The final barrier is the price of packaging and selling the computer. I expect even that to fall eventually: They will become "marginal", what many misinterpret as "free". They will literally become so cheap that they are given away with something else, which includes their cost in its overhead.
It's like elevator rides: Elevators are expensive to build, power, and maintain. But the cost of each ride is very low. So instead of a ticket-taker in the car you have the elevator included in the cost of rental or maintainence of a building, and rides are "free" to all comers - residents, customers, sightseers, pests, etc.
They used to say that about the alleged $1000 price barrier.
Walmart nicked the $200 "price barrier" in 2002 and it's still dropping.
If you don't need a screen and keyboard you can get a linux-based some-other-appliance, like a router, for under $50 no problem, and reload it with what you want to run. Or a linux-based fileserver appliance with a big disk, decent RAM, and a moderate CPU, and install all the apps you want.
if you have to re-convert anyway, 5V as intermediate voltage is not optimal. When converting to 5V, the voltage drop in the power diodes and in the wires to the mainboard eats a much higher proportion of the power than with 12V as intermediate voltage. 24V or even 48V would be even better.
Telephony has been running on redundant -48V DC supplies to the racks (typically from rooms full of floating storage batteries) since the early relay days. Much modern networking equipment also conforms to this standard, so it can be used in such racks with no local power supply (except the per-card isolation diodes and downconverters).
Power conversion modules running from 48V are in volume production.
Why does Google want to reinvent this wheel?
(However, if they do insist on using 12v, I hope they make it able to work from 11.75v to about 15V, with glitches, and shut off at stable levels below 11.75v. That way such boards could be used directly with 12v renewable energy systems, plugged directly into an automobile "cigarette lighter" power outlet, or easily wired into a vehicle or travel trailer as an appliance.)
It's a very rare gas station (usually the one with the lowest price in town) that is operating at full capacity 24/7.
This is the power grid. There is no storage. 24/7 doesn't matter. Peak matters. We're not talking enough generation to power the cars with slow charges. We're talking enough generation to replace the current number of gas pumps with an equivalent (though smaller) number of "juice pumps" to do the same job.
You also need to consider the overall reduction in "pump" demand if people can recharge overnight at home.
No I don't, since the exercise was to see what would happen in the case of simply replacing gasoline filling stations with electric ones.
Charging at home overnight would certainly reduce the number of plants needed by spreading the load over more time. Maybe cut it in half. It also allows the use of more of the off-peak capacity of the existing plants, further improving the situation on the generation side.
Unfortunately, it would multiply the required capacity of the residential grid by a factor of several. This would require a far bigger distribution buildout than just running primary wires to "juice station" locations (which are also mainly located adjacent to existing rights-of-way).
That's about 1,360 new nuclear plants to feed 'em.
You mean reactors. A plant can have multiple -- as you point out with Canyon Diablo.
Yep. Call it new nuclear reactor UNITS. Maybe half as many nuke PLANTS. Maybe a quarter as many if you average 4 units per plant, which you might if you decided to build out for such a load. (Density of plants is limited by cooling resources.)
And those reactors (or equivalents) would eliminate the need for how many billions of barrels of oil and the consequent political ramificaitions?
No question.
In fact, stationary plants are so much more efficient than automobile engines carried with the car they power that even burning the oil in them and accepting the transmission and storage losses would be an improvement.
Err... the station would undoubtedly have a bank of these capacitors continuously charging 24 hours a day in order to avoid running out at peak times, just as a conventional station has a storage tank for gasoline, rather than a direct pipeline all the way to the refinery sized for the peak demand.
Unlikely.
It wouldn't do them a lot of good: During rush hour they'd be running at capacity for hours. The storage required would be tremendous: A ten-"pump" station would require storage equivalent to the storage systems of 60 cars for each hour they had to buffer. For just a 4-hour "rush hour" you're talking 78 megawatt hours.
The capital cost would be prohibitive. Further: It would make one "juice station" comparable to a peaking power plant suitable for a small city, rather than an instalation you can put in on any street corner.
Then they would burn a LOT of power to the inefficiencies of charging one capacitor from another, which would raise the cost of the power - and the amount of power generated. It would also be dissipated as thermal pollution at the "juice station" site. A ten pump station with only 10% loss would be dumping almost two megawatts of heat during that time. Call it about a fifth of the heat you'd get if you went to an ordinary gas station, put a candle on the ground, started a pump, and left it pumping gasoline into the burning pool continuously for hours at a time.
