you claim something is true doesn't make it true, ya know.
"If God is infinite, and we are finite, it is impossible to fully comprehend Him."
what?
the set of natural numbers (1, 2, 3,...) is infinite. i'm finite. therefore i can not comprehend natural numbers. uhm, so we shouldn't teach mathematics, right?
When I talk to "scientists" (I'm not one either), and even hint that Einstein's theory might be wrong, it's as if I've shouted out a stream of profanities at church.
it's not because they thought you shouted profanities in church, it's because you're showing remarkable lack of understanding of the scientific method and how science is done. if you want to be taken serious, propose your model, show how it explains an observable fact that current theories don't.
being a physicist, i can't believe your post. do you really believe what you say? more importantly, does anyone else? not only is your 'discourse' lacking in supporting evidence, but it's clear you don't know how science is conducted.
one does not simply accept something that's published because it's published. yes, it makes a difference what journal, but that is what references are for- a paper which doesn't give enough details to be reproduced is useless.
have you heard of tenure? it's very hard to get fired if a professor has tenure regardless if he puts forth crackpot theories. to suggest scientists are priests and are afraid of dogma is unfounded and completely false. it's hard to disprove a standard theory simply because the standard theories are the cumulative sum of all of mankind understanding.
physics always consider alternatives. how else do you advance understanding? it's the willowing of all these alternatives that have given us relativity, quantum mechanics, statistical mechanics, electromagnetism, etc.
really, the natural sciences are not a religion. it has to do something, everytime. "objectivity" not "clinging to tradition" is what insures your computer works, electricity is on, gets a probe to Mars, etc.
obviously this is not "the end" of physics- quantum mechanics and general relativity need to combined in a consistent theory. perhaps strings, m-theory, etc will be it.
to challenge relativity (special or general) or quantum mechanics is simply very very hard. all experimental data that has been done and confirmed agree w/ the theory. more importantly, there is no data that directly contradicts the theories. therefore, to propose a new theory, it not only has to predict the same as the old theories (so it agrees w/ experimental facts), but it must also predict something in disagreement w/ the old theories for it to be tested.
to "mention challenging theories" is fine, but show me what's different. otherwise, yeah, you're a flake. show me and let's do the experiment- we'll be set for life!
yes, i'm a physicist. yes, the b.s. meter goes off when people say current theories are wrong but don't have any way to test that hypothesis.
pure Al resistivity is 24.2 nOhm m = 2.42e-8 Ohm m. if the inside of the cylinder is a and the outside is b and the length is L then the resistance of the cylinder is R = rho L / A = 2.42e-8 L / [ pi (b^2 - a^2) ].
but, it's wrong. by using Ampere's law, the magnetic field generated by a dc current of 19 million amp is B = mu_0 I / 2 pi b and if we use b=0.04445m (i.e. 3.5 inch diameter) B = 85.5T. if memory serves, there will be a force on the current (i.e. Al cylinder) goes as F = integral over space J x B = mu_0 I^2 L (b^3/3-a^2 b + 2 a^3 b/3)/[pi (b^2 - a^2)^2] > 1MNewton directed towards the center of the cylinder.
that is, the cylinder will not there anymore- it will collapse
if you dissamble a '3d' drive, it will look the same as the current '2d' drive. you still have platters which have magnetic storage layer. the platter still spins and the heads (read and write) fly over it.
in '2d' drives, which is really called longitudinal recording, the magnetic material have their poles in the plane of the platter. in '3d' drives, aka perpendicular recording, the poles are pointing out of the plane.
contrary to what i've read here, having the poles still out of the plane doesn't save space- each bit is a large collection of poles.
the problem w/ longitudinal recording is as you place the bits closer together, the interaction between the bits tend to want them to spontaneously align with each other. you know this. go to your frig, grab some magnets. if you align them properly, you'll see that you can't push them together. or if you align them properly, they stick together. the way the drive makers have over come this problem is to make the magnetic storage layer become very 'hard' magnetically. that is, as you pack the bits together, the crystal structure won't allow neighboring bits which have opposite polarity to spontaneously switch.
