What matters are the *rates*. For example, what *percentage* of residents are living in poverty. California has the #1 highest poverty rate in the country, according to the US Census Bureau.
Well, that depends on which metric you use. If you go by the official poverty rate, California is actually #30, behind 29 other states. If you go by the *supplemental* poverty rate, it is at #1. And if you look at the difference between those two metrics, you will conclude that the primary reason for that difference is the cost of housing in California. It has nothing to do with government programs and everything to do with the tech sector driving up the cost of housing to bats**t crazy levels for a large chunk of the state, and the entertainment industry doing the same for another large chunk of the state.
In the USA we lose less than 5% (last estimate I saw was 3%) in transmission, using traditional high-tension lines. We lose less than half as much energy in transmission as you imagine.
Sorry, I did the math. Your numbers are based on current distances, in a country where most power production is much, much closer to the consumer than what I proposed (putting nuclear plants in the dead center of the country). That 4–8% loss is a best-case scenario. Worst case is more like 20–30%.
With superconductors, that problem is solved. Just build superconducting underground power transmission lines from the middle of nowhere to where the power is needed.
Transmission losses are not that big a deal even with superconductors. Plus you talk about high temp superconductors as if that is something we can suddenly waive into existence.
I never mentioned high-temperature superconductors at all. And of course transmission losses are not a big deal with superconductors. That's why you use them. Did you mean "without"? Because if so, then yes, they are a big deal when you're talking about running that much power over hundreds or even thousands of miles. Say you bring 100 GW of nuclear plant capacity online in North Dakota, and your average distance to the customers is 800 miles (that's half the distance to the coast; I'm conservatively assuming a more even distribution of population that actually exists, so this is probably a very low estimate). With normal cables, you're talking about 4–8% loss.
Now whether the cost of the superconducting lines would exceed the cost of those losses is another question, but assuming more capacity comes online over time, and assuming that they continue to replace old plants with new ones such that the lines are a century-long investment, that seems almost certain.
We're certainly not going to use the sorts of superconductors available to us today.
I'm glad you're not in charge of our infrastructure. There are several superconducting transmission cables are in active use right now.
Won't happen. Politically it's just a non-starter for a variety of well understood reasons. People are scared of nuclear power regardless of whether or not they should be.
With superconductors, that problem is solved. Just build superconducting underground power transmission lines from the middle of nowhere to where the power is needed. After all, the United States has vast tracts of land with approximately zero people (with some small epsilon) living there, in states like Wyoming, Montana, North Dakota, South Dakota, New Mexico, etc. Let's put our nuclear power plants there, so that if things go wrong, we can just bomb and bury them.
Better yet, put them half a mile underground so that when things go wrong, we don't have to do anything beyond putting up a fence and collapsing the tunnel entrance.
Showing up in a conference room with a firebreathing 15" display and running 3D apps means that you plan to ignore the presenter.
Xcode is an IDE with a compiler toolchain, not a 3D app. I show up in the conference room with that laptop because that is the laptop I use at work to get work done.
Anyway, display efficiency keeps improving. OLED with 100 lumens per watt. [cnet.com] That's a lot, ever pointed a 100 lumen flashlight in your face? The real issue is just as I said, your laptop is a dinosaur.
Wrong again. OLED is nice in theory. In practice, for laptop use, it tends to use *more* power than LED-backlit LCDs, because computer screens tend to display large amounts of white (every web page, word processing document, etc.), which means they tend towards their worst-case power utilization, rather than their best as occurs on devices that are mostly used for media consumption. Between that and yield problems (and/or artificially constrained supply), OLD panels on laptops are more trouble than they are worth. And that's not even taking into account the significantly reduced panel lifespan when used to display seas of white. Those reasons could, perhaps, explain why laptop manufacturers that switched to OLED previously are now actually switching BACK to LCD panels in their newer models.
Larger screens require more power and a beefier GPU, which also requires more power.
You need power budget for external devices.
