A "degree" is only as good as the school that stands behind it. Degrees from degree mills won't get your foot in the door nearly as well as reputable colleges or unis.
"Get a masters in business by only spending 1 hour per week for 1 year from our online course!" Good luck getting a job with that.
I used an extreme example, but in the normal world, it seems most people don't properly understand their much simpler problem domains, then blame overly complex designs. I will agree that for every time this happens, it probably happens several times that someone does overly complicate a design, but that's is probably because they don't fully understand a problem domain or have a good ability to abstract ideas.
I had very little busy work in college. Most of my learning was in class having very intellectual discussions on pros, cons, theories, ideas, reasons, corner cases, devil's advocates, and other fun things. This was in almost every class, from biology, to communications, to history, system security, to data structures, to advanced database designs, to system analysis and design.
Lots of theory, lots of real world cases, lots of brain storming. Most of my learning had little to do with books or homework. Almost all of my learning was in-class discussions.
Everyone, stop doing particle research because someone somewhere doesn't understand it. If you have a complex problem that requires a complex solution, you don't dumb it down and expect things to work correctly. If the "lower level" individual can't even understand the problem domain because it's above their potential ability, then they shouldn't be there.
If you can't understand the problem domain, then GTFO, don't dumb it down until everyone can understand it.
Totally worth placing a $50,000 of piece of equipment in the field that needs power, UPS, cooling, heating, compared to just $500 of passive fiber. They may save money on not sending a fiber tech into a house, but they lose everywhere else. with fiber, a single GPON chassis can serve up to 5,000 people within 20km. These DSL setups are more like serving 100 people within 250m. But hey, you don't need to trench fiber. With Active Ethernet, you can serve about 500 people within 80km on single chassis. No need for expensive equipment in the field that needs to weather the elements, everything can be back in your datacenter.
While you have a lot of bandwidth, you still need some very expensive head units to manage each virtual channel. More limited by money than by physics. At some point it becomes cheaper to just use fiber, even if you can get near the same performance. And that's ignoring all of the other issues with copper vs fiber.
40mb/s Per virtual channel. A single 6mhz physical channel can support several virtual channels via CDMA. The official spec allows a single device to lock onto virtual channels within the same physical channel. The example given was to lock 8 40mb virtual channels within the same 6mhz physical channel, giving an aggregate of 320mb/s, and you can get near those speeds.
It is recommended to use separate physical channels as each virtual channel in use increases the floor noise, and a wide spectrum to use allows for more resiliency. But if you're a cable provider with few free channels, you can push over 1gb/s over 2-3 6mhz channels.
Upstream not allow for a single device to lock several virtual channels in the same physical channel for reasons like higher noise and less power because of the ranges the upstream uses.
Power consumption scales linearly with frequency and with the square of the voltage. "high end" low-power CPUs are not only low frequency, but low voltage. They are more rare than your more general CPUs that are between high-end low-power and high-end higher-performance. Lowering the voltage not only reduces power lost by oscillation of a conductor with capacitance(the square scaling), but also electrical leakage.
I wouldn't jump the gun too fast. FreeBSD just had a new network API added in 10 that doubles the packet throughput of legacy when using a wrapper and over 10x the throughput when using native. A single core ATOM cpu could handle full duplex routing of a 10gb interface while running in user mode, outside of the kernel.
The new interface will allow crazy low overhead for usermode programs to access the NICs.
FreeBSD 10 has a way to do user-mode firewall with performance that allows a dual-core 450mhz CPU to handle 10gb/s of 64byte packets. It involves no context switches and zero-copy of the packet-data.
At one point in the past few years, one of OCZ's popular SSD lines had a 20% failure rate. Most of their other products have consistently been about double to triple the failure rate of everyone else's.
Heck yeah. At least 10x more efficient than silicon per clock cycle, supports speeds into the 100ghz ranges, and can handle temperatures well over 200c.
RAM vs cache are two different things. Cache scales something like O(n^2), but is a small number, while RAM scales O(1), but is a large number. The latency of a huge sram cache would be horrendous. Very very generalized.
The retooling is crazy expensive, in the order of tens of billions. If you can get more money from your old machinery while refining your next gen, then to keep from competing with yourself.
Blizzard is like NewEgg, there will always be someone who got a bad experience. Given a large user base, it's a statistical guarantee, it just sucks when it's you.
I had a relatively small obsession with porn around 11 or 12, then I got sucked into Quake Team Fortress and anime. Porn was new and exciting, but ultimately boring. If I was to do a do-over with my life, it would be the amount of gaming without breaks I did as a kid.
I got a new USR modem.. ohh, conflicts with my mouse's IRQ.. derpa-da-derp.. There... Ohh, a printer, lets hook that up also. LPT1 conflicts with my mouse again, but there's not many options left.. hmmmm.. If I change the mouse to this IRQ, the modem to this IRQ, then the printer can be on this one! Woah... that sucked, hope I don't get an external zip drive, I may not have enough IRQs.
Slashdot Mirror
That wasn't hydrogen, that was a mixture of hydrogen and oxygen. Pure hydrogen does not explode.
But the Big Bang was an event. How does an event-less system have an event that creates events?
A "degree" is only as good as the school that stands behind it. Degrees from degree mills won't get your foot in the door nearly as well as reputable colleges or unis.
