At extremely low bit rates or packet loss, Opus kicks the crap out everything else. In my experience, Opus has really good reproduction of a wide range of frequencies. MP3s like to remove highs and lows. such that symbols and deep base drums in acoustic music sounds garbled when listening side by side.
Good luck caching when most traffic is encrypted per session and requires it to be signed. One cannot transparently cache with HTTPS without a self-signed CA.
Take fog, multiply by 0.08, then 10.0 and you're left with.. Fog. CO2 does drop about 3% every 300 meters. After a kilometer, it would be 90% what it was at the ground. Mostly unchanged. I'm sure it's effectively opaque. I know if I was to look through several milometers of fog, I'd see nothing but fog.
Natural climate changes takes hundreds of thousands of years, about the amount of time it took for humans to become distinctly humans. Given another 400,000 years of climate change, I'm sure we could change again. The issue is what normally takes 400,000 years is taking 1-2 centuries. The only times this has happened was during mass extinction events. Maybe this time is different.
A safer(not looking at the Sun) analogy is fog. It's only ~42ppm, yet I can't see very far through it. 420ppm of CO2 is effectively a dense fog for IR.
What about professions where study and practice does not affect their ability to do or not do their job? Not all jobs requires knowledge or experience to be a master at.
Even Intel's own white paper cited a best case performance improvement with hyperthreading was ~30%
That was a typical performance improvement in a mixed workload, aka lots of multitasking. Worst case decrease was about 5% reduction and best case increase was over 100%, super linear. My cousin was an admin at a datacenter where he saw all kinds of work loads. Some where he disabled HT because of negative performance, and other where he got over a 50% improvement in system throughput.
We've reached the point where you can't really make it run faster with shorter pipelines without having a massively wide execution path, which lots of software won't benefit much from, or you have lots of simple cores with the same issue. Execution concurrency or thread concurrency.
We're all human beings with the same capacity to learn and contribute
"Same" capacity? Everyone can be a first place winner! I know what you meant, but I think you over simplified it a bit. Smart people are too rare to care about what kind of kinks they have or what their "background" is.
They're elitist in the worst sort of way. The untalented elite. There are a few gems of regulars, some of them even have books where they say that they have to dumb down their answers to make the "safe" on sites they they frequent. Really funny when they person in question is exclusively known for being on StackOverflow. So many times I've had to read through all of the answers, all the way down to the zero vote ones that some times predate all of the others. You can't trust the votes.
Don't get me wrong, I use stackoverflow, but I no longer participate. There is always someone else who can answer a question, and even if the wrong answer gets voted up, you can't do against the will of the mob. You will get punished in one way or the other. One needs to get very good at sifting.
Not saying you're wrong, but I think you're coming about it the wrong way. First off, point-to-point fiber the pretty standard now, even if getting hooked up to passive optics. Run one fiber from each dwelling unit back to the CO or "fiber hut" for high density situations. We already have 40gb/40gb WDM-PON, which gives 32 end points 1.25Gb/1.25Gb each. In the near future, up it from 40Gb to 320Gb. In the semi-near future, up it to ~3Tb/s. Yeah, that's right, multi-terabit PON is working in labs. Imagine the "dread" of sharing "only" 3Tb/s with 31 other houses.
Now that we've established that current tech can easily support dedicated symmetrical 1Gb to each dwelling unit, lets move on to the next choke point. The chassis, which I've seen called a "fiber aggregator", into which the fiber connects, have anywhere from 1.5Tb/s to 5Tb/s of backplane last I checked a few years back. On the upper end, that's 2,500 customers with non-blocking 1Gb/1Gb. Of course you need the uplinks to do that, but 100-400Gb uplinks are available. Just break up your customers into groups of 100-400 and you're good to go.
Next potential choke point is the core router. This is what the "fiber aggregator" typically plugs into. I don't know what's "normal", but I'm in a very poor small town and I found out my ISP has a core router with the slowest port being 100Gb and it has 32 ports, and does line rate. That's at least 3.2Tb/s full-duplex. I've read that you can buy 1Pb/s routers where the linecards currently have dual 400Gb ports and will eventually have linecard options of a single 1Tb port. Imagine that right now, commercially available, you can purchase a router that supports up to 1,000 400Gb or 500 1Tb ports.
