The lone inventor is mostly something for a new area of research. Once an area of research is mature, it's near impossible to compete with larger commercial entities.
Probably near the distance where quantum tunneling becomes an issue. When your design requires a layer of insulation and electrons can pass through it, then your design breaks.
Until people start complaining again. How do you think we got those regulations in the first place? Magic? They're well understood problems in the voice calling industry, and have been mostly "solved" for a long while. If they suddenly crop up again, I'm sure the FCC will come down hard and fast since they already have prior experience.
Don't underestimate a bunch of 50+ year old people complaining en masse.
Depends on your definition of VOIP. I use Mumble for gaming and it sounds great and has low latency, especially with the new OPUS CODEC. I live in the Midwest and I play games with other people in the Midwest. We used an LA Mumble server for a long time and there was no noticeable latency, yet someone else's Chicago TeamSpeak server had noticeable latency in the audio, even though the ping and jitter to the server was much lower.
Don't allow VOIP over cable to replace POTS, only non-powered fiber networks. Cable networks need AMPs, fiber typically doesn't even need repeaters at all.
As a child, most of the time I lost power, I also lost phone. Something about the phone lines being strung up with the power lines, and when a tree falls over and takes out the entire pole, you lose both services.
I haven't owned a POTS in over 10 years now, just a call phone. I had POTS back in the 90s when I was a kid and I had the phone go down many times and had several times where the static was do bad I could not effectively communicate.
I don't know about cable/dsl, but any fiber system can have a lot of up time with a small UPS. You're talking about a fiber device that consumes 5watts under load, less when idle. A small car battery will get you almost a full week of up-time. You can get a small hand-crank generator that puts out 10 watts. Someone in another forum mentioned that their ISP is running DC power along with the fiber,so their ONTs can stay powered amidst an outage.
We probably need some regulations about VOIP services for anything that is meant to replace POTS.
Intel's new 250mw CPUs have more mips and flops than ARM with the same power rating, and the Intel CPU has better idle power draw. Intel is starting to beat ARM at their own game and with full x86 CPUs. ARM is trying to break into Intel territory and looks to have "decent" power draw and performance, but many of their current CPUs do not scale well past 1ghz. Like their ARM A9 consumes 4x the power just to gain an extra 10% performance on the high end. These "low power" CPUs are great at low frequencies, but are horrible at higher frequencies. People assumed they'd scale mostly linearly and compete with Intel by just clocking up.
ARM is still doing a lot of research and there is no reason they won't be able to compete with Intel on the desktop, but they're not there quite yet. They're just breaking into the server market, this will be interesting.
But IT doesn't change much. Same basic ideas that I read about from security and design still apply from the 80s as much as it does today. The tools change and give you more power to control, but the ideas are still the same.
Her time may slow down to a near halt, but she will keep moving relative to you. It's her movement relative to you that causes her time to slow down. In order for her time to nearly stop, she has to be moving away from you near c. She won't just "stop falling in".
When talking about quantum stuff, stating that there is "no chance" of something happening is pure bunk. There is a chance of anything happening, just highly unlikely.
Pretty much this. Every year my review has a question of what I like most about my job, and every year I say that I love knowing my work helps people and many times I even get to see or hear back about how much of a difference it makes.
In college, nearly every teacher that I had gave you until the end of the day to hand in, but at least the end of the class. Some of them went so far as "end of this semester". I was having a rough patch in my life and one teacher told me I had until the beginning of the NEXT semester, because that's the absolute deadline he had to fill in my grade.
Your Honor, my client has suffered major losses in paying me to pursue these theives. These people should have to cover the real losses of my exorbitant fees.
AMD GPUs can execute several kernels at a time and each Mantle task is a kernel. GPUs are meant to be high throughput and will keep itself busy as long as you can feed it work fast enough.
Single threaded per queue and you can create as many as you want per application. During setup, you create all of the queues you want to use, then register them. Once registered, the GPU can directly read from the queue instead of making a system call. This pretty much removes those pesky 30,000 clock cycle system calls. Kaveri goes a step further and registers queues directly with the hardware, because each queue item is exactly one cache line in size and the APU shares the cache-coherency and protected memory, the latency between the CPU and GPU is about the latency of the cache, which is around 10ns.
Because prior to Mantle, system calls were used, a 10ns communications latency was completely dwarfed by the 30,000 cycle system call. Now we're getting along the lines of 0 call overhead and 10ns latency. Obviously the GPU must have free resources to start working and some code may need to be ran in order for the GPU to schedule, but the main point still remains.
Slashdot Mirror
I like your summary better :)
The lone inventor is mostly something for a new area of research. Once an area of research is mature, it's near impossible to compete with larger commercial entities.
Probably near the distance where quantum tunneling becomes an issue. When your design requires a layer of insulation and electrons can pass through it, then your design breaks.
Even then, some things may be 4x more expensive, but I may be getting paid 5x more and other things may be 10,000x less expensive, like computers.