Then, with supercapacitors, there's the issue of catastrophic sudden discharge. This would make such an instalation the equivalent of a fuel-air explosive bomb containing about three tons of gasoline for the fuel.
Sounds like removal of a "lipoma". (I've had a few of those removed.)
Think of it as "cancer of the fat" - except benign. You get stiff fatty lumps (maybe one, maybe a scattering, maybe like a bunch of grapes). They're like regular fat with some kind of other tissue in them that makes them hard.
It's really annoying if it's above a muscle or some other easily hurt tissue: It's like a rock embedded in the fat that is SUPPOSED to be cushioning the tissue, so lying on it bruises the tissue instead.
They never go malignant so doctors will leave them in unless they're bruising something underneath or causing a disfiguring bump. They're near the surface of the skin so they're easy to cut out - usually by a dermatologist.
Sounds like the perfect test operation. Not a big deal if they don't get it all, near the surface so you don't have to cut through vital stuff or clamp stuff out of the way to get to it, etc. Easy to tell how well the op went. Much less opportunity for screwups than just about any other surgery.
Now that I look at it: They're claiming 500 mile range. Typical range for a 5min gas fillup is in the 250 mile ballpark.
So IMHO they're seriously puffing somewhere. (Probably by assuming regenerative braking and quoting city stop-and-go range, not cross-country or freeway commute.)
So figure it's really $29.20 to "juice up". (But you can get by with the same number of nuke plants and fewer "pumps" active at a time, presuming the drastically lower "gas price" doesn't lead to increased driving.)
A watt is about 3.41 BTU/hr, so we're talking 130 Megawatts.
Typo: 13 Megawatts (of heat). (Number is right in the next calculation.)
= = = =
I note that another poster, working from the article's $9 worth of juice in 5 min got 1.2 Megawatts. Converting that to a 3 minute charge makes it 2 Megawatts. Moderately close to my 3.25.
I read this as the supercapacitor guys estimating better efficiency for their electric car so they don't have to quote $14.60 for the fillup. B-)
I've timed pumps: they take about 10 seconds per galon, or feed about 6 gal/minute.
Lets use 124,000 BTU for a galon of gas. (Web searches give numbers from 123k to 125k.)
That's about 44.64 MILLION BTU/hr.
A watt is about 3.41 BTU/hr, so we're talking 130 Megawatts.
But that's heat. A gas engine is in the ballpark of 20-25% efficient turning it into HP, while electric is in the 80s to 90s, so divide by four. Now we're talking about 3 1/4 Megawatts to charge an electric car of the same range, weight, and profile at the same rate.
We'll assume we can neglect redesigns for better airodynamics, weight, tires, etc. since they could be applied equally to gas and electrics. We'll assume that the storage system plus engine/transmission is about a wash (i.e. the weight of the batteries/ultracapaictor bank at least cancels the weight advantage of a motor vs. an engine). And we'll assume we're talking about highway range rather than stop-and-go traffic, so regenerative braking doesn't enter the equation.
Let's take Sacremento CA's Ranco Seco and Diablo Canyon nuclear plants as "one nuclear plant" capacity. (Rancho Seco is located inland so it could presumably be cloned anywhere there's enough spare water to feed cooling towers.) Rancho Seco is 918,000 kW running flat out, and the two Diablo Canyon units are 1,084,000 and 1,106,000, so call a nuclear plant 1,000,000 kW max.
That says one nuclear plant unit can simultaneously drive only 306 "electric gas pumps". Utility electricity has to be generated as it's used and come rush hour the stations will often be "filling" on all pumps. At about 10 pumps per station that's only about 30 "electric gas stations" per nuclear plant. Say fifty if it takes about three minutes to fill and two minutes to pay.
A prof at University of Detroit Mercy counted the number of gas stations in the US in 2002 at 170,018. Presume they average about 4 pumps each (or would if they didn't need to stock multiple octanes of electricity B-) ). That's about 1,360 new nuclear plants to feed 'em.