there's a problem w/ pushing that two far. if you make it too hard (so you won't have spontaenous decay- i.e. data loss) you'll have problems generating a powerful enough magnetic field to actually write the information. ok, it's easy to generate a powerful magnetic field, but it's hard to confine it to a very small region.
so, perpendicular comes along. now perpendicular has the opposite characteristic. it likes having bits of opposite polarity next to each other. it does NOT like having bits of same polarity next to each other. also in perpendicular recording, the recording media usually has a soft magnetic material below the storage layer which helps to promote the write field strength and focus it (details are involved and are left as an excersize).
i spent (wasted?) three years of my life on modeling perpendicular recording. it turns out that this soft magnetic layer creates more problems than any one expected. in fact, if you could get the hard drive to write most of the poles to be in one direction, you'll find the hard drive will fail. essentially the read head will have have to cope w/ large dc shift due to all these fields coming from this magnetized layer. it also turns out that at high densities, the soft magnetic material doesn't actually help you to do writing since it forces the write field to be perpendicular to the platter and so while you have a strong field, there is very little torque acting on the magnetic spins to actually flip them.
from calculations i did, and from literature, it doesn't appear that perpendicular will buy you a whole lot. perhaps it can get you to 500 Gb/in^2 (about 4x greater than today's drive) if you can find a reader that can handle the dc shift (or you avoid the dc shift by encoding the data intelligently so you don't get large regions of same polarity).
as for laser assisted magnetic recording- not only do you have the same problem as perpendicular (most of these schemes are only for perpendicular recording), but now you have a laser so you have to handle the thermal time scales- you have to heat it fast enough to write, but also cool it down fast enough so you can keep the information. all the problems of perpendicular plus a nice thermal conduction/convection problem. ha!
IBM's millepede is fun- but data rates still appear to be too low.
what's important here is that if the industry does shift to perpendicular, there will be fewer players. it's not clear if Maxtor or Western Digital have enough money to spend on upgrading their fabs. you might be left w/ only Seagate and the Japan as makers of disc drives...
launched QM? i think Planck would disagree. he was the one who first postulated that photons came in discrete (i.e. quantum) frequencies. he didn't say why, but it did explain experimental facts and the clear failure of Maxwell's equations. what Einstein did was to accept the quantization and showed it explained the photoelectric effect. suddenly Planck's conjecture seemed to be more than just conjecture.
as for cosmology, yes, he replaced Newtons framework with relativity. but relativity was already contained within Maxwell's equations. he was the one that saw that. once that is explained, the rest 'falls out' (excuse the pun). once he had special relativity (which deals w/ reference frames not accelerating- i.e. inertial), it was clear there had to be general relativity (i.e. non-inertial reference frames). again, his great insight was that acceleration and gravitation was the same thing (something that every freshmen in physic courses assume when they write down F=ma and then say F=G m M/r^2- there is no reason to assume the two "m"'s are the same).
Einstein didn't show that the 'basic assumptions' were wrong. he gave a different interpretation of what the theories were saying. He didn't throw away anything. you can derive Netwon from relativity by going to the correct limits.
once again, i'm not dishing Einstein in any respect. his insights were absolutely fabulous. at the same time, if it wasn't him, it would have been another (or others).
when did i claim all physicists contributed the same? most physicists suck. really, they do.
while i might agree that he laid the ground work for strings (or whatever the 'final' theory might be) through the ideas of invariants, it's not like these things hadn't/won't have been discovered. for example, Newton invented calculus, but Leibniz was able to do it 20 years later, independently. science, especially physics, does this all the time- two (or more) independent teams find the same answer. that is, new ideas come from the work previous to it.