You're going to do much more CPU-intensive and GPU-intensive work with a laptop than a tablet, so you need a bigger battery, which means more wattage to avoid insanely long charge times.
You need lots of RAM for working with larger data sets (in this case, 4x the RAM of an iPad Pro), which draws a lot of power.
You need more disk storage for storing those larger data sets, which draws more power.
Add all of these things up, and even if you switched to an ARM CPU, you wouldn't really save that much power. For maximum charging, the MBP can draw up to 87W. It can run at full speed with slower charging at 60W. It can run at slightly reduced speed (no burst mode) at 45W (CPU TDP). So half the wattage is for charging speed and that extra little bit of performance. But below 45W, you aren't going to be able to power the device.
If you moved from Intel to ARM, you would be moving from a chip that can't be throttled enough to run below about 30W to a chip that currently draws 15W and is only two-thirds as fast and has a much less capable DRAM controller and northbridge. Bring that up to equivalent capabilities, and I doubt you'll save more than five watts, at a tremendous R&D cost.
The problem is the laptop, not the charger. Wintel has just fallen pitifully behind in terms of form factor. Displays are way more efficient than they used to be, processors too unless you are paying the Intel CISC tax. Are you running Fortnite on your laptop in the conference room? No? Then it is pathetic you can't run it on 15 watts.
The screen alone draws more than 15 Watts, I believe.
I'd love to, but other than shopping directly from the Benson Approved site, how can anyone know it's a quality cable that is "quality?" Most folks won't ever even know there's a difference. The huge variety in quality indicates that something in the USB-C spec is difficult to meet.
The huge variation in quality indicates that most Chinese manufacturers will build the cheapest cables that they can get away with, and if that means you're capped at two amperes instead of three, as long as 90% of users don't notice, they'll still get four out of five stars, so why bother to support fast charging for devices that prefer higher amperage, when that means using a thicker wire?
The bugs are not at the UI level. The "bugs" are exploitable code that is never actually executed. Black hats would find them when scanning the code, and try to exploit them, but the exploits wouldn't work because the code never executes.
Or, for Internet-based services, the bugs can be non-exploitable code that will fool fuzzing attacks into thinking that it is wreaking havoc, when in fact it is doing exactly what it was programmed to do.
For example, if your service can detect an inappropriate number of failed logins from a specific IP, you *could* make it throw an error that makes it obvious that the IP address is burned, or you could start delaying each response for an exponentially increasing amount of time so that it looks like the attack is working, but breaking the server, or you could just stop sending requests to your database entirely, and replace them with usleep() calls for about as long as a normal auth request would take, so that the bot is kept busy, but without actually expending resources meaningfully.
Or if your service stores data in a SQL database, it should always store data correctly in a way that prevents injection, but it could *also* detect obvious attempts at SQL injection and return data that makes it look like something weird is going on under the hood, so that attackers continue trying to find ways to attack working code, rather than looking for other attack vectors.
Or if your service stores data in JSON format, it could just reject malformed JSON, or you could make it deliberately drop the connection in a way that suggests a crash might have happened. (Bonus points if you provide some sort of unique server ID and can make that server ID "disappear" for a while whenever you get attacked, to fully simulate a crash.)
You could also create a fake admin account with no actual password that bots will attack continuously. You could create fake endpoints that sound like they do something juicy, but in fact, always return errors. And so on.
So is USB PD. You just have to have a PD-aware micro-USB cable (and, obviously, a PD-capable micro-USB device). That said, why would anyone want to continue using micro-B on new devices now that USB-C exists?
Even among the non-Apple devices there are still issues. The USB-Micro standard is fragile, uni-directional and has skinny-ass wires that can't cary much current for fast charging. The USB-C standard is anything but standard [digitaltrends.com] with a mixed bag of features and compatibility from device to device.
IMO, those problems are mostly overblown.
Yes, there are Thunderbolt cables that look like USB but don't pass USB data, but unless you're one of those rare people who actually uses Thunderbolt, you'll never encounter one, so that's not a meaningful reason to be concerned.