Worked it out once, assuming then-prevalant estimates of HP requirements for cars: You need about ten times as much generation for recharging one household car as you need for the household itself (and a two-income family will typically have two cars and two daily commutes, while a one-income family also needs to shop). That's a LOT of extra electric plants and distribution wiring.
That was a few years back. Tech has improved somewhat (as has other aspects of the utility environment), so a new estimate is needed.
But there are limits to how little power you can spend to carry a car full of people and their cargo down the road at reasonable speeds, or up a mountain.
You mean copyright isn't ENFORCED by the GAME OPERATORS.
Copyright exists in the countries of the real world in which the game is embedded, and can be enforced on the players by the legal system of those countries' governments.
Exception would be if the players had explicitly contracted away their copyrights as a condition of participating. In which case copyright could STILL be enforced - but enforcement would only be of interest when somebody without the rights to waive entered a derivative of someone else's designs, or cloned it from another player who didn't have the rights to waive.
Examples would be a work derived from a non-player's copyrighted design, or (depending on the contract wording) a design that a player had created externally and had not inserted into the game.
(IANAL: This is just my understanding of the law.)
Look at the direction of reflection of the "rays" - approximations of the actual wave propagation. Compute the momentum transfer to the walls from those reflections. You'll find that the component of it along the major axis matches and cancels the imbalance of the reflectons on the two ends.
His error was to ignore this reacton on the side of the cavity.
(Pity. An engine that turned electricity into thrust, violating the law of conservation of momentum in the process, would have been nice. B-) )
Is anyone else reminded...of those cartoons where Bugs Bunny or someone is sitting in a sailboat, pulls out a fan, aims it at the sail......and the boat moves?
That actually works. A little bit.
But it works MUCH BETTER if you just point the fan to the rear.
The fan sucks air from a lot of directions and ejects it in one direction, creating a net thrust (and reaction - backward - on the boat via the person holding the fan) and a net wind.
Diverting that wind to the rear via the sail produces somewhat more reaction forward on the boat via the sail and the mast than the reaction backward from the fan - IF the trim is good enough that the diverted wind ends up going backward rather than just off to the sides. Result: Slight net forward thrust on the boat.
But pointing the fan to the rear - using it as a jet - eliminates the inefficiencies of using the sail in this way, putting the fan's whole reaction into moving the boat forward.
There is no synchronized clock with regards to DOCSIS. PacketCable uses VoIP technology and, as the name implies, uses ip data packets for call transmission.
There is a synchronized clock with DOCSIS.
Note that all three of the supported data rates are even multiples of 8000 hz. This is not a coincidence - it's where the clock is carried.
Now look at, say, Annex A of the 3.0 spec ("Timing Requirements for Supporting Business Services over DOCSIS"):
A.1 CMTS
DOCSIS 3.0 CMTSs have the following timing and synchronization requirements. To fully support T1 and E1 services, the CMTS should, in addition, comply with the requirements outlined in [TEI]. Specifically, the CMTS should be able to send TDM timing information down the DOCSIS RF downstream...
- The CMTS MUST be able to lock the downstream DOCSIS symbol clock to an external Stratum 1 clock source (PRS or DTI). - The CMTS MUST be able to lock the downstream DOCSIS symbol clock to an external Stratum 1 clock source (PRS or DTI).
- The CMTS MUST be able to lock the downstream DOCSIS SYNC messages to an external Stratum 1 clock source (PRS or DTI)....
A.2...
- The CM SHOULD have a TDM clock output (in DOCSIS 3.0 operation, and in pre-3.0 DOCSIS S-CDMA operation, synchronzied to the downstream Symbol Clock) from which a T1 or E1 clock can be regenerated.
- The CM SHOULD have a TDM clock output (in pre-3.0 DOCSIS TDMA operation, synchronzied to the SYNC message timing information) from which a T1 or E1 clock can be regenerated.
And so on.
The "External Stratum 1 clock source" they're talking about is the tellco sampling clock, used to synchronize T1, E1, ISDN, and SONET carriers across the entire network - in order to control the timing of the A/D sampling and reconstruction.
DOCSIS uses time-division multiplexing to orchestrate the multiple boxes on the upstream path, so it needed to distribute an accurate clock to the remote boxes to synchronize them with the head end and each other. The designers were thinking ahead and chose the telephone industry's standard clock, specificially so they could transport telephone TDM signals, WITH timing, to/from the remote boxes. That includes, not just T1 and E1 carriers themselves, but also the individual DS0 "telephone call" signals. So the DOCSIS standard prescribes the abililty to propagate and regenerate the telco clock.
