They are assuming 4.5% efficiency and electricity as a 1:1 replacement for all energy needless to say their numbers are totally wacked. Average power per square meter is around 1000W * 8 hours / 24hours a day * % efficiency. So at 10% you get 33W/M not 15 but we already have 30% efferent solar cells. Beyond that using sunlight to generate electricity to heat water is stupid. Directly generating hot water from sunlight produces an average power output of around 300W/M which is 20 times higher than their estimate. Toss in some wind turbines and the amount of land required continues to drop.
Also electric cars require far less energy than gas cars so that 15TW figure is going to drop even more.
PS: The world's total supply of energy extractable from fossil fuels (Oil + Coal + Natural Gas) is less than the amount of energy extractable from renewable sources in one year. (http://en.wikipedia.org/wiki/World_energy_resources_and_consumption)
PPS: the above post was a joke, you don't reference memory on a single bit basis. 64 bit system can easily use 64bit word sizes and access 128exabyte's of ram.
Wait, 2^64 = 18,446,744,073,709,551,616 bits, but only 2,305,843,009,213,693,952 bytes. Anyway someone is going to want 2,097,152 terabytes of RAM before too long so I expect MS needs to look at 128bit systems before long.
That's overkill for 17 years. Just use a few USB sticks. Interfaces tend to stay backward compatable for a long time. So finding a 5 year old computer that supports a 12 year old standard in 2026 should be easy.
No, 0.1101001000100001000001... is irrational and does not encode useful information in any known computer language. Just because the number is not an integer fraction does not mean it contain all digit combinations.
To increase profit. The goal is to maximize (unit's sold * profit per unit) not just (unit's sold) or (profit per unit).
In theory if you lower the price you can attract more customers. This is offset by a loss in profit per customer. In the real world not everyone is willing to pay the same price. You see this most clearly with coupons. People who are not willing to use coupons are charged more because not as price sensitive. EX: I could not care less if my lettuce cost's 10% more so they charge me 10% more.
If you really balance the day night load then they are making money off of the float in energy costs over the day / night cycle.
It's stupid for individuals to pay for a battery back up system when it saves the utility companies money when you exchange daytime for nighttime usage. On the other hand if it's a significant fee, they they are going to drive some people off the grid which is a horrible idea for them.
PS: If day time rates ever dropped below night time rates then a fee might be a good idea.
I was talking about the site including the ship and flag etc. It's like the pyramid at Giza, it would be cool to keep a stone from there on your desk, but it's never going to be replaced. If they had landed several lenders and knew they would not randomly land in the middle of the site then cool, but for a first visit it seems risky.
Yes, but you can learn all that from looking at the 2nd moon landing site. There will only ever be one "first footprint on the moon" and when it's gone that's it.
A network of Radio telescopes http://en.wikipedia.org/wiki/Radio_telescope targeting fixed locations would have close to unlimited bandwidth. You are ultimately limited by diffraction for each part of the spectrum but that limit get's insane.
No offence, but you sound clueless. Convergence and systems hardening are not trends to be kept up with. What the market is trying to sell you has nothing to do with your job as CIO. Your real job is balancing R&D, cost, stability, security, and usability. Consider a classic convergence decision like moving to a VoIP phone system. To move is the high level decision, but the important part is understanding the benefits of a specific VoIP system which is independent from VoIP as a concept. A bad VoIP system is often more expensive and less stable than what was replaced. So VoIP as a concept is useless.
PS: Understanding what's coming 6 months from now always both useless and impossible, understanding what's available today is important and difficult.
To give you an idea how much computational power they could have using specilized hardware. Let's compare that to a 9800GTX.
65 megawatts / 140watts * 432gflops = ~200,000 TeraFlops or 200,000,000,000,000,000 Flops. For something like 40 to 80 million$.
Granted the accuracy of this estimate sucks as GTX's don't have networking suppport, and we need to cool things ect. But, they could also use more effecent hardware than the GTX.
No single country has the best treatment for all diseases. Rich Americans often leave the US to seek treatment in other countries. When scheduling a procedure more than 2 days in advance you can basically go to any hospital in the world. The real problem is not the peek quality of the US healthcare system it's all the gap's which kill people.
PS: People also spend over a million dollars to buy a car, suggesting that some people spend a lot of money on something says little about utility.
The problem with this approach is the nature of collage assignments. If you ignore variable names and comments then there is only so many ways to write a short efficient bubble sort. Schools with hundreds of students submitting work results in a large number of students submitting the same basic program.
I think the best solution for this is to have students fill out some functions in a program that mostly works. That way the teacher could change a few things each year and prevent students from submitting a generic solution.
10 years after a dam breaking you can use the land, 10 years after Chernobyl they where still guarding the wasteland. The real cost of CHernobyl was not the 56 direct deaths but the ~4,000 additional cancer deaths. The loss of a city and the 19 mi exclusion zone around the site (http://en.wikipedia.org/wiki/Zone_of_alienation). Plus the constant low enforcement issue.
