The original poster did say it's for FEA/CFD. You may not know what this means, but that doesn't mean there's no basic information. Incidentally, the suggestion for 28 core Xeon nodes is exactly what we got for weather prediction this year (and CFD would be similar - need for huge number crunching power, with lots of communication between threads every timestep).
The only obvious alternative would be a GPU-based solution, but to my knowledge most existing FEA & CFD codes don't use GPUs.
Except that Intel's GPUs just don't support some of their functionality on Linux. Like OpenCL. Or a modern OpenGL version.
Right, you might not care, if your usage pattern is mostly about websites and text files. For me, nVidia GPUs are the *only* thing that both brings the functionality I need (as a GPGPU software developer) and actually works.
AMD linux drivers are in a habit of losing functionality over time. Like all functionality (happened to me once). Others have complained that after updating the driver, some parts of the functionality that were present are no longer there. Because of the way Linux kernels work, you usually can't put an ancient driver to a new Linux distro.
Listening to the constant feedback from users was the problem.
Same thing happened with Nokia phones. After iPhone came out, most users switched over. Some still thought their Nokia phones were better suited to them, but majority liked iPhone better. So Nokia started making iPhone-like phones, losing their remaining customers.
most of the time I want to delete old code when I've written something newer to replace it. I keep the old code there for easy reference, until I'm confident the new code works at least as reliably as the old one.
I don't bother to find out about the available deductions. I don't mind paying some more taxes - that money also ends up benefiting me and others, I don't care I don't directly control how.
The police want laws to say they can't be filmed but they want to film everyone. They want maximum transparency of the population and none for themselves.
Nope. Nothing says it was the same policemen having both ideas. Besides, someone could prefer no filming, but still resort to filming oneself while filming remains legal.
You almost came to the right conclusion yourself. A drop-in replacement for oil does not exist now, although such are being developed in America and Europe. Finding such a "replacement" is very important. How could we support that kind of science and business? A carbon tax would work, wouldn't it? Energy's getting more and more expensive anyway, and if the rises in gas prices this far haven't "crippled the economy", then why would future rises do that?
No matter what we do, price of fossil fuels is going up. That is a requirement for using the tar sand reserves, as extracting oil from those is very expensive.
Well, the effect of humans depends on what humans do. What many people don't seem to realize is that different IPCC scenarios postulate different anthropogenic CO2 output. If the rate of pollution is soon stabilized and then slowly decreased, we're looking at something like 2 C in the next 100 years. If greenhouse gas emission rate increases exponentially (as this far), we're looking at something like 4 C. The arctic will probably warm double as much.
I've seen it in conferences, that if all fossil fuel use was stopped right now, the warming impact is expected to continue for at least 1000 years.
In my understanding, this would be mostly because oceanic circulation is so slow, one cycle of the thermohaline circulation taking 1600 years at max. As such, the oceans take thousands of years to warm up all the way (and in the meanwhile, they're net heat sinks).
The IPCC is a panel of experts on climate change. There's a couple thousand scientists contributing to IPCC reports, and only a handful of them have been found to make a mistake. The "climategate" guys were freed from all charges, investigations found they did nothing wrong. IPCC still is the most credible - nay, near the only credible - reviewer of climate science.
Actually, we have computer images that show the Earth's surface temperature, show it growing over time, and most importantly we can reproduce the effect of CO2 on longwave radiation in a lab. These cases are actually remarkably similar. In both cases, computer simulation is needed to determine how much of the lab findings would be expected in reality.
The one difference is that the ozone hole images are very clear, while our temperature image is noisy. That is to be expected, as satellites only see a very thin surface, and thermometers also only measure air temperature right around the thermometer. The heat capacity of the whole atmosphere is roughly equal to a three-meter layer of ocean.
I wasn't saying 2.5% was unsustainable. I bet some people are saying that, but I think they say that because they'd rather use that money for something else.
"Most greenies" might be advocating such things in order to pull in some direction. It's a widely accepted idea that if you want a 2% raise, you should ask for 4% and then negotiate. There was even a name for this, but I forget.
Sorry, you're wrong. He is saying different things.
