Yeah, that software that they just released *yesterday* catches them up to, what, 2003? Here's what games of five years ago (2005) looked like, and that beats the heck out of SL's latest client.
And yeah, cue SL's standard excuses in three, two, one... "Hey, that's not fair! Second life has to download its content!" Sorry, but that only applies to textures and poly counts. SL has never had problems with texture quality or poly count. It's always suffered from a pathetically low quality renderer to display them. Heck, they're *just now* adding support for displaying plausible normals shading in *2010*? No cast shadows (at least in the screenshots I've come across)? I mean, what is this, the dark ages?
You're acting like there's a single estimate of how much release there was. The Russians and the IAEA estimated 100m curies in 1986. Argonne estimated 30% of the core (3B curies) in 1986. Lawrence Livermore estiamted 50% of the core (5B curies) in 1986. The supervisor of the Russian cleanup effort estimated 80% of the core (6.4B curies) in 1991. And the Union of Concerned Scientists estimates that it all was vaporized (9B curies).
Since most of the estimates are in the billions range, that's what I used.
The problem is that the anti-nuclear types tend to want perfect technology and that of course can never exist.
Wrong. They want trillions to flow into producing today's tech, en-masse, right now. End of story.
Solar cell production releases NF3 a super greenhouse gas.
NF3 is only just now becoming recognized as a greenhouse gas for a reason -- such tiny amounts leak that they're barely even measurable. And it's being phased out anyway in favor of fluorine. And is only used in thin film photovoltaics, not photovoltaics in general, let alone solar thermal which is what most people refer to when they're talking about large-scale solar power plants.
Fusion requires Tritium (for now) which is only produced using nuclear fission reactors and a lithium blanket.
Fusion itself can breed tritium.
Geothermal can cause earthquakes.... Wind kills birds and bats.
Two very closely related nonsense claims. When will people stop pushing this FUD? Let's start with wind. There was ONE -- read it, ONE -- wind farm, built early on, that had bird kill problems: Altamont Pass. They built a bunch of fast-turning bladed turbines with towers shaped like perfect perches at low altitude in the middle of a bird flyway. I mean, if you were trying to design a raptor cuissanart, that's pretty much the design you'd use. NO wind farm today gets approved without a bird impact survey, and even if that wasn't the case, the way modern turbines are built has way less impact (much taller towers, slower rotation, and the towers are unattractive as perches). The bird kill rates of modern wind turbines are utterly dwarfed by that of other human activity, even taking into account the scale of wind farms versus the scale of other activities. The Audubon Society itself now supports wind farms because they impact birds less than most other forms of power generation. So, in short, this is FUD, plain and simple. Stop spreading it.
The exact same thing is happening with EGS, which is about a decade behind wind in terms of commercialization. One EGS project drilled straight into a faultline that had previously destroyed Basel, Switzerland, and started fracturing it. This caused a *small*, but tangible earthquake. EGS always causes earthquakes, but they're usually too small to even feel. Will people drill into active faults right next to cities anymore? Of course not; they learned the lesson with Basel. But now EGS is irreparably smeared in the same way wind was by Altamont Pass.
They all have problems and we will get no where by attacking every single technology that we have.
Right -- fair and balanced, as if bogus claims are on equal footing as valid ones. "Some say" wind kills birds. "Some say" coal causes health problems. Both have problems -- there's your equivalence, right?
You are out of touch with reality. Completely. Do you even comprehend how much pollution Coal plants produce?
Do you even comprehend how much radioactive material several to dozens of billions of curies per reactor is? Nuclear has great potential but also great risk. It *only makes sense* that it is so heavily regulated. And if such heavy regulations make it uneconomical? Tough cookies. There are plenty of other zero-carbon generation techs already available.
Wow -- this post is almost as though it came from an alternative universe where I *didn't* list what sort of power environmentalists typically support.
Wind power is noise pollution and dead birds. Geothermal causes earthquakes. Solar pollutes water streams because of the solvents used in silicon manufacturing. So yeah, make sure to keep burning that natural gas.
Almost all prominent environmentalists support wind. Most of the opposition from wind comes from rich people who are afraid of having their property values decreased by turbines. Even the audubon society supports wind power, despite the "OMG THEY KILLS BIRDZ!" FUD. And the "noise" issue is -- let's be blunt -- a lie. I've been to a big wind farm on a windy day. You can barely hear the things when you're standing at their base.