you must be joking about who could close to Einstein. take Maxwell- he took the work of others, postulated his own equation and was able to write down the classical equations of electromagnetism. the equations which really has the most important effect on the macroscopic scale. in fact, without his equations (which contain special relativity in), Einstein wouldn't have a leg to stand on. what about Boltzman and the formulation of the three laws of statistical mechanics. these two theories combined have more direct effect on your life and it's quality than anything Einstein ever did.
oh, you want contemporary? Dirac was able to predict the existence of anti-matter long before anyone could create/measure it. his equation (which included quantum mechanics and relativity) is what really describes the energy levels of atoms. as for cutting edge/breadth of work, Feynman was remarkable- nuclear, computation, statical mechanics (pick up is book on that- it's gorgeous), Feynman diagrams, path integrals, etc.. the quantity and quality rivals, if not surpasses Einstein.
it's precisely the statement " no one has been at the cutting edge of physics for so long or over such a wide range of topics" which drive physicists to moan. look, he was great. there are others who were/are just as great. it's because the media had to explain the atomic bomb and E=mc^2 did that (sort of) and therefore Einstein became more.
'most important physicist since Newton'. says who? pbs?
the fact remains, Einstein didn't 'invent' new mathematics as Newton did. Lorentz had the equations for special relativity. Plank had described black body radiation. Einstein needed help in writing down general relativity equations. His work on Brownian motion was built upon Boltzman and statistics.
his genius was to see a different way of explaining these things. time was not invariant was not obvious. that photons are quantized (i.e. particle-like) flew in the face of Maxwell's equations.
i'm not trying to take away anything from Einstein. it's just that the popular conception of him being 'the opera singer super model brain surgeion sports star' of physics is insulting to all the other great minds. he could not have done what he did without others.
it's not that physicists hate Einstein, it's more they hate how people view Einstein. it's mostly because Einstein became the poster-boy for the media about modern physics. as we know, the media tends to simplify things and so it suggests that Einstein did relativity, photoelectric effect, etc. by himself.
given that Newton said he had seen far only because he stood on the shoulders of giants, Einstein is even more indebted to others before and during his time.
look, i'm not saying that he wasn't a remarkable physicist- when i read some of his papers, i do feel like i'm reading something that's very close to 'god'- so clear, so elegant, so beautiful.
yes, i am a physicist. and yes, i do find myself using Einstein's results. i'm glad when it happens, because i *feel* i'm close to reality.
it's b.s. current high end disk drives are either at or above 100Gb/in^2. they had plans to push to 150Gb/in^2 w/ the current technology (called longitudinal recording). after that it's not clear it will work. essentially the magnetic bits (a collection of grains whose magnetic moments lie in roughly the same direction) will become unstable (spontaneous data loss).
one way to get around this (forget about IBM's pixie dust) is to do perpedicular recording. there you can pack muck higher linear densities, but you still have the problem of producing a very large magnetic field (to write) in a very small area (to keep the density you need). oh, and it gets harder to read smaller bits, of course.
i don't know where they get the data rate of kiloHz. that's b.s. too- high end drives run in the 100MHz. stuff on the drawing boards are pushing 1GHz (which, of course is a problem since magnetic media is limited by something called the gyromagnetic frequency).
yes, SOFC's do run at high temperature, so to convert the fuel into hydrogren is 'free'. it's all chemically driven.
as for producing carbon dioxide, CO2, well, that's not a problem, right? i mean, that's what i and you produce when we breathe and that's what plants use to create O2. it's much better than producing carbon monoxide, CO- since that's a greenhouse gas.
also, since SOFC's do run hot- the H2O you get out is steam which you can recover to run a turbine (like typical power plants do- burn coal or oil to produce steam to turn a turbine which then produces electricity). so SOFCs can be very efficient. more efficient than current technology.
be careful, fuel cells can just store energy, or they can produce it. there are mutiple types of such things for different applications.
the major problem is, yes, cost, but also lifetime. you need these things to survive for 40,000 hours to be feasiable. that's not trivial and turns into a material science problem. it's not impossible, but currently we don't know how. that is why i go to work;)
don't understand why everyone is so down about fuel cells.