There are two standards that deal with [voice or set-top boxes]: PacketCable is the cablelabs standard for voice. OpenCable is the cablelabs standard for settop boxes.
CableLabs' PacketCable standard is built on top of the DOCSIS 1.1. It uses the QoS features of 1.1 to provide guaranteed packet delivery.
I don't have the voice transport standard itself handy. But if you go to the OpenCable portion of their web site you'll see that they also wrote the DOCSIS timing interface and protocol standards - which distribute enough information to reconstruct, not just the 8 kHz rate, but even GPS timestamps, as part of synchronizing the head end and subscriber end boxes. (GPS is what essentially all the telcos use to synchronize their clocking these days, having graduated from their own distribution networks and LORAN long ago.)
With this clocking available - and CableLabs attention to it and understanding of the issues - I'd be flabbergasted if they're not using it in the standard to produce a telco-quality synchronized DS0 link for their POTS-over-cable standard.
Once you have the timing available, doing synchronized TDM is SO much easier than trying to do a timing-independent kludge.
Using bare G.711 will improve the quality. But it won't do a thing about packet drop, packet jitter, or clock inaccuracies.
To see the effect of clocking inaccuracies, try this: Configure your VoIP for G.711 and try to make a 56k modem connection over it and see what speed you get.
Obviously, HP went too far in their actions. Investigating within the corporation is one thing, but going outside the corporation, in the manner they did, is beyond the pale.
There was NOTHING wrong with investigating the outsiders involved with the leaks. Even (especially) if they were reporters.
What WAS wrong was USING ILLEGAL MEANS to do the investigation.
[leaking] insider info is illegal, highly damaging to the market, consumers, and stockholders.
Also damaging to the company's internal structure.
A few years ago the company I was at was having trouble with leaks. It finally came to a head when a trade journal practically quoted the CEO's presentation to the company meeting verbatim.
Net result was that upper management STOPPED TELLING THE REST OF THE COMPANY EMPLOYEES ANYTHING AT ALL about the company's forward planning.
The "cable phone service" in question is NOT VoIP. It is a TDM service over the DOCSIS digital cable standard.
This gives you an uncompressed DS0 digital phone connection, with no dropped packets, and with signal quality equivalent to having a few feet, rather than miles, of wire between the POTS instrument and the "phone company"'s D-A converter.
VoIP uses lossy compression, introduces much more latency, has dropped packets due to internet congestion, and has rotten timing for the A-D/D-A conversions.
You won't get "pin drop" signal quality with VoIP unless/until the network is reworked for QoS guarantees, the carriers are allowed to treat packets for phone streams better than they do those for services like FTP, and a billing model and contracts between carriers is in place for it to make sense for the carriers to extend these QoS guarantees to their competitor/partner's packets.
THIS is why any "Network Neutrality" proposals that treat all PACKETS equally - rather than packets of equivalent services from all PROVIDERS - breaks streaming services (among other things).
Cable providers also led the way in audio quality; the top firm in Keynote's study actually turned in an MOS of 4.24, above most real phone networks.
If I recall the DOCSYS standard correctly (that's the one for cable settop boxes), the framing provides the phone company TDM-style 8 kHz synchronized clock, and the phone signals are carried as full-rate uncompressed bytes.
In other words, POTS-over-cable is a 64 kbps synchronized digital signal, identical to what's carried on the phone companies' own ISDN, T, and SONET carriers, and is switched onmodified on and off the rest of the digital network unmodified. The A-D conversion happens in the settop box. It's like having your POTS phone at the switching center within wire-feet of the multiplexer. (The clocking is also good enough to encode analog signals from FAX and 56K computer modems. It has to be, as a side-effect of the need to time the upstream packets properly.)
POTS, on the other hand, is A-D converted at a central office or a "remote terminal" (in a box at the curb) and carried the rest of the way - blocks to miles - in one of a bundle of wires. This is subject to crosstalk, distortion (selective delay and attenuation of higher frequencies), and a number of other pathologies that lower the signal quality.