"Get a masters in business by only spending 1 hour per week for 1 year from our online course!" Good luck getting a job with that.
I used an extreme example, but in the normal world, it seems most people don't properly understand their much simpler problem domains, then blame overly complex designs. I will agree that for every time this happens, it probably happens several times that someone does overly complicate a design, but that's is probably because they don't fully understand a problem domain or have a good ability to abstract ideas.
I had very little busy work in college. Most of my learning was in class having very intellectual discussions on pros, cons, theories, ideas, reasons, corner cases, devil's advocates, and other fun things. This was in almost every class, from biology, to communications, to history, system security, to data structures, to advanced database designs, to system analysis and design.
Lots of theory, lots of real world cases, lots of brain storming. Most of my learning had little to do with books or homework. Almost all of my learning was in-class discussions.
Everyone, stop doing particle research because someone somewhere doesn't understand it. If you have a complex problem that requires a complex solution, you don't dumb it down and expect things to work correctly. If the "lower level" individual can't even understand the problem domain because it's above their potential ability, then they shouldn't be there.
If you can't understand the problem domain, then GTFO, don't dumb it down until everyone can understand it.
Totally worth placing a $50,000 of piece of equipment in the field that needs power, UPS, cooling, heating, compared to just $500 of passive fiber. They may save money on not sending a fiber tech into a house, but they lose everywhere else. with fiber, a single GPON chassis can serve up to 5,000 people within 20km. These DSL setups are more like serving 100 people within 250m. But hey, you don't need to trench fiber. With Active Ethernet, you can serve about 500 people within 80km on single chassis. No need for expensive equipment in the field that needs to weather the elements, everything can be back in your datacenter.
While you have a lot of bandwidth, you still need some very expensive head units to manage each virtual channel. More limited by money than by physics. At some point it becomes cheaper to just use fiber, even if you can get near the same performance. And that's ignoring all of the other issues with copper vs fiber.
40mb/s Per virtual channel. A single 6mhz physical channel can support several virtual channels via CDMA. The official spec allows a single device to lock onto virtual channels within the same physical channel. The example given was to lock 8 40mb virtual channels within the same 6mhz physical channel, giving an aggregate of 320mb/s, and you can get near those speeds.
It is recommended to use separate physical channels as each virtual channel in use increases the floor noise, and a wide spectrum to use allows for more resiliency. But if you're a cable provider with few free channels, you can push over 1gb/s over 2-3 6mhz channels.
Upstream not allow for a single device to lock several virtual channels in the same physical channel for reasons like higher noise and less power because of the ranges the upstream uses.
Power consumption scales linearly with frequency and with the square of the voltage. "high end" low-power CPUs are not only low frequency, but low voltage. They are more rare than your more general CPUs that are between high-end low-power and high-end higher-performance. Lowering the voltage not only reduces power lost by oscillation of a conductor with capacitance(the square scaling), but also electrical leakage.
I wouldn't jump the gun too fast. FreeBSD just had a new network API added in 10 that doubles the packet throughput of legacy when using a wrapper and over 10x the throughput when using native. A single core ATOM cpu could handle full duplex routing of a 10gb interface while running in user mode, outside of the kernel.
The new interface will allow crazy low overhead for usermode programs to access the NICs.
FreeBSD 10 has a way to do user-mode firewall with performance that allows a dual-core 450mhz CPU to handle 10gb/s of 64byte packets. It involves no context switches and zero-copy of the packet-data.
At one point in the past few years, one of OCZ's popular SSD lines had a 20% failure rate. Most of their other products have consistently been about double to triple the failure rate of everyone else's.
Heck yeah. At least 10x more efficient than silicon per clock cycle, supports speeds into the 100ghz ranges, and can handle temperatures well over 200c.
RAM vs cache are two different things. Cache scales something like O(n^2), but is a small number, while RAM scales O(1), but is a large number. The latency of a huge sram cache would be horrendous. Very very generalized.
The retooling is crazy expensive, in the order of tens of billions. If you can get more money from your old machinery while refining your next gen, then to keep from competing with yourself.
Smaller than a 0 dimensional point particle?
I only recently found out that I've been doing FBP, mostly. I naturally gravitated towards it as it makes multithreading relatively easy.
Quick! Throw it in the freezer!
Free market works when people want stuff, not when they NEED it. When people need something, they are forced to pay for it, which is monopolistic.
Free market health care is like a free market justice system.
None of that thought! This is 'merica! We would rather pay for a pound of cure than an ounce of prevention.
Society is a socialist idea. Nomads or go home.
Blizzard is like NewEgg, there will always be someone who got a bad experience. Given a large user base, it's a statistical guarantee, it just sucks when it's you.
I had a relatively small obsession with porn around 11 or 12, then I got sucked into Quake Team Fortress and anime. Porn was new and exciting, but ultimately boring. If I was to do a do-over with my life, it would be the amount of gaming without breaks I did as a kid.
I got a new USR modem.. ohh, conflicts with my mouse's IRQ.. derpa-da-derp.. There... Ohh, a printer, lets hook that up also. LPT1 conflicts with my mouse again, but there's not many options left.. hmmmm.. If I change the mouse to this IRQ, the modem to this IRQ, then the printer can be on this one! Woah... that sucked, hope I don't get an external zip drive, I may not have enough IRQs.