PS: The overall cost of the disaster is estimated at US$200 billion, taking inflation into account. This places the Chernobyl disaster as the most costly disaster in modern history.[5] There are a tiny number of dam's worldwide that could damage on that magnitude. However, most of those saves lives due to a reduction in annual flooding and a steady water source so they would exist even if they did not provide energy.
When a ship blows up assets are distryed not ISK, but buying more items from the market distroy's cach due to market transaction fees.
The more inflated the item the more cash is distroyed in the market. It's a vary stable self balancing system. If it get's to far out of whack you can make ISK by buying items from NPC's and distroying them for the minerals which also distroys ISK both to buy the item and in the market transactions fees.
PS: Cheep ships are next to meaningless on the overall market economy most cash is tied up in the T2 market and the really big ships. (I have not logged in for a while so I don't know if they added T3 yet.)
I think the analogy still stands. If some random person in China is looking for a desktop OS they are probably not willing to pay 50$ for it. However, said random person is not going to edit the Linux OS so the comparison becomes between to "Free" items that are basically identical to them.
And I am not the only person that takes that viewpoint so the post a few above mine:
"The poster's comment about Windows is that, much like open source Linux, you can go online and easily find and download the latest version for free. In context, his meaning is pretty clear."
So you might want to consider the idea rather than simply dismissing it as indefensible.
PS: I am a software developer but while it's theoretically possible that I could decide update the OS I also understand enough about software development to realize I would first have to understand what it's doing. So while I like the fact it's based on UNIX I don't care about the fact it's open source. In the same way that I would not care if it was written in strait ASM or Pascal.
When non coders hear open source they think free as in beer. So the fact it's easy to get a copy of Windows for free means people who want free still have the windows option.
Slashdot Mirror
They are assuming 4.5% efficiency and electricity as a 1:1 replacement for all energy needless to say their numbers are totally wacked. Average power per square meter is around 1000W * 8 hours / 24hours a day * % efficiency. So at 10% you get 33W/M not 15 but we already have 30% efferent solar cells. Beyond that using sunlight to generate electricity to heat water is stupid. Directly generating hot water from sunlight produces an average power output of around 300W/M which is 20 times higher than their estimate. Toss in some wind turbines and the amount of land required continues to drop.
Also electric cars require far less energy than gas cars so that 15TW figure is going to drop even more.
PS: The world's total supply of energy extractable from fossil fuels (Oil + Coal + Natural Gas) is less than the amount of energy extractable from renewable sources in one year. (http://en.wikipedia.org/wiki/World_energy_resources_and_consumption)
PPS: the above post was a joke, you don't reference memory on a single bit basis. 64 bit system can easily use 64bit word sizes and access 128exabyte's of ram.
Wait, 2^64 = 18,446,744,073,709,551,616 bits, but only 2,305,843,009,213,693,952 bytes. Anyway someone is going to want 2,097,152 terabytes of RAM before too long so I expect MS needs to look at 128bit systems before long.
PS: It's only 2exabytes, start to dream.
Don't worry, they might make the stuff chemically inert, but it's still radioactive.
That's overkill for 17 years. Just use a few USB sticks. Interfaces tend to stay backward compatable for a long time. So finding a 5 year old computer that supports a 12 year old standard in 2026 should be easy.
No, 0.1101001000100001000001... is irrational and does not encode useful information in any known computer language. Just because the number is not an integer fraction does not mean it contain all digit combinations.
The cost benefit curve in the consumer CPU space is strange. A 10% faster might be worth twice as much or be totally useless.
EX: Real time playback of HD movies. If your CPU can't do that it's a pain. However, if it can then extra speed is pointless for that task.
To increase profit. The goal is to maximize (unit's sold * profit per unit) not just (unit's sold) or (profit per unit).
In theory if you lower the price you can attract more customers. This is offset by a loss in profit per customer. In the real world not everyone is willing to pay the same price. You see this most clearly with coupons. People who are not willing to use coupons are charged more because not as price sensitive. EX: I could not care less if my lettuce cost's 10% more so they charge me 10% more.
FYI: Military spending on fuel is negligible in comparison to manpower.
The link said the Air Force uses 1/2 the fuel of our armed forces, which is probably also waked as the Army has more airplanes than the air force.
PS: Oil costs less than 5% of our GDP. We use lot's of oil because it's cheep but there are lot's of options including turning coal into gas.
If you really balance the day night load then they are making money off of the float in energy costs over the day / night cycle.
It's stupid for individuals to pay for a battery back up system when it saves the utility companies money when you exchange daytime for nighttime usage. On the other hand if it's a significant fee, they they are going to drive some people off the grid which is a horrible idea for them.