yes, pure hydrogen is hard/expensive to produce. but the next generation of fuel cells can use methane (or ethanol) for a source of fuel. ie, plug the fuel cell into the back end of a cow- suddenly wisconsin will be known for more than it's cheese.
for some reason, some are thinking fuel cells are going to replace gasoline engines in vechicles. well, ok. but what you really want to do is replace all the coal and oil burning power plants w/ fuel cells. so instead of acid rain and tons of greenhouse gases, you get H2O out, which you could use to water crops or drink. given that China seems to be building coal burning power plants as fast as they can, doesn't that sound like a good idea?
ok, fine, i might be biased. i am working on the next design of fuel cells (in particular solid oxide fuel cells- SOFC). but, still, the sooner we get to a place where producing energy is less harmful to the planet, i think we should. hell, we must.
albeit, my experiences w/ WD have all been older than 3 years (two cheapo WD died within two weeks after six months. and before,...). if it's something you don't want to lose and you don't have copies, you're the only one to blame.
Slashdot Mirror
Celsius, Fahrenheit, Kelvin or Rankine?
https://spectrum.ieee.org/ener...
for a white paper, it's pretty weak. no data. not even any theory. mostly just wishful thinking.
it strikes me that if you have to ask, there probably isn't a compelling story to tell. why do you feel a need to tell one? the best IT is an IT that no one is aware of. IT *should* be an utility. if it's not, you're not doing your job.
typical density of seawater, from wikipedia, is rho_s ~ 1025 kg/m^3. the density of fresh water is rho_w = 1000 kg/m^3. the density of frozen fresh water (ice) is rho_i = 917 kg/m^3 at 0C.
take any typical floating iceberg made out of fresh water, say it's volume is V_i. so, the iceberg's mass is rho_i V_i and it's weight is g rho_i V_i, where g is the acceleration from gravity (9.8m/s^2).
iceberg floats because it's weight can be offset by displacing a less volume of seawater. in fact, the volume of seawater displaced, V_s, is given by g rho_i V_i = g rho_s V_s, or simply V_s = rho_i V_i / rho_s = (917 / 1025) V_i.
when the iceberg completely melts, its mass (not it's volume) is conserved, so the volume of the melted iceberg, V_w, is given by rho_w V_w = rho_i V_i, or simply V_w = rho_i V_i / rho_w = (917 / 1000) V_i.
we see that while the volume of the melted iceberg is less than the volume of the iceberg, it's more than the originally displaced seawater needed to float it. in fact, the extra volume (which is directly proportional to the change of sea level) is given by rho_i / rho_w - rho_i / rho_s = rho_i ( rho_s - rho_w) / (rho_s rho_w) per unit of volume iceberg, or simply 0.0224 V_i.
we also note that if the iceberg is actually frozen seawater, then regardless of the density of the frozen seawater, rho_w is replaced by the rho_s and hence is no change in sealevel.
ice resting on land (not floating) an additional volume to the sea is given directly by (rho_i / rho_w) V_i = (917 / 1000) V_i = 0.917 V_i, where V_i is, again, the volume of ice melted (though not an iceberg this time)
assuming you do have an apple, final cut & dvd studio- it's pretty straight forward. final cut will import directly, compressor will encode and dvd studio will author the dvd. i had no idea what i was doing and was able to do it in a day.
do as few encoding/decodings as possible and burn the disk slow. make sure your (total) bit rate is below the dvd spec for stand-alone players. at least that's what i figured out...
when i got my phd in physics 12 years ago (sigh)- i felt the same way. perhaps being a physicist, attaching 'applied' in front of it rang as a 'sell-out'. so, off i went to a post-doc at LLNL. that environment was very similar to the university other than once every year, people ran around, formed new groups and tried to get funding. no one really planned for longer than a year, particularly when elections were close. but, really, it was easy to find outstanding people (a big plus in my book) and as a post-doc 15% of your time you're allowed to do whatever you damn well pleased. NIST is similar, better if you have problems w/ making weapons.