Until people start complaining again. How do you think we got those regulations in the first place? Magic? They're well understood problems in the voice calling industry, and have been mostly "solved" for a long while. If they suddenly crop up again, I'm sure the FCC will come down hard and fast since they already have prior experience.
Don't underestimate a bunch of 50+ year old people complaining en masse.
Depends on your definition of VOIP. I use Mumble for gaming and it sounds great and has low latency, especially with the new OPUS CODEC. I live in the Midwest and I play games with other people in the Midwest. We used an LA Mumble server for a long time and there was no noticeable latency, yet someone else's Chicago TeamSpeak server had noticeable latency in the audio, even though the ping and jitter to the server was much lower.
you could flip cable channels with maybe 0.25s latency
Kind of like YouTube or Netflix. Other than YouTube tends to have ads at the beginning of videos. uhgg.
Don't allow VOIP over cable to replace POTS, only non-powered fiber networks. Cable networks need AMPs, fiber typically doesn't even need repeaters at all.
As a child, most of the time I lost power, I also lost phone. Something about the phone lines being strung up with the power lines, and when a tree falls over and takes out the entire pole, you lose both services.
I haven't owned a POTS in over 10 years now, just a call phone. I had POTS back in the 90s when I was a kid and I had the phone go down many times and had several times where the static was do bad I could not effectively communicate.
I don't know about cable/dsl, but any fiber system can have a lot of up time with a small UPS. You're talking about a fiber device that consumes 5watts under load, less when idle. A small car battery will get you almost a full week of up-time. You can get a small hand-crank generator that puts out 10 watts. Someone in another forum mentioned that their ISP is running DC power along with the fiber,so their ONTs can stay powered amidst an outage.
We probably need some regulations about VOIP services for anything that is meant to replace POTS.
Intel's new 250mw CPUs have more mips and flops than ARM with the same power rating, and the Intel CPU has better idle power draw. Intel is starting to beat ARM at their own game and with full x86 CPUs. ARM is trying to break into Intel territory and looks to have "decent" power draw and performance, but many of their current CPUs do not scale well past 1ghz. Like their ARM A9 consumes 4x the power just to gain an extra 10% performance on the high end. These "low power" CPUs are great at low frequencies, but are horrible at higher frequencies. People assumed they'd scale mostly linearly and compete with Intel by just clocking up.
ARM is still doing a lot of research and there is no reason they won't be able to compete with Intel on the desktop, but they're not there quite yet. They're just breaking into the server market, this will be interesting.
But IT doesn't change much. Same basic ideas that I read about from security and design still apply from the 80s as much as it does today. The tools change and give you more power to control, but the ideas are still the same.
Her time may slow down to a near halt, but she will keep moving relative to you. It's her movement relative to you that causes her time to slow down. In order for her time to nearly stop, she has to be moving away from you near c. She won't just "stop falling in".
When talking about quantum stuff, stating that there is "no chance" of something happening is pure bunk. There is a chance of anything happening, just highly unlikely.
Pretty much this. Every year my review has a question of what I like most about my job, and every year I say that I love knowing my work helps people and many times I even get to see or hear back about how much of a difference it makes.
How do you know if your code is any good without a goal and a way to test it? But hey, it's free time, do whatever you think is fun.
I think they're talking about a blackhole that is not rotating and has no accretion disc. If you fell strait towards the blackhole, what would happen?
In college, nearly every teacher that I had gave you until the end of the day to hand in, but at least the end of the class. Some of them went so far as "end of this semester". I was having a rough patch in my life and one teacher told me I had until the beginning of the NEXT semester, because that's the absolute deadline he had to fill in my grade.
Your Honor, my client has suffered major losses in paying me to pursue these theives. These people should have to cover the real losses of my exorbitant fees.
AMD has been making commits to FreeBSD and Linux for Mantle.
AMD GPUs can execute several kernels at a time and each Mantle task is a kernel. GPUs are meant to be high throughput and will keep itself busy as long as you can feed it work fast enough.
WoW chokes, but it only uses about 30% of my CPU. My CPU is mostly idle.
What serious gamer is CPU bound these days?
Most of my games are thread bound. 80% idle CPU with a 90% idle GPU and getting 20fps.
Even Mantle is single threaded
Single threaded per queue and you can create as many as you want per application. During setup, you create all of the queues you want to use, then register them. Once registered, the GPU can directly read from the queue instead of making a system call. This pretty much removes those pesky 30,000 clock cycle system calls. Kaveri goes a step further and registers queues directly with the hardware, because each queue item is exactly one cache line in size and the APU shares the cache-coherency and protected memory, the latency between the CPU and GPU is about the latency of the cache, which is around 10ns.
Because prior to Mantle, system calls were used, a 10ns communications latency was completely dwarfed by the 30,000 cycle system call. Now we're getting along the lines of 0 call overhead and 10ns latency. Obviously the GPU must have free resources to start working and some code may need to be ran in order for the GPU to schedule, but the main point still remains.
And I thought paying $80/tire for high end all-seasons was expensive. WTF do you people drive that have such expensive tires?!