And yes, it is possible for a crappy cable to reduce your charging speed by limiting the current. Buy quality cables and you won't have this problem.
But realistically, so far, the only real problems I've had were caused by plugging blindly into a USB-C connection that goes through some random conference room table at work, only to find out that it is a 15W cell phone charger that obviously won't power a laptop. This, of course, doesn't affect charging phones at all, and has nothing to do with USB-C except insofar as my laptop just happens to use the same charge connector as cell phones.
Hearthstone, too. If I'm killing time for an afternoon, I'll play on the phone, and I keep it continuously connected to power, because as you say, the batteries are an order of magnitude too small to be practical for gaming.
You don't need a liberal arts degree to effectively write software for composing music, and you don't need to be an artist to create digital painting software
Also, when robots, machine learning, outsourcing, or H1-Bs replace people who only know how to do one particular thing, those folks are screwed. People who instead focus on breadth with only limited depth end up being in a much better position to acquire just enough depth to start doing something different.
And he's not wrong, though the examples don't really work for me. Why do I think liberal arts are important for CS majors? Because software has to be used by people who don't write software. Musicians need software for creating music. Artists need software for creating art. Writers need software for writing. Programmers who also understand those secondary fields are likely to be better at creating such software than programmers who don't.
And the same is true for other, non-liberal-arts fields; pretty much every science field has some sort of software for collecting data, visualizing data, simulating complex interactions like protein folding, etc. (And arguably, data visualization is an art unto itself, upon which all sciences depend to some degree.)
I think every computer programmer should have at least a minor in a non-tech field, if not a second major, whether in a liberal arts field or a science or something else entirely, if only because of the opportunities for specialization that such outside interests offer. Also, if computers get to the point where software writes itself, at least they'll have something to fall back on.:-D
Is it still the Left's stance that Pai rolling back Title-II NN was because of corruption?
Corruption or merely tending to favor the same positions that you held while working in the industry, which are skewed by the types of companies you worked for. Either way.
Could the Left have been...*gasp!*...wrong??
Unlikely. As someone else pointed out elsewhere in the thread, the very telecom company that Pai used to work for, Verizon, will benefit greatly from this decision during their 5G rollout, because it will require adding orders of magnitude more cells in densely populated areas, which must be fed with fiber, and Verizon isn't the primary wire-line provider in much of its coverage area.
They didn't want the rules when it benefitted Google over them. They do want the rules when it benefits them over somebody else. Simple enough.
Zebras don't change their stripes. Ajit Pai is an asshole, clearly and objectively, therefore someone must have stopped bribing him in some way to get this 'ruling'. The question remains: who?
Of course, Google Fiber basically stopped expanding more than a year ago. It may be that Pai's FCC stopped resisting this change only because his major-telecom overlords concluded that the threat has passed.
I the driver where to grab the wheel to adjust steering - BUT keeps the feet hovering above the pedals without pushing them (Yet) - would this disengage only the autopilot steering ? While keeping engage the usual distance keeping / emergency autonomous braking features (FCAS) ?
If you press the gas pedal too much, you'll disable automatic braking, and nothing else. If you touch the brake, it disengages everything. (This is, IMO, a bug; it is too easy to accidentally tap it on a curve and have the wheels suddenly straighten. But I digress.)
Also, how good are the FCAS in city ?
The non-steering stuff works fairly well in a city, though you'll have to manually stop at stop signs and traffic lights (unless you're the second car back at a light).
Most drivers never tried to use AutoSteer in places where it mattered. Those of us who drive CA-17 and other similar roads regularly were the only ones swearing about it, but we were very much swearing.
And I made fairly public comments here on Slashdot and Tesla discussion boards about it, noting that it was unlikely that the software being used for AutoSteer at the time was actually the base code for their self-driving feature, but rather temporary code that they were patching and tweaking minimally, in maintenance mode, until they could get the real software out. Unsurprisingly, I was correct in that statement.:-)
They massively redesigned their machine learning setup in an update that came out earlier this year, and the difference is night and day.
