Actually, widely separating the frequencies is good. It simplifies the filtering necessary to do full-duplex.
Approximately three-to-one frequency split lets you use the same antenna, too. A quarter-wave for the lower frequency is about three-quarters for the higher, and will load up reasonably well - especially with a the odd loading coil or capacitor here and there - which just might fall out as a side-effect of the band splitter.
You want the talk channel to be better than the listen channel, so you don't keep yattering away at somebody that YOU can hear just fine but who can't hear you.
I'm not a Physicist, but I'll try to play one on the net. Here's what I think is the current theory:
The vacuum is full of virtual particle-antiparticle pairs, constantly forming and annihilating, with a mass-time product less than the uncertainty principle's magic number. When a pair occurs near an event horizon, one of the particles can tunnel deep enough into it to be annihilated by its antiparticle below the horizon, allowing the partner to escape - as if the particle below the horizon had tunneled out. (If there isn't an antiparticle available, there isn't energy available to kick loose the particle that didn't penetrate the event horzon. So it falls in, too, and the virtual particle-antiparticle pair disappear back into the vacuum.)
So black holes evaporate. Bigger black holes have a bigger separation between the mass and the event horizon, and thus a lower mass density just under it. So the smaller the black hole the faster it eveporates. "Evaporate" means emit a spray of energetic subatomic particles.
If I have the constants right, a stellar-sized black hole emits the odd particle now and then, a mountain-mass black hole is a good approximation of a nuclear power plant's core, and so on. But radiation reduces their mass, so the faster they radiate, the faster they shrink, and the FASTER their radiation increases, until the event horizon suddenly disappears and the remaining particles come blasting out of the former cage at nearly lightspeed. It goes BANG big-time - because this happens when there's still a lot of stuff in there. Current high-end H-bombs would blush with envy.
A black hole with the mass of a couple heavy ions would have a very short lifetime, even as compared with other subnuclear processes. Making one that would have a lifetime in seconds would consist of creating a density of matter that would push stuff through the event horizon faster than it tunnels out. That's equivalent to making a BIG atomic fireball and squeezing it down to the size of a single nucleus.
So we might see black holes as screwier-than-usual short-lived composite particles acting as intermediate steps in sunuclear reactions. But we shouldn't see a baby black hole falling quietly out of the accellerator and eating the earth.
Of course, my understanding of the model could be wrong. B-)
Or the model could be wrong. In which case, other predictions from it (such as the hole forming in the first place) are also up for grabs.
A DS3 (About 44 Mbps, or 5.5 MByte/sec) runs about $1500 PER MONTH. This gadget is the bandwidth of about 227 of them. The equivalent of 100 times the bandwidth of a 100Mbps ethernet, vs a T3 which is a bit less than half of one.
If it costs a few grand per box, one-time, a company with two or more buildings within line-of-sight of each other but not on contiguous land, in a region with even moderately good weather, might buy them to connect the building LANs, and fall back to a puny T1 (1.5 Mbps) on stormy days. BIG bargain.
I think it was Control Data that once used the spin of the disks to provide enough power to write the disk cache and otherwise get the system shut down gracefully in a sudden power failure.
BIG, HEAVY disks in those days. About the time Cray was just finishing up there before striking out on his own.
(Part of the trick is to have your write clock come off a servo track and the PLL able to follow it down a few percent as the disk slows.)
It's already being looked at. A recent conference I attended had a (not entirely tongue-in-cheek) discussion on powering some of the new low-drain laptops with various battery-alternatives, including mechanicals such as the wound-spring generators you see in portable radios, or similar stuff with a large weight hung by a pulley from a tree.
A problem with most current keyboards is that they don't do a good job absorbing the (considerable) energy imparted to the keys. Some of it gets dissipated putting wear on the keyboard, but most of it puts wear on the tendons, muscles, and ligaments.
Soaking it up in the keyboard is good for the hands. Soaking it up by turning it into juice and shoving it somewhere, rather than flexing and heating keyboard moving parts, is good for a hand-friendly keyboard.
And as long as you've turned it into juice, why not shove it into the battery. It may not be enough to power your machine, but it might make a noticeable difference in battery life. (And you might be surprised: There's a LOT of power in physical motion, and electromagnetic generators with losses of a few percent are considered inefficient.)
Amdahl has been doing this for years - splitting up multiprocessor mainframes and the attached equipment into "domains" - separate virtual machines. One domain might be running VM on 4 CPUs, another running UTS (mainframe Unix) on a half-CPU timeslice, and so on.