We know some things for certain. Global average climate has changed by a measurable, albeit small, amount. We know that emitting CO2 changes the climate. It is happening alright, and unless there are huge volcanic eruptions or other catastrophic natural disasters, it will keep happening. The rate of its happening has been projected to be quite modest, by IPCC, in 2007. Even before, the best scientific scenarios have been realistic - and more and more realistic all the time. Certainly we don't know exactly what the climate is doing, but our idea is getting better and better.
It seems to me that James Lovelock has just taken 20 years to admit he was wrong in the eighties.
I agree with you that we shouldn't dismantle civilization. Let's instead make a small effort, and put 5% of GDP into minimizing our contribution to climate change. That'd be a small contribution, hardly noticable, but would already do something.
The article was published in 2006. How is this 'new?'
Well, I agree with your point. But six years is a good time to let scientific papers simmer. Less than that is not enough time for other scientists to evaluate the correctness and value of some paper.
Transportation can be (and has historically been) powered with non-fossil-fuels. Likewise, heat can be produced without fossil fuels. But you've got a point - transportation is probably the most painful casehere. That's why I'm hoping that the usage of fossil fuels will be brought down gradually within 10 - 20 years, and not instantaneously. Most vehicles are replaced in 10 - 20 years anyway.
Btw, electric cars are almost as good as fossil-fuel-powered ones. People are not changing over very much, because they're lazy and reluctant to lose even the tiniest bit of convenience. Switching over now would definitely be felt everywhere, but I wouldn't call it suffering. Btw, I'm lazy too and no better than the next guy, and that's why I'm hoping we'll get strong governance directing more environmentally friendly solutions to become the cheapest and most convenient ones, so that people will actually switch over. As a first remedy, governments globally should put heavy taxes on all cars that pollute more than the most fuel-efficient (combustion engine) alternative.
As a final note, I don't claim to know for fact that electric cars would be less polluting overall, even though the text above assumes so.
We agree on one thing: fossil fuels need to be replaced with something else. The issue is not as catastrophic as you make it sound - many countries already produce half of their energy in some non-fossil-fuel way. But how can we get human populations to switch to something else? Humans are ingenious, resourceful and lazy. The change comes either from political pressure, or the depletion of fossil fuel resources. The former can be a controlled process, the latter cannot.
By the way - estimates on how long fossil fuel resources last are generally based on the "guess", that their use will remain constant until we run out. In reality, energy usage is growing exponentially. http://www.youtube.com/watch?v=umFnrvcS6AQ (The Most IMPORTANT Video You'll Ever See) describes the situation with proper gravity.
That was my problem with this too. Pretty much the only one I have. I don't mind doctors charging from their work, but I greatly object anyone making money from my organs, just because there's scarcity on the "market" and they can.
Trust those who find errors in their own processes and report them openly. Do not trust those who hide errors, or don't even find them in the first place. Everybody makes errors.
I thought that the exam questions change from exam to exam! Based on this assumption, I'd call the deed just "studying". Sometimes going through ten previous exams is the only way to figure out what the lecturer thinks is important about a course. Sure, this is a problem that could be solved by other methods as well, but it isn't. Go figure.
The exams should test if students know what they should. If an exam succeeds in this, then students can not study exams without learning exactly the things they were supposed to. Thus studying old exams is in principle no different from studying course exercises or course material.
What Really Happened Aboard Air France 447 tells the story as it stands after investigations. It's a rather chilling read. But it makes one thing clear: it was about human error. The plane was even fully operational when it crashed, as an anti-icing system had managed to bring air speed sensors back to operation before it.
Two years after the Airbus 330 plunged into the Atlantic Ocean, Air France 447's flight-data recorders finally turned up. The revelations from the pilot transcript paint a surprising picture of chaos in the cockpit, and confusion between the pilots that led to the crash.
Slashdot Mirror
The original poster did say it's for FEA/CFD. You may not know what this means, but that doesn't mean there's no basic information. Incidentally, the suggestion for 28 core Xeon nodes is exactly what we got for weather prediction this year (and CFD would be similar - need for huge number crunching power, with lots of communication between threads every timestep). The only obvious alternative would be a GPU-based solution, but to my knowledge most existing FEA & CFD codes don't use GPUs.