I also made comments a few months back saying that it was unlikely that AP2 would actually be the hardware used for self-driving, and that they would probably be replacing it. That also proved to be correct (just now).
The problem is not that people aren't telling users the truth, but rather that the average person doesn't notice the problems, because the average person isn't sufficiently technically inclined to know what to look for. Those of us who are sufficiently skilled in software engineering to recognize the issues get ignored, because people automatically assume that the negative opinions are the result of rare problems, rather than because of common problems that few people notice. And the result is that outside of a few Tesla message boards, everybody thinks that everything is sunshine and roses, where in fact, there are still a lot of problems.
The other thing that most people don't realize is how often your car will need repairs during the first year for stupid assembly mistakes and lack of proper quality control, but that's another rant for another day.
Maybe after a point, but up until that point, the main risk is reacting too slowly. Ask anybody with an AP2 Tesla how well it handled curves prior to earlier this year. Of they don't use the word "lag", they don't know software, and if their eyes don't bug out in abject terror, they don't know how to drive.
Basically, it had (and still has, to a lesser extent) trouble with lane keeping, because its reactions lagged behind reality, and it started turning way too late, resulting in uncomfortable turns, getting dangerously close to barriers and center lines, etc. This is better in current versions, but I still get scared enough to take manual control a couple of times per day.
So right now, performance is still their main problem. This is a very welcome announcement.
Slashdot Mirror
Well, that depends on which metric you use. If you go by the official poverty rate, California is actually #30, behind 29 other states. If you go by the *supplemental* poverty rate, it is at #1. And if you look at the difference between those two metrics, you will conclude that the primary reason for that difference is the cost of housing in California. It has nothing to do with government programs and everything to do with the tech sector driving up the cost of housing to bats**t crazy levels for a large chunk of the state, and the entertainment industry doing the same for another large chunk of the state.
Sorry, I did the math. Your numbers are based on current distances, in a country where most power production is much, much closer to the consumer than what I proposed (putting nuclear plants in the dead center of the country). That 4–8% loss is a best-case scenario. Worst case is more like 20–30%.
I never mentioned high-temperature superconductors at all. And of course transmission losses are not a big deal with superconductors. That's why you use them. Did you mean "without"? Because if so, then yes, they are a big deal when you're talking about running that much power over hundreds or even thousands of miles. Say you bring 100 GW of nuclear plant capacity online in North Dakota, and your average distance to the customers is 800 miles (that's half the distance to the coast; I'm conservatively assuming a more even distribution of population that actually exists, so this is probably a very low estimate). With normal cables, you're talking about 4–8% loss.
Now whether the cost of the superconducting lines would exceed the cost of those losses is another question, but assuming more capacity comes online over time, and assuming that they continue to replace old plants with new ones such that the lines are a century-long investment, that seems almost certain.
I'm glad you're not in charge of our infrastructure. There are several superconducting transmission cables are in active use right now.
With superconductors, that problem is solved. Just build superconducting underground power transmission lines from the middle of nowhere to where the power is needed. After all, the United States has vast tracts of land with approximately zero people (with some small epsilon) living there, in states like Wyoming, Montana, North Dakota, South Dakota, New Mexico, etc. Let's put our nuclear power plants there, so that if things go wrong, we can just bomb and bury them.
Better yet, put them half a mile underground so that when things go wrong, we don't have to do anything beyond putting up a fence and collapsing the tunnel entrance.
Yeah, this kind of makes me want to trade my Tesla in for a Hummer.
Xcode is an IDE with a compiler toolchain, not a 3D app. I show up in the conference room with that laptop because that is the laptop I use at work to get work done.
Wrong again. OLED is nice in theory. In practice, for laptop use, it tends to use *more* power than LED-backlit LCDs, because computer screens tend to display large amounts of white (every web page, word processing document, etc.), which means they tend towards their worst-case power utilization, rather than their best as occurs on devices that are mostly used for media consumption. Between that and yield problems (and/or artificially constrained supply), OLD panels on laptops are more trouble than they are worth. And that's not even taking into account the significantly reduced panel lifespan when used to display seas of white. Those reasons could, perhaps, explain why laptop manufacturers that switched to OLED previously are now actually switching BACK to LCD panels in their newer models.