But wait, there's more. DWMD currently allows 30Tb/s+ down a single fiber, two fibers for full duplex at that rate. What's more is the 600km ranges. You can even trade bandwidth for distance. 11,000km range for a bit over 1Tb/s. In the labs, they've got 1Pb/s fiber with 100Km range and 10Pb/s fiber with 40km range.
We currently have the ability of unlimited bandwidth up to the 1Gb range using commodity parts. Fiber is virtually unlimited. Getting beyond 1Gb is mostly an exercise of getting smaller fab processes freed up for networking instead of higher margin CPUs and other parts. Beyond 10Gb and being affordable to a typical person will take some time, but the technology is ready, it's just a matter of dropping the costs.
While the cost of all of this fiber infrastructure sounds expensive, ISPs like Sonic.Net claim infrastructure and transit only accounts for 1%-2% of your bill for any competent established ISP. That includes maintenance and regular upgrades. 98% of your internet bill is not even for the internet. What does it cover? Support and Marketing for the most part. What support you ask? You say you never call in or have anyone stop out. Well, you're paying for the support costs of all of the other people who do call in. People complain about "data hogs" and how they should be charged more. Data hogs are so cheap, the management overhead of monitoring them and charging them costs more than buying more bandwidth. Even then, their cost has to be less than the total cost, and that's only 1-2% of your bill. Maybe ISPs should start nickle and diming for support. then again, your mere existence as a customer carries a certain amount of fixed overhead. I know I would rather have a $40 bill instead of a $2 bill, where that time I need support, I don't get a $1,000 fee.
Depends on which type of dreams we're talking about. If you're talking about the kind you don't remember in any way, that's a normal part of the sleep cycle. If you're talking about lucid dreaming, that requires certain parts of the brain to be active, denying them the ability to rest.
I haven't had a nightmare of any kind since I learned to lucid dream. As soon as a dream get anywhere near scary, I become god like. This is actually the problem I have with all of my dreams. I immediately become an unmatched all powerful being and instantly bored. The only way to drag it out is to think of ways to make the dream fun, but that takes too much effort.
What's funny is the dream will keep trying to move forward, but I keep stopping it.
The only times I don't interfere with the dream is when it has something to do with a hard problem I'm working on in the real world. I can tell while I'm dreaming when it has something to do with problem solving. Something a person or a thing that I see in my dream reminds me of the problem I am stuck on, like how you see someone's doppelganger and they remind you of someone you know. Every time this has happened, I had the answer to my problem in the morning.
A recent development is like a person in my dream, who is part of the dream, yet somehow not. In the past, I day dreamed a lot and learned how to create persons in my dreams that I did not know what they were thinking. they always played a part of my dream, but I could not directly control them, I could only configure them. This has kind of returned, without any effort of my own, but they're in-between. Sometimes they'll narrate the dream in a way that's hilarious to me or even stop the dream and say something directly to me that is purely unexpected. Some of their witty remarks make for great stories at work that make other people laugh.
I've been lucid dreaming since I was 3. Had to learn to control my horrible nightmares. Gotten to the point that when I dream, I try to shut down the dream so I can sleep. I've dreamed nearly everything I've wanted to dream, and dragging out a dream just means I get less sleep.
What's your plan when there are no more jobs? Unemployment is becoming more and more of a norm, and in the future, most everyone will be unemployed. I would like to what what ethical plan(doesn't involve mass killings or imprisoning the majority of people because they need to steal to survive) you have.
Work has been an interesting topic in general. A substantiate portion of jobs are "busy jobs" that offer no or negative value, but have been created for the sole purpose of creating a job. These types of jobs are becoming more and more common. In your ideal world, everyone will just become a "ditch digger", digging, then filling them back in, then digging them again, just so your ideal of "everyone must have a job" is fulfilled.
Definitely trolling. No one writes perfect code, but I do see magnitudes difference among different people in both bugs per LOC and performance. Noticing bugs is the first battle. Too many times to count, I've had to dig into someone code for one reason or another, and I many times will find several bugs, many times critical. And many times this code has gone through several rounds of QA and code review over several months. Maybe I'm not the best to compare against, my co-worker says I have a super human attention to detail.