...) and, in general, good highly-motivated people. another plus, for the 2.5 years here, i've worked on 5 different project- i get to constantly learn. not just new physics, but how it fits into the real world. this year one of my projects is product driven, but the research is so bloody basic that there is nothing in the literature to help. if this project works (it will go on for at least 2 more years), i'll be helping to change the world for the better, slightly (yeah, sure, GE gets to make money too).
after some consulting jobs for private industries, universities and various government labs, i ended up at an industrial research lab. while the work was applied, the focus was 6-10 years to product. in some sense, a dream job. i hated it. some because i found the work boring and mostly because my peers, in general, sucked. if i'm the brightest one in the room, there is a problem. management had no vision nor could they understand the implications of the research. but, the research was quite basic.
for the past couple years i've been w/ GE global research (upstate NY). yes, it's applied work and yes you are expected to deliver- on time and accurate results. there are projects here which are 'basic research' since it appears the business knows that if they miss out, it could hurt them. most of the work is directly product related and are typically year long duration. at the same time perhaps only 20% of the work done here get into a product. management isn't concerned- they are more concerned that the science is done correctly so that the correct business decision can be derived. i get to work w/ a variety of people (other physicists, chemists, material scientists, various engineers,
really, 'applied' and 'pure' doesn't mean much. it's the people you work with (at least in my case). don't go to a place where they only do one type of research (regardless of applied or pure)- you may end up hating it and then what?
under 10.4 (and even 10.3), i can run large multiple fp calculations (~3.5 GB data size) on a dual g5 2.0GHz pegging both cpus at 100% and still do anything else w/o seeing any performance hit for interactive use. it's much better than the equivalent situation on w2k...
i prefer the green Kool-aid, please.
ok, it's clear we're not playing w/ logic here.
hold whatever beliefs you want. i'll stick to logic.
i may be a perfect example of something, but not what you claim.
which part of of the natural numbers don't i know? it's clear i know every part i care to think about, therefore i must know it all. remarkable- a finite being can understand something infinite.
the 'largest number' is a non-question since i can prove by induction no such number exists: say A is the largest number. A+1 is larger than A, therefore A can not exist. are you claiming i don't understand that proof?
i can prove many things about infinity. i also know there are different types of infinity. E.g. the number of real numbers and the number of natural numbers are both infinite, but there are more real numbers than natural numbers. would you like a proof?
just because you claim something is true doesn't make it true, ya know.
"If God is infinite, and we are finite, it is impossible to fully comprehend Him."
...) is infinite. i'm finite. therefore i can not comprehend natural numbers. uhm, so we shouldn't teach mathematics, right?
what?
the set of natural numbers (1, 2, 3,
sigh.
ramjet has lifetime replacment- never a problem on g5 systems. did have problems w/ crucial's 1GB kit (multiple times).
When I talk to "scientists" (I'm not one either), and even hint that Einstein's theory might be wrong, it's as if I've shouted out a stream of profanities at church.
it's not because they thought you shouted profanities in church, it's because you're showing remarkable lack of understanding of the scientific method and how science is done. if you want to be taken serious, propose your model, show how it explains an observable fact that current theories don't.
being a physicist, i can't believe your post. do you really believe what you say? more importantly, does anyone else? not only is your 'discourse' lacking in supporting evidence, but it's clear you don't know how science is conducted.
one does not simply accept something that's published because it's published. yes, it makes a difference what journal, but that is what references are for- a paper which doesn't give enough details to be reproduced is useless.
have you heard of tenure? it's very hard to get fired if a professor has tenure regardless if he puts forth crackpot theories. to suggest scientists are priests and are afraid of dogma is unfounded and completely false. it's hard to disprove a standard theory simply because the standard theories are the cumulative sum of all of mankind understanding.