It really wouldn't matter what chipset they use.
Add all of these things up, and even if you switched to an ARM CPU, you wouldn't really save that much power. For maximum charging, the MBP can draw up to 87W. It can run at full speed with slower charging at 60W. It can run at slightly reduced speed (no burst mode) at 45W (CPU TDP). So half the wattage is for charging speed and that extra little bit of performance. But below 45W, you aren't going to be able to power the device.
If you moved from Intel to ARM, you would be moving from a chip that can't be throttled enough to run below about 30W to a chip that currently draws 15W and is only two-thirds as fast and has a much less capable DRAM controller and northbridge. Bring that up to equivalent capabilities, and I doubt you'll save more than five watts, at a tremendous R&D cost.
My laptop is a 15" MacBook Pro, so no, it isn't.
The screen alone draws more than 15 Watts, I believe.
The huge variation in quality indicates that most Chinese manufacturers will build the cheapest cables that they can get away with, and if that means you're capped at two amperes instead of three, as long as 90% of users don't notice, they'll still get four out of five stars, so why bother to support fast charging for devices that prefer higher amperage, when that means using a thicker wire?
Or, for Internet-based services, the bugs can be non-exploitable code that will fool fuzzing attacks into thinking that it is wreaking havoc, when in fact it is doing exactly what it was programmed to do.
For example, if your service can detect an inappropriate number of failed logins from a specific IP, you *could* make it throw an error that makes it obvious that the IP address is burned, or you could start delaying each response for an exponentially increasing amount of time so that it looks like the attack is working, but breaking the server, or you could just stop sending requests to your database entirely, and replace them with usleep() calls for about as long as a normal auth request would take, so that the bot is kept busy, but without actually expending resources meaningfully.
Or if your service stores data in a SQL database, it should always store data correctly in a way that prevents injection, but it could *also* detect obvious attempts at SQL injection and return data that makes it look like something weird is going on under the hood, so that attackers continue trying to find ways to attack working code, rather than looking for other attack vectors.
Or if your service stores data in JSON format, it could just reject malformed JSON, or you could make it deliberately drop the connection in a way that suggests a crash might have happened. (Bonus points if you provide some sort of unique server ID and can make that server ID "disappear" for a while whenever you get attacked, to fully simulate a crash.)
You could also create a fake admin account with no actual password that bots will attack continuously. You could create fake endpoints that sound like they do something juicy, but in fact, always return errors. And so on.
So is USB PD. You just have to have a PD-aware micro-USB cable (and, obviously, a PD-capable micro-USB device). That said, why would anyone want to continue using micro-B on new devices now that USB-C exists?
IMO, those problems are mostly overblown.
Yes, there are Thunderbolt cables that look like USB but don't pass USB data, but unless you're one of those rare people who actually uses Thunderbolt, you'll never encounter one, so that's not a meaningful reason to be concerned.
And yes, it is possible for a crappy cable to reduce your charging speed by limiting the current. Buy quality cables and you won't have this problem.
But realistically, so far, the only real problems I've had were caused by plugging blindly into a USB-C connection that goes through some random conference room table at work, only to find out that it is a 15W cell phone charger that obviously won't power a laptop. This, of course, doesn't affect charging phones at all, and has nothing to do with USB-C except insofar as my laptop just happens to use the same charge connector as cell phones.
USB 3 was also originally going to be optical. These mistakes have a tendency to correct themselves.
Hearthstone, too. If I'm killing time for an afternoon, I'll play on the phone, and I keep it continuously connected to power, because as you say, the batteries are an order of magnitude too small to be practical for gaming.
Need, no, but it helps.
Also, when robots, machine learning, outsourcing, or H1-Bs replace people who only know how to do one particular thing, those folks are screwed. People who instead focus on breadth with only limited depth end up being in a much better position to acquire just enough depth to start doing something different.