Geothermal causes tiny earthquakes, usually impossible to feel. The biggest one on record occurred from *drilling into a freaking fault line that had once destroyed the city* (something nobody's going to be dumb enough to do again), and even still, the quake was quite small. The opposition to EGS is quite limited and definitely isn't coming from the environmental movement.
Almost all prominent environmentalists support solar. The amount of waste from solar manufacturing is tiny compared to the amount of waste in pretty much every other kind of power generation on a per-watt-hour basis. And is way overplayed, at that.
Natural gas isn't perfect, but it's a good low carbon/low health-pollution peaking source, pairs nicely with solar thermal, and is semi-renewable.
NIMBY and Greenpeace. Coal is responsile for the deaths of over 40,000 people a year in the United States due to various pollution caused by Coal plants. I wonder why people are screaming so loud to stop nuclear plants when Coal is by far the greater danger to life and the environment.
Show me a single anti-nuclear environmentalist of note who's not equally also anti-coal. This false dichotomy you people paint gets old fast. Almost all anti-nuclear environments are pro-solar, pro-wind, pro-wave, and pro-geothermal. Many are pro-tidal. A good number are pro-natural gas (esp. for peaking). Some are pro-hydro. Almost none are pro-coal. I doubt you could find a single one of note.
Again, red tape.
Again, for good reason. And see my post higher up comparing the radiation exposure from a single nuclear accident with that from all of the coal emissions in US history, as well as the part above about your false dichotomy. Anti-nuclear environmentalists want both coal *and* nuclear shut down. And your blaming them for the lack of new nuclear plants in the past two decades makes them out to be way more powerful than they actually are. There's been no *market demand* for building them.
The relative impotence of the environmental movement over the past several decades is reflected in their stunning lack of success at blocking coal plants, mountaintop removal, and valley filling despite high profile campaigns to do so during this time.
The primary concerns about nuclear power plants and health aren't from their normal operation; they're from their failure modes. To put it into perspective: Chernobyl released several billion curies (the exact number is widely disputed, but somewhere in that range; the core contained 9 billion). Coal ash is 5-6 picocuries per gram. That's 4.98951607e-6 curies per ton. US coal plants currently produce about 125 million tons of various ashes each year. Let's say 7.5 billion tons over the US's history -- does that sound reasonable? That means that the US's coal plants have emitted about 40,000 curies of radioactive material. So the single failure at Chernobyl utterly dwarfs the radiation emissions of all of the US's coal plants running throughout the country's entire history. And Chernobyl was hardly the only nuclear accident in world history -- Chazhma Bay, Windscale, Tomsk-7, etc.
Nuclear carries both great potential and great risk. That's why significant caution is required.
Actually, no it's not. First off, HTO is only one of the forms tritium can take, and the one with the shortest-term residency in the body. When the tritium becomes bioincorporated before ingestion, it tends to last in the body much longer. Secondly, 2.4 liters/day (average adult water consumption) * 2 millicuries/liter (the level of this leak) * 64 mrem/mcurie (exposure for HTO ingestion) = 307 mrem/day of ingestion. So once you account for the rate it leaves the body, that means a constant exposure to about 2 rem, or ~730 rem/year. The normal average dose is 360 mrem/year. So if this was your water supply, you'd be exposed to about 2000 times the normal human's exposure to radiation. If it was diluted at a 100:1 ratio, it'd still 20x your normal exposure. This, as mentioned, assumes that the tritium remains as HTO rather than becoming bioincorporated.
Nuclear "red tape" is reasonable. "With great power comes great responsibility". Nuclear failures have the potential to render uninhabitable vast swaths of land. A high level of caution is called for. That's not saying that nuclear is a bad idea -- only that it needs to be highly regulated and monitored.
As for it being "the cheapest source of electricity", the market sure didn't consider it to be for the past couple decades, and a lot of the recent market enthusiasm for it has cooled off as some of the new plants have been running behind schedule and over budget. Nuclear has low operating costs, but their capital costs are high, and those must be amortized. And a lot of that is amortizing the cost of risk.
Nuclear power plant capital repayment is presaged on very long lifespans.