physics always consider alternatives. how else do you advance understanding? it's the willowing of all these alternatives that have given us relativity, quantum mechanics, statistical mechanics, electromagnetism, etc.
really, the natural sciences are not a religion. it has to do something, everytime. "objectivity" not "clinging to tradition" is what insures your computer works, electricity is on, gets a probe to Mars, etc.
i do hope you are simply a troll...
obviously this is not "the end" of physics- quantum mechanics and general relativity need to combined in a consistent theory. perhaps strings, m-theory, etc will be it.
to challenge relativity (special or general) or quantum mechanics is simply very very hard. all experimental data that has been done and confirmed agree w/ the theory. more importantly, there is no data that directly contradicts the theories. therefore, to propose a new theory, it not only has to predict the same as the old theories (so it agrees w/ experimental facts), but it must also predict something in disagreement w/ the old theories for it to be tested.
to "mention challenging theories" is fine, but show me what's different. otherwise, yeah, you're a flake. show me and let's do the experiment- we'll be set for life!
yes, i'm a physicist. yes, the b.s. meter goes off when people say current theories are wrong but don't have any way to test that hypothesis.
sigh.
pure Al resistivity is 24.2 nOhm m = 2.42e-8 Ohm m. if the inside of the cylinder is a and the outside is b and the length is L then the resistance of the cylinder is R = rho L / A = 2.42e-8 L / [ pi (b^2 - a^2) ].
but, it's wrong. by using Ampere's law, the magnetic field generated by a dc current of 19 million amp is B = mu_0 I / 2 pi b and if we use b=0.04445m (i.e. 3.5 inch diameter) B = 85.5T. if memory serves, there will be a force on the current (i.e. Al cylinder) goes as F = integral over space J x B = mu_0 I^2 L (b^3/3-a^2 b + 2 a^3 b/3)/[pi (b^2 - a^2)^2] > 1MNewton directed towards the center of the cylinder.
that is, the cylinder will not there anymore- it will collapse
if you dissamble a '3d' drive, it will look the same as the current '2d' drive. you still have platters which have magnetic storage layer. the platter still spins and the heads (read and write) fly over it.
in '2d' drives, which is really called longitudinal recording, the magnetic material have their poles in the plane of the platter. in '3d' drives, aka perpendicular recording, the poles are pointing out of the plane.
contrary to what i've read here, having the poles still out of the plane doesn't save space- each bit is a large collection of poles.
the problem w/ longitudinal recording is as you place the bits closer together, the interaction between the bits tend to want them to spontaneously align with each other. you know this. go to your frig, grab some magnets. if you align them properly, you'll see that you can't push them together. or if you align them properly, they stick together. the way the drive makers have over come this problem is to make the magnetic storage layer become very 'hard' magnetically. that is, as you pack the bits together, the crystal structure won't allow neighboring bits which have opposite polarity to spontaneously switch.
there's a problem w/ pushing that two far. if you make it too hard (so you won't have spontaenous decay- i.e. data loss) you'll have problems generating a powerful enough magnetic field to actually write the information. ok, it's easy to generate a powerful magnetic field, but it's hard to confine it to a very small region.
so, perpendicular comes along. now perpendicular has the opposite characteristic. it likes having bits of opposite polarity next to each other. it does NOT like having bits of same polarity next to each other. also in perpendicular recording, the recording media usually has a soft magnetic material below the storage layer which helps to promote the write field strength and focus it (details are involved and are left as an excersize).
i spent (wasted?) three years of my life on modeling perpendicular recording. it turns out that this soft magnetic layer creates more problems than any one expected. in fact, if you could get the hard drive to write most of the poles to be in one direction, you'll find the hard drive will fail. essentially the read head will have have to cope w/ large dc shift due to all these fields coming from this magnetized layer. it also turns out that at high densities, the soft magnetic material doesn't actually help you to do writing since it forces the write field to be perpendicular to the platter and so while you have a strong field, there is very little torque acting on the magnetic spins to actually flip them.