Except that Intel's GPUs just don't support some of their functionality on Linux. Like OpenCL. Or a modern OpenGL version.
Right, you might not care, if your usage pattern is mostly about websites and text files. For me, nVidia GPUs are the *only* thing that both brings the functionality I need (as a GPGPU software developer) and actually works.
AMD linux drivers are in a habit of losing functionality over time. Like all functionality (happened to me once). Others have complained that after updating the driver, some parts of the functionality that were present are no longer there. Because of the way Linux kernels work, you usually can't put an ancient driver to a new Linux distro.
"Hey, you've done nothing wrong, but you could still end up in jail for a long time. How about only going to jail for 4 years?"
Yeah, sounds about fair.
Listening to the constant feedback from users was the problem.
Same thing happened with Nokia phones. After iPhone came out, most users switched over. Some still thought their Nokia phones were better suited to them, but majority liked iPhone better. So Nokia started making iPhone-like phones, losing their remaining customers.
most of the time I want to delete old code when I've written something newer to replace it. I keep the old code there for easy reference, until I'm confident the new code works at least as reliably as the old one.
I don't bother to find out about the available deductions. I don't mind paying some more taxes - that money also ends up benefiting me and others, I don't care I don't directly control how.
The police want laws to say they can't be filmed but they want to film everyone. They want maximum transparency of the population and none for themselves.
Nope. Nothing says it was the same policemen having both ideas. Besides, someone could prefer no filming, but still resort to filming oneself while filming remains legal.
You almost came to the right conclusion yourself. A drop-in replacement for oil does not exist now, although such are being developed in America and Europe. Finding such a "replacement" is very important. How could we support that kind of science and business? A carbon tax would work, wouldn't it? Energy's getting more and more expensive anyway, and if the rises in gas prices this far haven't "crippled the economy", then why would future rises do that?
No matter what we do, price of fossil fuels is going up. That is a requirement for using the tar sand reserves, as extracting oil from those is very expensive.
Well, the effect of humans depends on what humans do. What many people don't seem to realize is that different IPCC scenarios postulate different anthropogenic CO2 output. If the rate of pollution is soon stabilized and then slowly decreased, we're looking at something like 2 C in the next 100 years. If greenhouse gas emission rate increases exponentially (as this far), we're looking at something like 4 C. The arctic will probably warm double as much.
I've seen it in conferences, that if all fossil fuel use was stopped right now, the warming impact is expected to continue for at least 1000 years.
In my understanding, this would be mostly because oceanic circulation is so slow, one cycle of the thermohaline circulation taking 1600 years at max. As such, the oceans take thousands of years to warm up all the way (and in the meanwhile, they're net heat sinks).
The IPCC is a panel of experts on climate change. There's a couple thousand scientists contributing to IPCC reports, and only a handful of them have been found to make a mistake. The "climategate" guys were freed from all charges, investigations found they did nothing wrong. IPCC still is the most credible - nay, near the only credible - reviewer of climate science.
Actually, we have computer images that show the Earth's surface temperature, show it growing over time, and most importantly we can reproduce the effect of CO2 on longwave radiation in a lab. These cases are actually remarkably similar. In both cases, computer simulation is needed to determine how much of the lab findings would be expected in reality.
The one difference is that the ozone hole images are very clear, while our temperature image is noisy. That is to be expected, as satellites only see a very thin surface, and thermometers also only measure air temperature right around the thermometer. The heat capacity of the whole atmosphere is roughly equal to a three-meter layer of ocean.
I wasn't saying 2.5% was unsustainable. I bet some people are saying that, but I think they say that because they'd rather use that money for something else.
"Most greenies" might be advocating such things in order to pull in some direction. It's a widely accepted idea that if you want a 2% raise, you should ask for 4% and then negotiate. There was even a name for this, but I forget.
So we shouldn't have saved the ozone layer from CFCs, because nobody was certain about it? Because in science, nothing is ever really certain.
Sorry, you're wrong. He is saying different things.