I just don't see PWRs as the future. Now, I'd be more interested if we were talking about lead-bismuth breeders like BREST. Someone, please go ahead and NIMBY this:
* A "bathtub" design sunk into the ground, so for any radiation to reach people outside the plant, it has to go through an awful lot of ground first. * A breeder, so the amount of fuel available is huge and the burnup is great * A type of reprocessing that, combined with the high burnup of the breeder, means that any waste subject to burial has less radiation than the natural uranium mined for the plant's fuel. * Can burn existing nuclear waste * Highly proliferation-resistant * The coolant is a molten lead-bismuth mixture, unlike sodium in normal breeders. I.e., it's not very reactive and has a high boiling point. * The coolant can circulate passively in a failure scenario, but unlike with PBMRs, there *is* a containment structure, there's no graphite involved, and no failure mode involves oxygen reaching the fuel. * Worst case scenario? Your core is *already* entombed in lead, by default! You just need to let it solidify.
That's reminiscent of the furor over Scott Brown voting for the jobs bill. The conservative blogs are ready to burn him at the stake. Lots of people furiously asking when he took his 30 pieces of silver, stuff like that. For what -- for voting for a bill primarily comprised tax cuts for small businesses? Isn't that the sort of stuff that conservatives like to vote for?
It's sad what our political discourse has turned into.
..It tested at the leak at a whopping 2 million picocuries, which is a bullshit measurement that's clearly chosen because it's more shocking than 2 microcuries. 2 microcuries is about what you'd get for a basic thyroid test at the docs office.
1) That's 2 million picocuries *per liter*. The average adult human drinks 2.4 liters of water *per day*. 2) The human body naturally contains about 0.1 microcuries. So yes, combining that with above, this amount would be significant if it were to contaminate drinking water. 3) A curie is a very large unit -- 37 billion Bq. 4) You don't get "curies" of radiation during a test; curies are a measure of emission *rate*. That's like saying that your meter reader recorded that you used 80kW of electricity this month. Radiation doses are properly measured in gray or sievert (formerly rad and rem).
You always see this: in any field where there's lots of announcements about new tech, people berating them for that tech not showing up instantly on the consumer market... while meanwhile, the consumer market *does* continue to advance behind the scenes.
No, the laws of thermodynamics apply to *everything*, and are limited by entropy. 2 H2 + O2 has a greater entropy state than 2 H2O. Therefore, entropy must increase in the system elsewhere (i.e., waste heat).
There is no getting around the Second Law. Carnot's Law for heat engines is just a consequence of the Second Law. If you can break the Second Law, you can create a perpetual motion machine, so it's kind of an issue.
If "they" here refers to Amazon/EC2, that's what I was talking about: EC2 doesn't do any kind of scaling for you. If not, then who does it refer to?
I was under the impression that EC2 starts and stops new instances of your system, up to whatever limit you specify, based on load.
* Bake the data directly into your AMI, if it's small and doesn't change often.
* Store the data on an EBS volume (basically a NAS unit): decent performance and transparent to your apps, but can potentially fail and so should be backed up regularly to S3. Can only be attached to one running instance at a time, though of course you can run an NFS server on top of it if you want.
* Store the data on S3 directly: slow and requires special tools to access, but you can just throw data on there and forget it, as well as access it from all your instances.
Thanks -- that really helps with my storage questions. I appreciate it.:)
And there are also benefits to local generation. According to the articles, the Bloom Box is supposed to be a more efficient electric generator than a full-size power plant.
I already mentioned that. It's not about the size; it's about the generator type. And the difference isn't really that much.
It becomes even more efficient (at the site) without transmission losses.
Transmission losses are tiny. They average 7.2% in the US, and commercial power needs will have lower losses than average.
It's more nimble to changes in fuel prices (switch from natural gas to syn-gas, ethanol, etc) than a power plant
[[Citation Needed]]. Really, is someone going to take your natural gas line leading up to your building and start pushing ethanol through it? Centralized locations are much more adaptive to change than countless smaller sites that need everything distributed to them in far lesser quantities.
as well as being under your own control
Relevance to the discussion of pricing?
You also only get one markup for buying the hardware and recurring costs for maintenance, rather than both of those costs (maintenance subsidized) with an additional markup for the power companies profits.
The power company's hardware is a *lot* cheaper per watt. Orders of magnitude.
An equivalent of the Carnot limit exists for thermochemical cycles as well -- Gibbs free energy. The Second Law is not merely constrained to heat engines. H2 + O2 has a higher entropy state than H2O (esp. if the H2O is liquid or solid). So the reaction is a reduction in entropy. It must correspond with a greater increase in entropy for the equation to be balanced -- i.e., waste heat. The maximum theoretical efficiency for a fuel cell can be calculated as described here. You'll notice that it *does* depend on the operating temperature. Also note that in practice, fuel cells don't get anywhere close to their theoretical, esp. in real-world conditions where you're not running them at low loads and where they're not being fed air rather than pure, pre-compressed oxygen as one of the feedstocks, plus all of the parasitic losses.