from calculations i did, and from literature, it doesn't appear that perpendicular will buy you a whole lot. perhaps it can get you to 500 Gb/in^2 (about 4x greater than today's drive) if you can find a reader that can handle the dc shift (or you avoid the dc shift by encoding the data intelligently so you don't get large regions of same polarity).
as for laser assisted magnetic recording- not only do you have the same problem as perpendicular (most of these schemes are only for perpendicular recording), but now you have a laser so you have to handle the thermal time scales- you have to heat it fast enough to write, but also cool it down fast enough so you can keep the information. all the problems of perpendicular plus a nice thermal conduction/convection problem. ha!
IBM's millepede is fun- but data rates still appear to be too low.
what's important here is that if the industry does shift to perpendicular, there will be fewer players. it's not clear if Maxtor or Western Digital have enough money to spend on upgrading their fabs. you might be left w/ only Seagate and the Japan as makers of disc drives...
launched QM? i think Planck would disagree. he was the one who first postulated that photons came in discrete (i.e. quantum) frequencies. he didn't say why, but it did explain experimental facts and the clear failure of Maxwell's equations. what Einstein did was to accept the quantization and showed it explained the photoelectric effect. suddenly Planck's conjecture seemed to be more than just conjecture.
/r^2- there is no reason to assume the two "m"'s are the same).
as for cosmology, yes, he replaced Newtons framework with relativity. but relativity was already contained within Maxwell's equations. he was the one that saw that. once that is explained, the rest 'falls out' (excuse the pun). once he had special relativity (which deals w/ reference frames not accelerating- i.e. inertial), it was clear there had to be general relativity (i.e. non-inertial reference frames). again, his great insight was that acceleration and gravitation was the same thing (something that every freshmen in physic courses assume when they write down F=ma and then say F=G m M
Einstein didn't show that the 'basic assumptions' were wrong. he gave a different interpretation of what the theories were saying. He didn't throw away anything. you can derive Netwon from relativity by going to the correct limits.
once again, i'm not dishing Einstein in any respect. his insights were absolutely fabulous. at the same time, if it wasn't him, it would have been another (or others).
i did check out the article...
when did i claim all physicists contributed the same? most physicists suck. really, they do.
while i might agree that he laid the ground work for strings (or whatever the 'final' theory might be) through the ideas of invariants, it's not like these things hadn't/won't have been discovered. for example, Newton invented calculus, but Leibniz was able to do it 20 years later, independently. science, especially physics, does this all the time- two (or more) independent teams find the same answer. that is, new ideas come from the work previous to it.
you must be joking about who could close to Einstein. take Maxwell- he took the work of others, postulated his own equation and was able to write down the classical equations of electromagnetism. the equations which really has the most important effect on the macroscopic scale. in fact, without his equations (which contain special relativity in), Einstein wouldn't have a leg to stand on. what about Boltzman and the formulation of the three laws of statistical mechanics. these two theories combined have more direct effect on your life and it's quality than anything Einstein ever did.
oh, you want contemporary? Dirac was able to predict the existence of anti-matter long before anyone could create/measure it. his equation (which included quantum mechanics and relativity) is what really describes the energy levels of atoms. as for cutting edge/breadth of work, Feynman was remarkable- nuclear, computation, statical mechanics (pick up is book on that- it's gorgeous), Feynman diagrams, path integrals, etc.. the quantity and quality rivals, if not surpasses Einstein.
it's precisely the statement " no one has been at the cutting edge of physics for so long or over such a wide range of topics" which drive physicists to moan. look, he was great. there are others who were/are just as great. it's because the media had to explain the atomic bomb and E=mc^2 did that (sort of) and therefore Einstein became more.
it's the frequency that's quantized. energy is simply h v in your notation. you're over counting n (the number of modes).