I've got 8 year old code that is still running in production that I wrote several months after graduation when I got my first job and I had 2 weeks of programming experience. It was my first project in the real world. I was to fix some bugs that several senior programming could not fix over several months. I re-wrote the entire ~5kloc program in ~2kloc, made it multi-threaded and wrote my own lock free data structures and algorithms. Over the past 8 years, only 2 bugs have ever been reported, both related to the threading, and both fixed without debugging.
Then there's some other programmers that have been programming for 10+ years and manage to write a simple 200loc application that manages to have 20+ bugs discovered over several months in prod, runs slowly, and somehow manages to gobble memory.
I rarely use the debugger. On numerous occasions, I've had co-workers come to me after days of debugging and a group of people trying to figure it out, and all I do is ask questions like what is responsible for X feature, show me the code so we can see what it guarantees, what guarantees do we have about the data inputs, etc. And just reading the code, walking through it with the other programmers, not only will I find the bug without ever running the code, but many times they find they independently notice the bug before I get to say anything. I know they're capable, seeing that they many times come to the same conclusions as me when presented with the same data, but for whatever reason they don't ask the questions I ask. They look at their code and say "what could it be", while I look at the code and say "what can't it be".
I look for guarantees in code and I use those guarantees to narrow it down. Single point of responsibly is another aspect that I heavily rely upon. If I can logically assert the state has certain guarantees at certain points in the code, there is only so many places for the bug, or any bug for that matter, to hide.
Many managed switches do Layer 3 routing and default to routing among the VLANs. I think that's a horrible default. So many times people have issues with asymmetric routes between overlapping subnets in different vlans because their switch is routing one way, and the actual router the other way, but the stateful firewall is having a spazz attack about only seeing half of the traffic.
No idea why people like to use overlapping subnets, especially in different broadcast domains.
Death is also a natural thing, but no need to make it happen several orders of magnitude faster.
The last time the climate changed a fraction as fast as it is now, a world wide extinction event occurred, nearly killed all life.This is faster. Even that time it was caused by a cataclysmic event, like a huge meteor or something. There is nothing normal about what is happening. Maybe it was bound to happen. Probably not. Hard to tell. No second Earth with no humans to compare against.
Correct, the Earth has done quite a few warm and cold cycles. What worries most is why this warming cycle is happening about 10,000x faster than any of the prior.
Can you coil the wire in a way that it generates a self-opposing magnetic field that limits the amps? I remember when I was young, my uncle was showing me how to make a light bulb dimmer by either using a resistor or a coil of wire, but the coil didn't generate heat like the resistor did and the light lasted quite a bit longer on the small capacitor. One issue was when the power finally ended, the magnetic field collapsed and caused a burst of very bright light, indicating a voltage spike.
Rated for 5000 cycles to be at their rated capacity. Their 100KWH bank starts off with ~120KWH of storage, and over 5,000 cycles, will be down to about 100KWH of storage. If you want to throw out your 100KWH battery bank because it can only hold 100KWH of power, that's on you. Use it another 10-20 years before it's not worth it.
I don't think you understand how durable these batteries are. Using the argument of "there a limited number of useful cycles" is like arguing we need to conserve our usage of atoms because there's a limited number of atoms in the Universe. These batteries already have up-to 20%+ unused capacity built in to act like unused space in an SSD for wear leveling. Li-ion lasts much much longer when not fully charged or discharged, and Tesla knows this. They under-spec the batteries and program the chargers to keep the charge in the optimal range to prolong the life of the batteries. Less so with the cars than the power walls, I assume because of weight.
Except the cost of the battery is a sunk cost if you need the car and the wear&tear to the battery will probably be less than 5% difference. The battery will probably out live the car and be replaced because it's no longer supported than too worn.
Very much so. I was using my limited experience to make my decision of using an average in discussion. In my case, I only use about 500KWH/m during the Summer, and I live in a duplex with almost no insulation and the AC unit is from the 90s, pre energy star. My AC runs constantly during the Summer. I also have several computers that run 24/7, and a TV or two that run 24/7 for the pets because of anxiety, better than drugs. Someone with a modern AC unit and decent insulation should use less power than I do and I use about 1/2 the average. I am not sure if the average includes all industrial usage averaged over the capita or the "average" person likes to have a chandelier with incandescent bulbs.