Not necessarily. Many things favor central generation, including end-user distribution infrastructure, bulk buys, centralized maintenance, and -- here's a big one -- much longer lifespans than SOFCs.
And even if that wasn't the case, and even if we assume your scenario, the payback would be *at best* 30 years (not considering the time-value of money). To get under 30 years, you have to assume that the electricity costs rise *faster* than the NG costs, not at the same speed.
That's typical for ceramic fuel cells. These are very different from the PEMFCs that go into cars. They perform better and are cheaper per unit power than PEMFCs, but they generally only work for bulky, stationary installs.
Betting on natural gas prices rising slower than electricity prices seems a pretty dumb bet to me, personally.
Slashdot Mirror
Yeah, that software that they just released *yesterday* catches them up to, what, 2003? Here's what games of five years ago (2005) looked like, and that beats the heck out of SL's latest client.
And yeah, cue SL's standard excuses in three, two, one... "Hey, that's not fair! Second life has to download its content!" Sorry, but that only applies to textures and poly counts. SL has never had problems with texture quality or poly count. It's always suffered from a pathetically low quality renderer to display them. Heck, they're *just now* adding support for displaying plausible normals shading in *2010*? No cast shadows (at least in the screenshots I've come across)? I mean, what is this, the dark ages?
I just hope that things be quite different from SL by then :)
By that, do you mean, "with graphics that don't look like they're from the late '90s", or do you mean "with not so many flying penises"?
You're off by a factor of about 20.
You're acting like there's a single estimate of how much release there was. The Russians and the IAEA estimated 100m curies in 1986. Argonne estimated 30% of the core (3B curies) in 1986. Lawrence Livermore estiamted 50% of the core (5B curies) in 1986. The supervisor of the Russian cleanup effort estimated 80% of the core (6.4B curies) in 1991. And the Union of Concerned Scientists estimates that it all was vaporized (9B curies).
Since most of the estimates are in the billions range, that's what I used.
The problem is that the anti-nuclear types tend to want perfect technology and that of course can never exist.
Wrong. They want trillions to flow into producing today's tech, en-masse, right now. End of story.
Solar cell production releases NF3 a super greenhouse gas.
NF3 is only just now becoming recognized as a greenhouse gas for a reason -- such tiny amounts leak that they're barely even measurable. And it's being phased out anyway in favor of fluorine. And is only used in thin film photovoltaics, not photovoltaics in general, let alone solar thermal which is what most people refer to when they're talking about large-scale solar power plants.
Fusion requires Tritium (for now) which is only produced using nuclear fission reactors and a lithium blanket.
Fusion itself can breed tritium.
Geothermal can cause earthquakes. ... Wind kills birds and bats.
Two very closely related nonsense claims. When will people stop pushing this FUD? Let's start with wind. There was ONE -- read it, ONE -- wind farm, built early on, that had bird kill problems: Altamont Pass. They built a bunch of fast-turning bladed turbines with towers shaped like perfect perches at low altitude in the middle of a bird flyway. I mean, if you were trying to design a raptor cuissanart, that's pretty much the design you'd use. NO wind farm today gets approved without a bird impact survey, and even if that wasn't the case, the way modern turbines are built has way less impact (much taller towers, slower rotation, and the towers are unattractive as perches). The bird kill rates of modern wind turbines are utterly dwarfed by that of other human activity, even taking into account the scale of wind farms versus the scale of other activities. The Audubon Society itself now supports wind farms because they impact birds less than most other forms of power generation. So, in short, this is FUD, plain and simple. Stop spreading it.
The exact same thing is happening with EGS, which is about a decade behind wind in terms of commercialization. One EGS project drilled straight into a faultline that had previously destroyed Basel, Switzerland, and started fracturing it. This caused a *small*, but tangible earthquake. EGS always causes earthquakes, but they're usually too small to even feel. Will people drill into active faults right next to cities anymore? Of course not; they learned the lesson with Basel. But now EGS is irreparably smeared in the same way wind was by Altamont Pass.
They all have problems and we will get no where by attacking every single technology that we have.
Right -- fair and balanced, as if bogus claims are on equal footing as valid ones. "Some say" wind kills birds. "Some say" coal causes health problems. Both have problems -- there's your equivalence, right?
You are out of touch with reality. Completely. Do you even comprehend how much pollution Coal plants produce?