'most important physicist since Newton'. says who? pbs?
the fact remains, Einstein didn't 'invent' new mathematics as Newton did. Lorentz had the equations for special relativity. Plank had described black body radiation. Einstein needed help in writing down general relativity equations. His work on Brownian motion was built upon Boltzman and statistics.
his genius was to see a different way of explaining these things. time was not invariant was not obvious. that photons are quantized (i.e. particle-like) flew in the face of Maxwell's equations.
i'm not trying to take away anything from Einstein. it's just that the popular conception of him being 'the opera singer super model brain surgeion sports star' of physics is insulting to all the other great minds. he could not have done what he did without others.
it's not that physicists hate Einstein, it's more they hate how people view Einstein. it's mostly because Einstein became the poster-boy for the media about modern physics. as we know, the media tends to simplify things and so it suggests that Einstein did relativity, photoelectric effect, etc. by himself.
given that Newton said he had seen far only because he stood on the shoulders of giants, Einstein is even more indebted to others before and during his time.
look, i'm not saying that he wasn't a remarkable physicist- when i read some of his papers, i do feel like i'm reading something that's very close to 'god'- so clear, so elegant, so beautiful.
yes, i am a physicist. and yes, i do find myself using Einstein's results. i'm glad when it happens, because i *feel* i'm close to reality.
it's b.s. current high end disk drives are either at or above 100Gb/in^2. they had plans to push to 150Gb/in^2 w/ the current technology (called longitudinal recording). after that it's not clear it will work. essentially the magnetic bits (a collection of grains whose magnetic moments lie in roughly the same direction) will become unstable (spontaneous data loss).
one way to get around this (forget about IBM's pixie dust) is to do perpedicular recording. there you can pack muck higher linear densities, but you still have the problem of producing a very large magnetic field (to write) in a very small area (to keep the density you need). oh, and it gets harder to read smaller bits, of course.
i don't know where they get the data rate of kiloHz. that's b.s. too- high end drives run in the 100MHz. stuff on the drawing boards are pushing 1GHz (which, of course is a problem since magnetic media is limited by something called the gyromagnetic frequency).
yes, SOFC's do run at high temperature, so to convert the fuel into hydrogren is 'free'. it's all chemically driven.
;)
as for producing carbon dioxide, CO2, well, that's not a problem, right? i mean, that's what i and you produce when we breathe and that's what plants use to create O2. it's much better than producing carbon monoxide, CO- since that's a greenhouse gas.
also, since SOFC's do run hot- the H2O you get out is steam which you can recover to run a turbine (like typical power plants do- burn coal or oil to produce steam to turn a turbine which then produces electricity). so SOFCs can be very efficient. more efficient than current technology.
be careful, fuel cells can just store energy, or they can produce it. there are mutiple types of such things for different applications.
the major problem is, yes, cost, but also lifetime. you need these things to survive for 40,000 hours to be feasiable. that's not trivial and turns into a material science problem. it's not impossible, but currently we don't know how. that is why i go to work
don't understand why everyone is so down about fuel cells.
yes, pure hydrogen is hard/expensive to produce. but the next generation of fuel cells can use methane (or ethanol) for a source of fuel. ie, plug the fuel cell into the back end of a cow- suddenly wisconsin will be known for more than it's cheese.
for some reason, some are thinking fuel cells are going to replace gasoline engines in vechicles. well, ok. but what you really want to do is replace all the coal and oil burning power plants w/ fuel cells. so instead of acid rain and tons of greenhouse gases, you get H2O out, which you could use to water crops or drink. given that China seems to be building coal burning power plants as fast as they can, doesn't that sound like a good idea?
ok, fine, i might be biased. i am working on the next design of fuel cells (in particular solid oxide fuel cells- SOFC). but, still, the sooner we get to a place where producing energy is less harmful to the planet, i think we should. hell, we must.
albeit, my experiences w/ WD have all been older than 3 years (two cheapo WD died within two weeks after six months. and before,...). if it's something you don't want to lose and you don't have copies, you're the only one to blame.