So it is not at ALL surprising that cable POTS signal quality beats telco POTS. Cable's signal is about as pristine as you can get.
(And VoIP isn't in the same ballpark, due to both compression and timing uncertainties.)
Slashdot Mirror
The final barrier is the price of packaging and selling the computer. I expect even that to fall eventually: They will become "marginal", what many misinterpret as "free". They will literally become so cheap that they are given away with something else, which includes their cost in its overhead.
It's like elevator rides: Elevators are expensive to build, power, and maintain. But the cost of each ride is very low. So instead of a ticket-taker in the car you have the elevator included in the cost of rental or maintainence of a building, and rides are "free" to all comers - residents, customers, sightseers, pests, etc.
They used to say that about the alleged $1000 price barrier.
Walmart nicked the $200 "price barrier" in 2002 and it's still dropping.
If you don't need a screen and keyboard you can get a linux-based some-other-appliance, like a router, for under $50 no problem, and reload it with what you want to run. Or a linux-based fileserver appliance with a big disk, decent RAM, and a moderate CPU, and install all the apps you want.
if you have to re-convert anyway, 5V as intermediate voltage is not optimal. When converting to 5V, the voltage drop in the power diodes and in the wires to the mainboard eats a much higher proportion of the power than with 12V as intermediate voltage. 24V or even 48V would be even better.
Telephony has been running on redundant -48V DC supplies to the racks (typically from rooms full of floating storage batteries) since the early relay days. Much modern networking equipment also conforms to this standard, so it can be used in such racks with no local power supply (except the per-card isolation diodes and downconverters).
Power conversion modules running from 48V are in volume production.
Why does Google want to reinvent this wheel?
(However, if they do insist on using 12v, I hope they make it able to work from 11.75v to about 15V, with glitches, and shut off at stable levels below 11.75v. That way such boards could be used directly with 12v renewable energy systems, plugged directly into an automobile "cigarette lighter" power outlet, or easily wired into a vehicle or travel trailer as an appliance.)
It's a very rare gas station (usually the one with the lowest price in town) that is operating at full capacity 24/7.
This is the power grid. There is no storage. 24/7 doesn't matter. Peak matters. We're not talking enough generation to power the cars with slow charges. We're talking enough generation to replace the current number of gas pumps with an equivalent (though smaller) number of "juice pumps" to do the same job.
You also need to consider the overall reduction in "pump" demand if people can recharge overnight at home.
No I don't, since the exercise was to see what would happen in the case of simply replacing gasoline filling stations with electric ones.
Charging at home overnight would certainly reduce the number of plants needed by spreading the load over more time. Maybe cut it in half. It also allows the use of more of the off-peak capacity of the existing plants, further improving the situation on the generation side.
Unfortunately, it would multiply the required capacity of the residential grid by a factor of several. This would require a far bigger distribution buildout than just running primary wires to "juice station" locations (which are also mainly located adjacent to existing rights-of-way).
That's about 1,360 new nuclear plants to feed 'em.
You mean reactors. A plant can have multiple -- as you point out with Canyon Diablo.
Yep. Call it new nuclear reactor UNITS. Maybe half as many nuke PLANTS. Maybe a quarter as many if you average 4 units per plant, which you might if you decided to build out for such a load. (Density of plants is limited by cooling resources.)
And those reactors (or equivalents) would eliminate the need for how many billions of barrels of oil and the consequent political ramificaitions?
No question.
In fact, stationary plants are so much more efficient than automobile engines carried with the car they power that even burning the oil in them and accepting the transmission and storage losses would be an improvement.
Thanks for that. (It's 'way different from what I was told.)
Err... the station would undoubtedly have a bank of these capacitors continuously charging 24 hours a day in order to avoid running out at peak times, just as a conventional station has a storage tank for gasoline, rather than a direct pipeline all the way to the refinery sized for the peak demand.
Unlikely.
It wouldn't do them a lot of good: During rush hour they'd be running at capacity for hours. The storage required would be tremendous: A ten-"pump" station would require storage equivalent to the storage systems of 60 cars for each hour they had to buffer. For just a 4-hour "rush hour" you're talking 78 megawatt hours.
The capital cost would be prohibitive. Further: It would make one "juice station" comparable to a peaking power plant suitable for a small city, rather than an instalation you can put in on any street corner.