Do you even comprehend how much radioactive material several to dozens of billions of curies per reactor is? Nuclear has great potential but also great risk. It *only makes sense* that it is so heavily regulated. And if such heavy regulations make it uneconomical? Tough cookies. There are plenty of other zero-carbon generation techs already available.
Wow -- this post is almost as though it came from an alternative universe where I *didn't* list what sort of power environmentalists typically support.
Wind power is noise pollution and dead birds. Geothermal causes earthquakes. Solar pollutes water streams because of the solvents used in silicon manufacturing. So yeah, make sure to keep burning that natural gas.
Almost all prominent environmentalists support wind. Most of the opposition from wind comes from rich people who are afraid of having their property values decreased by turbines. Even the audubon society supports wind power, despite the "OMG THEY KILLS BIRDZ!" FUD. And the "noise" issue is -- let's be blunt -- a lie. I've been to a big wind farm on a windy day. You can barely hear the things when you're standing at their base.
Geothermal causes tiny earthquakes, usually impossible to feel. The biggest one on record occurred from *drilling into a freaking fault line that had once destroyed the city* (something nobody's going to be dumb enough to do again), and even still, the quake was quite small. The opposition to EGS is quite limited and definitely isn't coming from the environmental movement.
Almost all prominent environmentalists support solar. The amount of waste from solar manufacturing is tiny compared to the amount of waste in pretty much every other kind of power generation on a per-watt-hour basis. And is way overplayed, at that.
Natural gas isn't perfect, but it's a good low carbon/low health-pollution peaking source, pairs nicely with solar thermal, and is semi-renewable.
Curses! Bill GATES, you win this time! But next time, my human-powered doomsday machine will SEQUESTER its carbon! Muahahaha!!
That's what I get for not proofreading. Both numbers above should read microcuries, both in terms of per liter and mrem per microcurie.
NIMBY and Greenpeace. Coal is responsile for the deaths of over 40,000 people a year in the United States due to various pollution caused by Coal plants. I wonder why people are screaming so loud to stop nuclear plants when Coal is by far the greater danger to life and the environment.
Show me a single anti-nuclear environmentalist of note who's not equally also anti-coal. This false dichotomy you people paint gets old fast. Almost all anti-nuclear environments are pro-solar, pro-wind, pro-wave, and pro-geothermal. Many are pro-tidal. A good number are pro-natural gas (esp. for peaking). Some are pro-hydro. Almost none are pro-coal. I doubt you could find a single one of note.
Again, red tape.
Again, for good reason. And see my post higher up comparing the radiation exposure from a single nuclear accident with that from all of the coal emissions in US history, as well as the part above about your false dichotomy. Anti-nuclear environmentalists want both coal *and* nuclear shut down. And your blaming them for the lack of new nuclear plants in the past two decades makes them out to be way more powerful than they actually are. There's been no *market demand* for building them.
The relative impotence of the environmental movement over the past several decades is reflected in their stunning lack of success at blocking coal plants, mountaintop removal, and valley filling despite high profile campaigns to do so during this time.
2.4 liters per day at 2 picocuries per liter is 4.8 picocuries per day.
Why did you just drop the dose by six orders of magnitude?
The primary concerns about nuclear power plants and health aren't from their normal operation; they're from their failure modes. To put it into perspective: Chernobyl released several billion curies (the exact number is widely disputed, but somewhere in that range; the core contained 9 billion). Coal ash is 5-6 picocuries per gram. That's 4.98951607e-6 curies per ton. US coal plants currently produce about 125 million tons of various ashes each year. Let's say 7.5 billion tons over the US's history -- does that sound reasonable? That means that the US's coal plants have emitted about 40,000 curies of radioactive material. So the single failure at Chernobyl utterly dwarfs the radiation emissions of all of the US's coal plants running throughout the country's entire history. And Chernobyl was hardly the only nuclear accident in world history -- Chazhma Bay, Windscale, Tomsk-7, etc.
Nuclear carries both great potential and great risk. That's why significant caution is required.
Actually, no, its pretty harmless.