Not long ago they rehacked their machine identity system so that you could license software for the number of CPUs in a domain rather than the number in the box.
So they're only paying a 75-cent premium over current market for the shares. This means the major talents' stock options just turned into shit.
If IBM doesn't come up with an additional incentive for them within the next couple weeks, watch for them to start taking a hike as soon as they've had a chance to evaluate offers, line up financing for their next startup, and/or the minute their next vesting increment hits.
Re:What about underweight hackers?!
on
Hacker's Diet
·
· Score: 1
There are also a vitamin or two that are short in veggies. B12 in particular: It's almost impossible to become B12 deficient - the requirement is in micrograms. But you can do it with an ill-chosen strict vegitarian diet.
(That's why Folic Acid supliments are limited by FDA regulations: Folic Acid masks the early symptoms of B12 deficiency, so a vegitarian taking them can acquire serious and irreparable neurological damage before the other symptoms show up. Stupid on the FDA's part: It condems a lot of pregnant women to morning sickness - which is largely a Folic Acid deficiency resulting from the Foetus' rapid cell replication and the resulting Folic Acid uptake. If they must protect vegitarians from themselves, they should instead require a trace of B12 in high-dose Folic Acid supliments.)
Of course a strict cooked-meat diet can also lead to deficiencies - notably some heat-sensitive vitamins such as C and E.
People are omnivores, and cursorial hunters. We can survive for quite a while, and even prosper, on just veggies or just meat. But we do best on a mix.
Re:Some of that book's thinking is simplistic
on
Hacker's Diet
·
· Score: 1
I've seen John post-diet - several times over several years - and he managed to come out of it with permanent weight loss, no visible excess fat, no skin wrinkles, and no other debilitating symptoms (other than male pattern baldness, which was probably a coincidence from his heredity and the timing of his diet).
I must admit that I did NOT see him with his clothes off. But there wasn't room under there to hide any significant fat - not even a bicycle-sized spare tire. Neither emaciation nor excess rounding.
(Sigh. Reminds me of the figure I had in my late teens - right up until I got my first car and stopped jogging everywhere I went.)
Slashdot Mirror
Actually, widely separating the frequencies is good. It simplifies the filtering necessary to do full-duplex.
Approximately three-to-one frequency split lets you use the same antenna, too. A quarter-wave for the lower frequency is about three-quarters for the higher, and will load up reasonably well - especially with a the odd loading coil or capacitor here and there - which just might fall out as a side-effect of the band splitter.
You want the talk channel to be better than the listen channel, so you don't keep yattering away at somebody that YOU can hear just fine but who can't hear you.
Interesting that you can buy a whole Apple base station for the price of a PC plug-in card, and an apple plug-in card for a third of that.
I suspect this will put some price pressure on PC cards. Expect a price drop shortly.
I'm not a Physicist, but I'll try to play one on the net. Here's what I think is the current theory:
The vacuum is full of virtual particle-antiparticle pairs, constantly forming and annihilating, with a mass-time product less than the uncertainty principle's magic number. When a pair occurs near an event horizon, one of the particles can tunnel deep enough into it to be annihilated by its antiparticle below the horizon, allowing the partner to escape - as if the particle below the horizon had tunneled out. (If there isn't an antiparticle available, there isn't energy available to kick loose the particle that didn't penetrate the event horzon. So it falls in, too, and the virtual particle-antiparticle pair disappear back into the vacuum.)
So black holes evaporate. Bigger black holes have a bigger separation between the mass and the event horizon, and thus a lower mass density just under it. So the smaller the black hole the faster it eveporates. "Evaporate" means emit a spray of energetic subatomic particles.
If I have the constants right, a stellar-sized black hole emits the odd particle now and then, a mountain-mass black hole is a good approximation of a nuclear power plant's core, and so on. But radiation reduces their mass, so the faster they radiate, the faster they shrink, and the FASTER their radiation increases, until the event horizon suddenly disappears and the remaining particles come blasting out of the former cage at nearly lightspeed. It goes BANG big-time - because this happens when there's still a lot of stuff in there. Current high-end H-bombs would blush with envy.
A black hole with the mass of a couple heavy ions would have a very short lifetime, even as compared with other subnuclear processes. Making one that would have a lifetime in seconds would consist of creating a density of matter that would push stuff through the event horizon faster than it tunnels out. That's equivalent to making a BIG atomic fireball and squeezing it down to the size of a single nucleus.