Then they would burn a LOT of power to the inefficiencies of charging one capacitor from another, which would raise the cost of the power - and the amount of power generated. It would also be dissipated as thermal pollution at the "juice station" site. A ten pump station with only 10% loss would be dumping almost two megawatts of heat during that time. Call it about a fifth of the heat you'd get if you went to an ordinary gas station, put a candle on the ground, started a pump, and left it pumping gasoline into the burning pool continuously for hours at a time.
Then, with supercapacitors, there's the issue of catastrophic sudden discharge. This would make such an instalation the equivalent of a fuel-air explosive bomb containing about three tons of gasoline for the fuel.
No, I don't think that's a viable solution.
Sounds like removal of a "lipoma". (I've had a few of those removed.)
Think of it as "cancer of the fat" - except benign. You get stiff fatty lumps (maybe one, maybe a scattering, maybe like a bunch of grapes). They're like regular fat with some kind of other tissue in them that makes them hard.
It's really annoying if it's above a muscle or some other easily hurt tissue: It's like a rock embedded in the fat that is SUPPOSED to be cushioning the tissue, so lying on it bruises the tissue instead.
They never go malignant so doctors will leave them in unless they're bruising something underneath or causing a disfiguring bump. They're near the surface of the skin so they're easy to cut out - usually by a dermatologist.
Sounds like the perfect test operation. Not a big deal if they don't get it all, near the surface so you don't have to cut through vital stuff or clamp stuff out of the way to get to it, etc. Easy to tell how well the op went. Much less opportunity for screwups than just about any other surgery.
Now that I look at it: They're claiming 500 mile range. Typical range for a 5min gas fillup is in the 250 mile ballpark.
So IMHO they're seriously puffing somewhere. (Probably by assuming regenerative braking and quoting city stop-and-go range, not cross-country or freeway commute.)
So figure it's really $29.20 to "juice up". (But you can get by with the same number of nuke plants and fewer "pumps" active at a time, presuming the drastically lower "gas price" doesn't lead to increased driving.)
A watt is about 3.41 BTU/hr, so we're talking 130 Megawatts.
Typo: 13 Megawatts (of heat). (Number is right in the next calculation.)
= = = =
I note that another poster, working from the article's $9 worth of juice in 5 min got 1.2 Megawatts. Converting that to a 3 minute charge makes it 2 Megawatts. Moderately close to my 3.25.
I read this as the supercapacitor guys estimating better efficiency for their electric car so they don't have to quote $14.60 for the fillup. B-)
I've timed pumps: they take about 10 seconds per galon, or feed about 6 gal/minute.
Lets use 124,000 BTU for a galon of gas. (Web searches give numbers from 123k to 125k.)
That's about 44.64 MILLION BTU/hr.
A watt is about 3.41 BTU/hr, so we're talking 130 Megawatts.
But that's heat. A gas engine is in the ballpark of 20-25% efficient turning it into HP, while electric is in the 80s to 90s, so divide by four. Now we're talking about 3 1/4 Megawatts to charge an electric car of the same range, weight, and profile at the same rate.
We'll assume we can neglect redesigns for better airodynamics, weight, tires, etc. since they could be applied equally to gas and electrics. We'll assume that the storage system plus engine/transmission is about a wash (i.e. the weight of the batteries/ultracapaictor bank at least cancels the weight advantage of a motor vs. an engine). And we'll assume we're talking about highway range rather than stop-and-go traffic, so regenerative braking doesn't enter the equation.
Let's take Sacremento CA's Ranco Seco and Diablo Canyon nuclear plants as "one nuclear plant" capacity. (Rancho Seco is located inland so it could presumably be cloned anywhere there's enough spare water to feed cooling towers.) Rancho Seco is 918,000 kW running flat out, and the two Diablo Canyon units are 1,084,000 and 1,106,000, so call a nuclear plant 1,000,000 kW max.
That says one nuclear plant unit can simultaneously drive only 306 "electric gas pumps". Utility electricity has to be generated as it's used and come rush hour the stations will often be "filling" on all pumps. At about 10 pumps per station that's only about 30 "electric gas stations" per nuclear plant. Say fifty if it takes about three minutes to fill and two minutes to pay.