And he's not wrong, though the examples don't really work for me. Why do I think liberal arts are important for CS majors? Because software has to be used by people who don't write software. Musicians need software for creating music. Artists need software for creating art. Writers need software for writing. Programmers who also understand those secondary fields are likely to be better at creating such software than programmers who don't.
And the same is true for other, non-liberal-arts fields; pretty much every science field has some sort of software for collecting data, visualizing data, simulating complex interactions like protein folding, etc. (And arguably, data visualization is an art unto itself, upon which all sciences depend to some degree.)
I think every computer programmer should have at least a minor in a non-tech field, if not a second major, whether in a liberal arts field or a science or something else entirely, if only because of the opportunities for specialization that such outside interests offer. Also, if computers get to the point where software writes itself, at least they'll have something to fall back on. :-D
Add three or four zeroes and they can serve as an example for others.
Corruption or merely tending to favor the same positions that you held while working in the industry, which are skewed by the types of companies you worked for. Either way.
Unlikely. As someone else pointed out elsewhere in the thread, the very telecom company that Pai used to work for, Verizon, will benefit greatly from this decision during their 5G rollout, because it will require adding orders of magnitude more cells in densely populated areas, which must be fed with fiber, and Verizon isn't the primary wire-line provider in much of its coverage area.
They didn't want the rules when it benefitted Google over them. They do want the rules when it benefits them over somebody else. Simple enough.
FTFY
Of course, Google Fiber basically stopped expanding more than a year ago. It may be that Pai's FCC stopped resisting this change only because his major-telecom overlords concluded that the threat has passed.
Even a stopped clock is right twice a day.
If you press the gas pedal too much, you'll disable automatic braking, and nothing else. If you touch the brake, it disengages everything. (This is, IMO, a bug; it is too easy to accidentally tap it on a curve and have the wheels suddenly straighten. But I digress.)
The non-steering stuff works fairly well in a city, though you'll have to manually stop at stop signs and traffic lights (unless you're the second car back at a light).
Most drivers never tried to use AutoSteer in places where it mattered. Those of us who drive CA-17 and other similar roads regularly were the only ones swearing about it, but we were very much swearing.
And I made fairly public comments here on Slashdot and Tesla discussion boards about it, noting that it was unlikely that the software being used for AutoSteer at the time was actually the base code for their self-driving feature, but rather temporary code that they were patching and tweaking minimally, in maintenance mode, until they could get the real software out. Unsurprisingly, I was correct in that statement. :-)
They massively redesigned their machine learning setup in an update that came out earlier this year, and the difference is night and day.
I also made comments a few months back saying that it was unlikely that AP2 would actually be the hardware used for self-driving, and that they would probably be replacing it. That also proved to be correct (just now).
The problem is not that people aren't telling users the truth, but rather that the average person doesn't notice the problems, because the average person isn't sufficiently technically inclined to know what to look for. Those of us who are sufficiently skilled in software engineering to recognize the issues get ignored, because people automatically assume that the negative opinions are the result of rare problems, rather than because of common problems that few people notice. And the result is that outside of a few Tesla message boards, everybody thinks that everything is sunshine and roses, where in fact, there are still a lot of problems.
The other thing that most people don't realize is how often your car will need repairs during the first year for stupid assembly mistakes and lack of proper quality control, but that's another rant for another day.
Maybe after a point, but up until that point, the main risk is reacting too slowly. Ask anybody with an AP2 Tesla how well it handled curves prior to earlier this year. Of they don't use the word "lag", they don't know software, and if their eyes don't bug out in abject terror, they don't know how to drive.
Basically, it had (and still has, to a lesser extent) trouble with lane keeping, because its reactions lagged behind reality, and it started turning way too late, resulting in uncomfortable turns, getting dangerously close to barriers and center lines, etc. This is better in current versions, but I still get scared enough to take manual control a couple of times per day.
So right now, performance is still their main problem. This is a very welcome announcement.