Slashdot Mirror
At extremely low bit rates or packet loss, Opus kicks the crap out everything else. In my experience, Opus has really good reproduction of a wide range of frequencies. MP3s like to remove highs and lows. such that symbols and deep base drums in acoustic music sounds garbled when listening side by side.
Good luck caching when most traffic is encrypted per session and requires it to be signed. One cannot transparently cache with HTTPS without a self-signed CA.
Take fog, multiply by 0.08, then 10.0 and you're left with.. Fog. CO2 does drop about 3% every 300 meters. After a kilometer, it would be 90% what it was at the ground. Mostly unchanged. I'm sure it's effectively opaque. I know if I was to look through several milometers of fog, I'd see nothing but fog.
Natural climate changes takes hundreds of thousands of years, about the amount of time it took for humans to become distinctly humans. Given another 400,000 years of climate change, I'm sure we could change again. The issue is what normally takes 400,000 years is taking 1-2 centuries. The only times this has happened was during mass extinction events. Maybe this time is different.
A safer(not looking at the Sun) analogy is fog. It's only ~42ppm, yet I can't see very far through it. 420ppm of CO2 is effectively a dense fog for IR.
What about professions where study and practice does not affect their ability to do or not do their job? Not all jobs requires knowledge or experience to be a master at.
Even Intel's own white paper cited a best case performance improvement with hyperthreading was ~30%
That was a typical performance improvement in a mixed workload, aka lots of multitasking. Worst case decrease was about 5% reduction and best case increase was over 100%, super linear. My cousin was an admin at a datacenter where he saw all kinds of work loads. Some where he disabled HT because of negative performance, and other where he got over a 50% improvement in system throughput.
We've reached the point where you can't really make it run faster with shorter pipelines without having a massively wide execution path, which lots of software won't benefit much from, or you have lots of simple cores with the same issue. Execution concurrency or thread concurrency.
We're all human beings with the same capacity to learn and contribute
"Same" capacity? Everyone can be a first place winner! I know what you meant, but I think you over simplified it a bit. Smart people are too rare to care about what kind of kinks they have or what their "background" is.
They're elitist in the worst sort of way. The untalented elite. There are a few gems of regulars, some of them even have books where they say that they have to dumb down their answers to make the "safe" on sites they they frequent. Really funny when they person in question is exclusively known for being on StackOverflow. So many times I've had to read through all of the answers, all the way down to the zero vote ones that some times predate all of the others. You can't trust the votes.
Don't get me wrong, I use stackoverflow, but I no longer participate. There is always someone else who can answer a question, and even if the wrong answer gets voted up, you can't do against the will of the mob. You will get punished in one way or the other. One needs to get very good at sifting.
Not saying you're wrong, but I think you're coming about it the wrong way. First off, point-to-point fiber the pretty standard now, even if getting hooked up to passive optics. Run one fiber from each dwelling unit back to the CO or "fiber hut" for high density situations. We already have 40gb/40gb WDM-PON, which gives 32 end points 1.25Gb/1.25Gb each. In the near future, up it from 40Gb to 320Gb. In the semi-near future, up it to ~3Tb/s. Yeah, that's right, multi-terabit PON is working in labs. Imagine the "dread" of sharing "only" 3Tb/s with 31 other houses. Now that we've established that current tech can easily support dedicated symmetrical 1Gb to each dwelling unit, lets move on to the next choke point. The chassis, which I've seen called a "fiber aggregator", into which the fiber connects, have anywhere from 1.5Tb/s to 5Tb/s of backplane last I checked a few years back. On the upper end, that's 2,500 customers with non-blocking 1Gb/1Gb. Of course you need the uplinks to do that, but 100-400Gb uplinks are available. Just break up your customers into groups of 100-400 and you're good to go.
Next potential choke point is the core router. This is what the "fiber aggregator" typically plugs into. I don't know what's "normal", but I'm in a very poor small town and I found out my ISP has a core router with the slowest port being 100Gb and it has 32 ports, and does line rate. That's at least 3.2Tb/s full-duplex. I've read that you can buy 1Pb/s routers where the linecards currently have dual 400Gb ports and will eventually have linecard options of a single 1Tb port. Imagine that right now, commercially available, you can purchase a router that supports up to 1,000 400Gb or 500 1Tb ports.