A prof at University of Detroit Mercy counted the number of gas stations in the US in 2002 at 170,018. Presume they average about 4 pumps each (or would if they didn't need to stock multiple octanes of electricity B-) ). That's about 1,360 new nuclear plants to feed 'em.
One downside is you need more electric plants.
Worked it out once, assuming then-prevalant estimates of HP requirements for cars: You need about ten times as much generation for recharging one household car as you need for the household itself (and a two-income family will typically have two cars and two daily commutes, while a one-income family also needs to shop). That's a LOT of extra electric plants and distribution wiring.
That was a few years back. Tech has improved somewhat (as has other aspects of the utility environment), so a new estimate is needed.
But there are limits to how little power you can spend to carry a car full of people and their cargo down the road at reasonable speeds, or up a mountain.
Copyright ... Doesn't exist in game
You mean copyright isn't ENFORCED by the GAME OPERATORS.
Copyright exists in the countries of the real world in which the game is embedded, and can be enforced on the players by the legal system of those countries' governments.
Exception would be if the players had explicitly contracted away their copyrights as a condition of participating. In which case copyright could STILL be enforced - but enforcement would only be of interest when somebody without the rights to waive entered a derivative of someone else's designs, or cloned it from another player who didn't have the rights to waive.
Examples would be a work derived from a non-player's copyrighted design, or (depending on the contract wording) a design that a player had created externally and had not inserted into the game.
(IANAL: This is just my understanding of the law.)
Will real-world designers start to steal from the virtual one? Is that a copyright violation?
Yes.
It's called a "derived work".
Same thing as one designer doing a sketch for a work in progress and another copying the sketch into cloth.
Right.
You can even see it in figure 2.4.
Look at the direction of reflection of the "rays" - approximations of the actual wave propagation. Compute the momentum transfer to the walls from those reflections. You'll find that the component of it along the major axis matches and cancels the imbalance of the reflectons on the two ends.
His error was to ignore this reacton on the side of the cavity.
(Pity. An engine that turned electricity into thrust, violating the law of conservation of momentum in the process, would have been nice. B-) )
Is anyone else reminded...of those cartoons where Bugs Bunny or someone is sitting in a sailboat, pulls out a fan, aims it at the sail... ...and the boat moves?
That actually works. A little bit.
But it works MUCH BETTER if you just point the fan to the rear.
The fan sucks air from a lot of directions and ejects it in one direction, creating a net thrust (and reaction - backward - on the boat via the person holding the fan) and a net wind.
Diverting that wind to the rear via the sail produces somewhat more reaction forward on the boat via the sail and the mast than the reaction backward from the fan - IF the trim is good enough that the diverted wind ends up going backward rather than just off to the sides. Result: Slight net forward thrust on the boat.
But pointing the fan to the rear - using it as a jet - eliminates the inefficiencies of using the sail in this way, putting the fan's whole reaction into moving the boat forward.
I don't have the voice transport standard itself handy. But if you go to the OpenCable portion of their web site ...
Meant PacketCable (the relevant standard).
There is a synchronized clock with DOCSIS.
Note that all three of the supported data rates are even multiples of 8000 hz. This is not a coincidence - it's where the clock is carried.
Now look at, say, Annex A of the 3.0 spec ("Timing Requirements for Supporting Business Services over DOCSIS"):
And so on.
The "External Stratum 1 clock source" they're talking about is the tellco sampling clock, used to synchronize T1, E1, ISDN, and SONET carriers across the entire network - in order to control the timing of the A/D sampling and reconstruction.
DOCSIS uses time-division multiplexing to orchestrate the multiple boxes on the upstream path, so it needed to distribute an accurate clock to the remote boxes to synchronize them with the head end and each other. The designers were thinking ahead and chose the telephone industry's standard clock, specificially so they could transport telephone TDM signals, WITH timing, to/from the remote boxes. That includes, not just T1 and E1 carriers themselves, but also the individual DS0 "telephone call" signals. So the DOCSIS standard prescribes the abililty to propagate and regenerate the telco clock.
There are two standards that deal with [voice or set-top boxes]:
PacketCable is the cablelabs standard for voice.
OpenCable is the cablelabs standard for settop boxes.
CableLabs' PacketCable standard is built on top of the DOCSIS 1.1. It uses the QoS features of 1.1 to provide guaranteed packet delivery.