PS: If day time rates ever dropped below night time rates then a fee might be a good idea.
I was talking about the site including the ship and flag etc. It's like the pyramid at Giza, it would be cool to keep a stone from there on your desk, but it's never going to be replaced. If they had landed several lenders and knew they would not randomly land in the middle of the site then cool, but for a first visit it seems risky.
Yes, but you can learn all that from looking at the 2nd moon landing site. There will only ever be one "first footprint on the moon" and when it's gone that's it.
A network of Radio telescopes http://en.wikipedia.org/wiki/Radio_telescope targeting fixed locations would have close to unlimited bandwidth. You are ultimately limited by diffraction for each part of the spectrum but that limit get's insane.
A bad decision is still a bad decision even if it's the most common choice.
No offence, but you sound clueless. Convergence and systems hardening are not trends to be kept up with. What the market is trying to sell you has nothing to do with your job as CIO. Your real job is balancing R&D, cost, stability, security, and usability. Consider a classic convergence decision like moving to a VoIP phone system. To move is the high level decision, but the important part is understanding the benefits of a specific VoIP system which is independent from VoIP as a concept. A bad VoIP system is often more expensive and less stable than what was replaced. So VoIP as a concept is useless.
PS: Understanding what's coming 6 months from now always both useless and impossible, understanding what's available today is important and difficult.
To give you an idea how much computational power they could have using specilized hardware. Let's compare that to a 9800GTX.
65 megawatts / 140watts * 432gflops = ~200,000 TeraFlops or 200,000,000,000,000,000 Flops. For something like 40 to 80 million$.
Granted the accuracy of this estimate sucks as GTX's don't have networking suppport, and we need to cool things ect. But, they could also use more effecent hardware than the GTX.
The LCD clock could probably run for a few months per housefly. It's mostly a question of digestion effecency, and other losses.
They want 50GFLOPS/w so you start talking about GFLOP people assumed it's a typo.
PS: Hand them something that does 50GFLOP per year / w and they will say it's out of spec because they said 50GFLOPS/W.
If that's what's going on then you would need to cool the ocean or as the ice melts it starts to escape.
No single country has the best treatment for all diseases. Rich Americans often leave the US to seek treatment in other countries. When scheduling a procedure more than 2 days in advance you can basically go to any hospital in the world. The real problem is not the peek quality of the US healthcare system it's all the gap's which kill people.
PS: People also spend over a million dollars to buy a car, suggesting that some people spend a lot of money on something says little about utility.
The problem with this approach is the nature of collage assignments. If you ignore variable names and comments then there is only so many ways to write a short efficient bubble sort. Schools with hundreds of students submitting work results in a large number of students submitting the same basic program.
I think the best solution for this is to have students fill out some functions in a program that mostly works. That way the teacher could change a few things each year and prevent students from submitting a generic solution.
10 years after a dam breaking you can use the land, 10 years after Chernobyl they where still guarding the wasteland. The real cost of CHernobyl was not the 56 direct deaths but the ~4,000 additional cancer deaths. The loss of a city and the 19 mi exclusion zone around the site (http://en.wikipedia.org/wiki/Zone_of_alienation). Plus the constant low enforcement issue.
PS: The overall cost of the disaster is estimated at US$200 billion, taking inflation into account. This places the Chernobyl disaster as the most costly disaster in modern history.[5] There are a tiny number of dam's worldwide that could damage on that magnitude. However, most of those saves lives due to a reduction in annual flooding and a steady water source so they would exist even if they did not provide energy.
When a ship blows up assets are distryed not ISK, but buying more items from the market distroy's cach due to market transaction fees.
The more inflated the item the more cash is distroyed in the market. It's a vary stable self balancing system. If it get's to far out of whack you can make ISK by buying items from NPC's and distroying them for the minerals which also distroys ISK both to buy the item and in the market transactions fees.
PS: Cheep ships are next to meaningless on the overall market economy most cash is tied up in the T2 market and the really big ships. (I have not logged in for a while so I don't know if they added T3 yet.)
I think the analogy still stands. If some random person in China is looking for a desktop OS they are probably not willing to pay 50$ for it. However, said random person is not going to edit the Linux OS so the comparison becomes between to "Free" items that are basically identical to them.
And I am not the only person that takes that viewpoint so the post a few above mine:
"The poster's comment about Windows is that, much like open source Linux, you can go online and easily find and download the latest version for free. In context, his meaning is pretty clear."
So you might want to consider the idea rather than simply dismissing it as indefensible.
PS: I am a software developer but while it's theoretically possible that I could decide update the OS I also understand enough about software development to realize I would first have to understand what it's doing. So while I like the fact it's based on UNIX I don't care about the fact it's open source. In the same way that I would not care if it was written in strait ASM or Pascal.
When non coders hear open source they think free as in beer. So the fact it's easy to get a copy of Windows for free means people who want free still have the windows option.