But wait, there's more. DWMD currently allows 30Tb/s+ down a single fiber, two fibers for full duplex at that rate. What's more is the 600km ranges. You can even trade bandwidth for distance. 11,000km range for a bit over 1Tb/s. In the labs, they've got 1Pb/s fiber with 100Km range and 10Pb/s fiber with 40km range.
We currently have the ability of unlimited bandwidth up to the 1Gb range using commodity parts. Fiber is virtually unlimited. Getting beyond 1Gb is mostly an exercise of getting smaller fab processes freed up for networking instead of higher margin CPUs and other parts. Beyond 10Gb and being affordable to a typical person will take some time, but the technology is ready, it's just a matter of dropping the costs.
While the cost of all of this fiber infrastructure sounds expensive, ISPs like Sonic.Net claim infrastructure and transit only accounts for 1%-2% of your bill for any competent established ISP. That includes maintenance and regular upgrades. 98% of your internet bill is not even for the internet. What does it cover? Support and Marketing for the most part. What support you ask? You say you never call in or have anyone stop out. Well, you're paying for the support costs of all of the other people who do call in. People complain about "data hogs" and how they should be charged more. Data hogs are so cheap, the management overhead of monitoring them and charging them costs more than buying more bandwidth. Even then, their cost has to be less than the total cost, and that's only 1-2% of your bill. Maybe ISPs should start nickle and diming for support. then again, your mere existence as a customer carries a certain amount of fixed overhead. I know I would rather have a $40 bill instead of a $2 bill, where that time I need support, I don't get a $1,000 fee.
Depends on which type of dreams we're talking about. If you're talking about the kind you don't remember in any way, that's a normal part of the sleep cycle. If you're talking about lucid dreaming, that requires certain parts of the brain to be active, denying them the ability to rest.
I haven't had a nightmare of any kind since I learned to lucid dream. As soon as a dream get anywhere near scary, I become god like. This is actually the problem I have with all of my dreams. I immediately become an unmatched all powerful being and instantly bored. The only way to drag it out is to think of ways to make the dream fun, but that takes too much effort.
What's funny is the dream will keep trying to move forward, but I keep stopping it.
The only times I don't interfere with the dream is when it has something to do with a hard problem I'm working on in the real world. I can tell while I'm dreaming when it has something to do with problem solving. Something a person or a thing that I see in my dream reminds me of the problem I am stuck on, like how you see someone's doppelganger and they remind you of someone you know. Every time this has happened, I had the answer to my problem in the morning.
A recent development is like a person in my dream, who is part of the dream, yet somehow not. In the past, I day dreamed a lot and learned how to create persons in my dreams that I did not know what they were thinking. they always played a part of my dream, but I could not directly control them, I could only configure them. This has kind of returned, without any effort of my own, but they're in-between. Sometimes they'll narrate the dream in a way that's hilarious to me or even stop the dream and say something directly to me that is purely unexpected. Some of their witty remarks make for great stories at work that make other people laugh.
I've been lucid dreaming since I was 3. Had to learn to control my horrible nightmares. Gotten to the point that when I dream, I try to shut down the dream so I can sleep. I've dreamed nearly everything I've wanted to dream, and dragging out a dream just means I get less sleep.
What's your plan when there are no more jobs? Unemployment is becoming more and more of a norm, and in the future, most everyone will be unemployed. I would like to what what ethical plan(doesn't involve mass killings or imprisoning the majority of people because they need to steal to survive) you have.
Work has been an interesting topic in general. A substantiate portion of jobs are "busy jobs" that offer no or negative value, but have been created for the sole purpose of creating a job. These types of jobs are becoming more and more common. In your ideal world, everyone will just become a "ditch digger", digging, then filling them back in, then digging them again, just so your ideal of "everyone must have a job" is fulfilled.
Definitely trolling. No one writes perfect code, but I do see magnitudes difference among different people in both bugs per LOC and performance. Noticing bugs is the first battle. Too many times to count, I've had to dig into someone code for one reason or another, and I many times will find several bugs, many times critical. And many times this code has gone through several rounds of QA and code review over several months. Maybe I'm not the best to compare against, my co-worker says I have a super human attention to detail.