We know some things for certain. Global average climate has changed by a measurable, albeit small, amount. We know that emitting CO2 changes the climate. It is happening alright, and unless there are huge volcanic eruptions or other catastrophic natural disasters, it will keep happening. The rate of its happening has been projected to be quite modest, by IPCC, in 2007. Even before, the best scientific scenarios have been realistic - and more and more realistic all the time. Certainly we don't know exactly what the climate is doing, but our idea is getting better and better.
It seems to me that James Lovelock has just taken 20 years to admit he was wrong in the eighties.
I agree with you that we shouldn't dismantle civilization. Let's instead make a small effort, and put 5% of GDP into minimizing our contribution to climate change. That'd be a small contribution, hardly noticable, but would already do something.
The article was published in 2006. How is this 'new?'
Well, I agree with your point. But six years is a good time to let scientific papers simmer. Less than that is not enough time for other scientists to evaluate the correctness and value of some paper.
Transportation can be (and has historically been) powered with non-fossil-fuels. Likewise, heat can be produced without fossil fuels. But you've got a point - transportation is probably the most painful casehere. That's why I'm hoping that the usage of fossil fuels will be brought down gradually within 10 - 20 years, and not instantaneously. Most vehicles are replaced in 10 - 20 years anyway.
Btw, electric cars are almost as good as fossil-fuel-powered ones. People are not changing over very much, because they're lazy and reluctant to lose even the tiniest bit of convenience. Switching over now would definitely be felt everywhere, but I wouldn't call it suffering. Btw, I'm lazy too and no better than the next guy, and that's why I'm hoping we'll get strong governance directing more environmentally friendly solutions to become the cheapest and most convenient ones, so that people will actually switch over. As a first remedy, governments globally should put heavy taxes on all cars that pollute more than the most fuel-efficient (combustion engine) alternative.
As a final note, I don't claim to know for fact that electric cars would be less polluting overall, even though the text above assumes so.
We agree on one thing: fossil fuels need to be replaced with something else. The issue is not as catastrophic as you make it sound - many countries already produce half of their energy in some non-fossil-fuel way. But how can we get human populations to switch to something else? Humans are ingenious, resourceful and lazy. The change comes either from political pressure, or the depletion of fossil fuel resources. The former can be a controlled process, the latter cannot.
By the way - estimates on how long fossil fuel resources last are generally based on the "guess", that their use will remain constant until we run out. In reality, energy usage is growing exponentially. http://www.youtube.com/watch?v=umFnrvcS6AQ (The Most IMPORTANT Video You'll Ever See) describes the situation with proper gravity.
That was my problem with this too. Pretty much the only one I have. I don't mind doctors charging from their work, but I greatly object anyone making money from my organs, just because there's scarcity on the "market" and they can.
Trust those who find errors in their own processes and report them openly. Do not trust those who hide errors, or don't even find them in the first place. Everybody makes errors.
Just wanted to add: even if the cable might have to be 10 meters thick at its thickest, it does not have to be equally thick everywhere.
You'd also see a couple of lakes and fields!
I thought that the exam questions change from exam to exam! Based on this assumption, I'd call the deed just "studying". Sometimes going through ten previous exams is the only way to figure out what the lecturer thinks is important about a course. Sure, this is a problem that could be solved by other methods as well, but it isn't. Go figure.
The exams should test if students know what they should. If an exam succeeds in this, then students can not study exams without learning exactly the things they were supposed to. Thus studying old exams is in principle no different from studying course exercises or course material.
pilot error as in hiding a bug in airbus autopilot or it reading faulty gauges.
http://www.thedailybeast.com/articles/2011/05/27/air-france-flight-447-crash-report-airbus-autopilot-to-blame.html
What Really Happened Aboard Air France 447 tells the story as it stands after investigations. It's a rather chilling read. But it makes one thing clear: it was about human error. The plane was even fully operational when it crashed, as an anti-icing system had managed to bring air speed sensors back to operation before it.
Two years after the Airbus 330 plunged into the Atlantic Ocean, Air France 447's flight-data recorders finally turned up. The revelations from the pilot transcript paint a surprising picture of chaos in the cockpit, and confusion between the pilots that led to the crash.