I don't have the voice transport standard itself handy. But if you go to the OpenCable portion of their web site you'll see that they also wrote the DOCSIS timing interface and protocol standards - which distribute enough information to reconstruct, not just the 8 kHz rate, but even GPS timestamps, as part of synchronizing the head end and subscriber end boxes. (GPS is what essentially all the telcos use to synchronize their clocking these days, having graduated from their own distribution networks and LORAN long ago.)
With this clocking available - and CableLabs attention to it and understanding of the issues - I'd be flabbergasted if they're not using it in the standard to produce a telco-quality synchronized DS0 link for their POTS-over-cable standard.
Once you have the timing available, doing synchronized TDM is SO much easier than trying to do a timing-independent kludge.
Using bare G.711 will improve the quality. But it won't do a thing about packet drop, packet jitter, or clock inaccuracies.
To see the effect of clocking inaccuracies, try this: Configure your VoIP for G.711 and try to make a 56k modem connection over it and see what speed you get.
Or try to send a FAX, ditto.
Obviously, HP went too far in their actions. Investigating within the corporation is one thing, but going outside the corporation, in the manner they did, is beyond the pale.
There was NOTHING wrong with investigating the outsiders involved with the leaks. Even (especially) if they were reporters.
What WAS wrong was USING ILLEGAL MEANS to do the investigation.
[leaking] insider info is illegal, highly damaging to the market, consumers, and stockholders.
Also damaging to the company's internal structure.
A few years ago the company I was at was having trouble with leaks. It finally came to a head when a trade journal practically quoted the CEO's presentation to the company meeting verbatim.
Net result was that upper management STOPPED TELLING THE REST OF THE COMPANY EMPLOYEES ANYTHING AT ALL about the company's forward planning.
This caused all sorts of problems.
Yes.
(IF that's what your cable provider is selling - not some VoIP service.)
The "cable phone service" in question is NOT VoIP. It is a TDM service over the DOCSIS digital cable standard.
This gives you an uncompressed DS0 digital phone connection, with no dropped packets, and with signal quality equivalent to having a few feet, rather than miles, of wire between the POTS instrument and the "phone company"'s D-A converter.
VoIP uses lossy compression, introduces much more latency, has dropped packets due to internet congestion, and has rotten timing for the A-D/D-A conversions.
You won't get "pin drop" signal quality with VoIP unless/until the network is reworked for QoS guarantees, the carriers are allowed to treat packets for phone streams better than they do those for services like FTP, and a billing model and contracts between carriers is in place for it to make sense for the carriers to extend these QoS guarantees to their competitor/partner's packets.
THIS is why any "Network Neutrality" proposals that treat all PACKETS equally - rather than packets of equivalent services from all PROVIDERS - breaks streaming services (among other things).
(Which goes to show how long ago I was looking into it. B-) )
Cable providers also led the way in audio quality; the top firm in Keynote's study actually turned in an MOS of 4.24, above most real phone networks.
If I recall the DOCSYS standard correctly (that's the one for cable settop boxes), the framing provides the phone company TDM-style 8 kHz synchronized clock, and the phone signals are carried as full-rate uncompressed bytes.
In other words, POTS-over-cable is a 64 kbps synchronized digital signal, identical to what's carried on the phone companies' own ISDN, T, and SONET carriers, and is switched onmodified on and off the rest of the digital network unmodified. The A-D conversion happens in the settop box. It's like having your POTS phone at the switching center within wire-feet of the multiplexer. (The clocking is also good enough to encode analog signals from FAX and 56K computer modems. It has to be, as a side-effect of the need to time the upstream packets properly.)
POTS, on the other hand, is A-D converted at a central office or a "remote terminal" (in a box at the curb) and carried the rest of the way - blocks to miles - in one of a bundle of wires. This is subject to crosstalk, distortion (selective delay and attenuation of higher frequencies), and a number of other pathologies that lower the signal quality.
So it is not at ALL surprising that cable POTS signal quality beats telco POTS. Cable's signal is about as pristine as you can get.
(And VoIP isn't in the same ballpark, due to both compression and timing uncertainties.)
While the next model apparently will be a USB-powered plugin, the product documentation isn't clear about where THIS model gets its power.
Does it need a wall-wart, is it powered via power-over-ethernet, or what?