I've got 8 year old code that is still running in production that I wrote several months after graduation when I got my first job and I had 2 weeks of programming experience. It was my first project in the real world. I was to fix some bugs that several senior programming could not fix over several months. I re-wrote the entire ~5kloc program in ~2kloc, made it multi-threaded and wrote my own lock free data structures and algorithms. Over the past 8 years, only 2 bugs have ever been reported, both related to the threading, and both fixed without debugging.
Then there's some other programmers that have been programming for 10+ years and manage to write a simple 200loc application that manages to have 20+ bugs discovered over several months in prod, runs slowly, and somehow manages to gobble memory.
I rarely use the debugger. On numerous occasions, I've had co-workers come to me after days of debugging and a group of people trying to figure it out, and all I do is ask questions like what is responsible for X feature, show me the code so we can see what it guarantees, what guarantees do we have about the data inputs, etc. And just reading the code, walking through it with the other programmers, not only will I find the bug without ever running the code, but many times they find they independently notice the bug before I get to say anything. I know they're capable, seeing that they many times come to the same conclusions as me when presented with the same data, but for whatever reason they don't ask the questions I ask. They look at their code and say "what could it be", while I look at the code and say "what can't it be".
I look for guarantees in code and I use those guarantees to narrow it down. Single point of responsibly is another aspect that I heavily rely upon. If I can logically assert the state has certain guarantees at certain points in the code, there is only so many places for the bug, or any bug for that matter, to hide.
Many managed switches do Layer 3 routing and default to routing among the VLANs. I think that's a horrible default. So many times people have issues with asymmetric routes between overlapping subnets in different vlans because their switch is routing one way, and the actual router the other way, but the stateful firewall is having a spazz attack about only seeing half of the traffic.
No idea why people like to use overlapping subnets, especially in different broadcast domains.
Death is also a natural thing, but no need to make it happen several orders of magnitude faster.
The last time the climate changed a fraction as fast as it is now, a world wide extinction event occurred, nearly killed all life.This is faster. Even that time it was caused by a cataclysmic event, like a huge meteor or something. There is nothing normal about what is happening. Maybe it was bound to happen. Probably not. Hard to tell. No second Earth with no humans to compare against.
Correct, the Earth has done quite a few warm and cold cycles. What worries most is why this warming cycle is happening about 10,000x faster than any of the prior.
80% of what I watch on Netflix is "Netflix original", but 80% of what I watch is not on Netflix.
Can you coil the wire in a way that it generates a self-opposing magnetic field that limits the amps? I remember when I was young, my uncle was showing me how to make a light bulb dimmer by either using a resistor or a coil of wire, but the coil didn't generate heat like the resistor did and the light lasted quite a bit longer on the small capacitor. One issue was when the power finally ended, the magnetic field collapsed and caused a burst of very bright light, indicating a voltage spike.
Rated for 5000 cycles to be at their rated capacity. Their 100KWH bank starts off with ~120KWH of storage, and over 5,000 cycles, will be down to about 100KWH of storage. If you want to throw out your 100KWH battery bank because it can only hold 100KWH of power, that's on you. Use it another 10-20 years before it's not worth it.
I don't think you understand how durable these batteries are. Using the argument of "there a limited number of useful cycles" is like arguing we need to conserve our usage of atoms because there's a limited number of atoms in the Universe. These batteries already have up-to 20%+ unused capacity built in to act like unused space in an SSD for wear leveling. Li-ion lasts much much longer when not fully charged or discharged, and Tesla knows this. They under-spec the batteries and program the chargers to keep the charge in the optimal range to prolong the life of the batteries. Less so with the cars than the power walls, I assume because of weight.
Except the cost of the battery is a sunk cost if you need the car and the wear&tear to the battery will probably be less than 5% difference. The battery will probably out live the car and be replaced because it's no longer supported than too worn.
one has to be careful talking about averages
Very much so. I was using my limited experience to make my decision of using an average in discussion. In my case, I only use about 500KWH/m during the Summer, and I live in a duplex with almost no insulation and the AC unit is from the 90s, pre energy star. My AC runs constantly during the Summer. I also have several computers that run 24/7, and a TV or two that run 24/7 for the pets because of anxiety, better than drugs. Someone with a modern AC unit and decent insulation should use less power than I do and I use about 1/2 the average. I am not sure if the average includes all industrial usage averaged over the capita or the "average" person likes to have a chandelier with incandescent bulbs.