So we might see black holes as screwier-than-usual short-lived composite particles acting as intermediate steps in sunuclear reactions. But we shouldn't see a baby black hole falling quietly out of the accellerator and eating the earth.
Of course, my understanding of the model could be wrong. B-)
Or the model could be wrong. In which case, other predictions from it (such as the hole forming in the first place) are also up for grabs.
A DS3 (About 44 Mbps, or 5.5 MByte/sec) runs about $1500 PER MONTH. This gadget is the bandwidth of about 227 of them. The equivalent of 100 times the bandwidth of a 100Mbps ethernet, vs a T3 which is a bit less than half of one.
If it costs a few grand per box, one-time, a company with two or more buildings within line-of-sight of each other but not on contiguous land, in a region with even moderately good weather, might buy them to connect the building LANs, and fall back to a puny T1 (1.5 Mbps) on stormy days. BIG bargain.
I think it was Control Data that once used the spin of the disks to provide enough power to write the disk cache and otherwise get the system shut down gracefully in a sudden power failure.
BIG, HEAVY disks in those days. About the time Cray was just finishing up there before striking out on his own.
(Part of the trick is to have your write clock come off a servo track and the PLL able to follow it down a few percent as the disk slows.)
It's already being looked at. A recent conference I attended had a (not entirely tongue-in-cheek) discussion on powering some of the new low-drain laptops with various battery-alternatives, including mechanicals such as the wound-spring generators you see in portable radios, or similar stuff with a large weight hung by a pulley from a tree.
A problem with most current keyboards is that they don't do a good job absorbing the (considerable) energy imparted to the keys. Some of it gets dissipated putting wear on the keyboard, but most of it puts wear on the tendons, muscles, and ligaments.
Soaking it up in the keyboard is good for the hands. Soaking it up by turning it into juice and shoving it somewhere, rather than flexing and heating keyboard moving parts, is good for a hand-friendly keyboard.
And as long as you've turned it into juice, why not shove it into the battery. It may not be enough to power your machine, but it might make a noticeable difference in battery life. (And you might be surprised: There's a LOT of power in physical motion, and electromagnetic generators with losses of a few percent are considered inefficient.)
Amdahl has been doing this for years - splitting up multiprocessor mainframes and the attached equipment into "domains" - separate virtual machines. One domain might be running VM on 4 CPUs, another running UTS (mainframe Unix) on a half-CPU timeslice, and so on.
Not long ago they rehacked their machine identity system so that you could license software for the number of CPUs in a domain rather than the number in the box.
So they're only paying a 75-cent premium over current market for the shares. This means the major talents' stock options just turned into shit.
If IBM doesn't come up with an additional incentive for them within the next couple weeks, watch for them to start taking a hike as soon as they've had a chance to evaluate offers, line up financing for their next startup, and/or the minute their next vesting increment hits.
There are also a vitamin or two that are short in veggies. B12 in particular: It's almost impossible to become B12 deficient - the requirement is in micrograms. But you can do it with an ill-chosen strict vegitarian diet.
(That's why Folic Acid supliments are limited by FDA regulations: Folic Acid masks the early symptoms of B12 deficiency, so a vegitarian taking them can acquire serious and irreparable neurological damage before the other symptoms show up. Stupid on the FDA's part: It condems a lot of pregnant women to morning sickness - which is largely a Folic Acid deficiency resulting from the Foetus' rapid cell replication and the resulting Folic Acid uptake. If they must protect vegitarians from themselves, they should instead require a trace of B12 in high-dose Folic Acid supliments.)
Of course a strict cooked-meat diet can also lead to deficiencies - notably some heat-sensitive vitamins such as C and E.
People are omnivores, and cursorial hunters. We can survive for quite a while, and even prosper, on just veggies or just meat. But we do best on a mix.
I've seen John post-diet - several times over several years - and he managed to come out of it with permanent weight loss, no visible excess fat, no skin wrinkles, and no other debilitating symptoms (other than male pattern baldness, which was probably a coincidence from his heredity and the timing of his diet).
I must admit that I did NOT see him with his clothes off. But there wasn't room under there to hide any significant fat - not even a bicycle-sized spare tire. Neither emaciation nor excess rounding.
(Sigh. Reminds me of the figure I had in my late teens - right up until I got my first car and stopped jogging everywhere I went.)