Actually, no it's not. First off, HTO is only one of the forms tritium can take, and the one with the shortest-term residency in the body. When the tritium becomes bioincorporated before ingestion, it tends to last in the body much longer. Secondly, 2.4 liters/day (average adult water consumption) * 2 millicuries/liter (the level of this leak) * 64 mrem/mcurie (exposure for HTO ingestion) = 307 mrem/day of ingestion. So once you account for the rate it leaves the body, that means a constant exposure to about 2 rem, or ~730 rem/year. The normal average dose is 360 mrem/year. So if this was your water supply, you'd be exposed to about 2000 times the normal human's exposure to radiation. If it was diluted at a 100:1 ratio, it'd still 20x your normal exposure. This, as mentioned, assumes that the tritium remains as HTO rather than becoming bioincorporated.
Nuclear "red tape" is reasonable. "With great power comes great responsibility". Nuclear failures have the potential to render uninhabitable vast swaths of land. A high level of caution is called for. That's not saying that nuclear is a bad idea -- only that it needs to be highly regulated and monitored.
As for it being "the cheapest source of electricity", the market sure didn't consider it to be for the past couple decades, and a lot of the recent market enthusiasm for it has cooled off as some of the new plants have been running behind schedule and over budget. Nuclear has low operating costs, but their capital costs are high, and those must be amortized. And a lot of that is amortizing the cost of risk.
Nuclear power plant capital repayment is presaged on very long lifespans.
I just don't see PWRs as the future. Now, I'd be more interested if we were talking about lead-bismuth breeders like BREST. Someone, please go ahead and NIMBY this:
* A "bathtub" design sunk into the ground, so for any radiation to reach people outside the plant, it has to go through an awful lot of ground first.
* A breeder, so the amount of fuel available is huge and the burnup is great
* A type of reprocessing that, combined with the high burnup of the breeder, means that any waste subject to burial has less radiation than the natural uranium mined for the plant's fuel.
* Can burn existing nuclear waste
* Highly proliferation-resistant
* The coolant is a molten lead-bismuth mixture, unlike sodium in normal breeders. I.e., it's not very reactive and has a high boiling point.
* The coolant can circulate passively in a failure scenario, but unlike with PBMRs, there *is* a containment structure, there's no graphite involved, and no failure mode involves oxygen reaching the fuel.
* Worst case scenario? Your core is *already* entombed in lead, by default! You just need to let it solidify.
That's reminiscent of the furor over Scott Brown voting for the jobs bill. The conservative blogs are ready to burn him at the stake. Lots of people furiously asking when he took his 30 pieces of silver, stuff like that. For what -- for voting for a bill primarily comprised tax cuts for small businesses? Isn't that the sort of stuff that conservatives like to vote for?
It's sad what our political discourse has turned into.
..It tested at the leak at a whopping 2 million picocuries, which is a bullshit measurement that's clearly chosen because it's more shocking than 2 microcuries. 2 microcuries is about what you'd get for a basic thyroid test at the docs office.
1) That's 2 million picocuries *per liter*. The average adult human drinks 2.4 liters of water *per day*.
2) The human body naturally contains about 0.1 microcuries. So yes, combining that with above, this amount would be significant if it were to contaminate drinking water.
3) A curie is a very large unit -- 37 billion Bq.
4) You don't get "curies" of radiation during a test; curies are a measure of emission *rate*. That's like saying that your meter reader recorded that you used 80kW of electricity this month. Radiation doses are properly measured in gray or sievert (formerly rad and rem).
Tritium is pretty safe outside your body. Not so safe inside it.
And meanwhile, consumer solar tech *does* continue to advance. Have you priced solar panels lately? Solarbuzz lists:
Lowest Mono-crystalline Module Price: $2.37/Wp
Lowest Multi-crystalline Module Price: $1.98/Wp
Lowest Thin Film Module price: $1.76/Wp
You always see this: in any field where there's lots of announcements about new tech, people berating them for that tech not showing up instantly on the consumer market... while meanwhile, the consumer market *does* continue to advance behind the scenes.
No, the laws of thermodynamics apply to *everything*, and are limited by entropy. 2 H2 + O2 has a greater entropy state than 2 H2O. Therefore, entropy must increase in the system elsewhere (i.e., waste heat).
There is no getting around the Second Law. Carnot's Law for heat engines is just a consequence of the Second Law. If you can break the Second Law, you can create a perpetual motion machine, so it's kind of an issue.
If "they" here refers to Amazon/EC2, that's what I was talking about: EC2 doesn't do any kind of scaling for you. If not, then who does it refer to?
I was under the impression that EC2 starts and stops new instances of your system, up to whatever limit you specify, based on load.
* Bake the data directly into your AMI, if it's small and doesn't change often.
* Store the data on an EBS volume (basically a NAS unit): decent performance and transparent to your apps, but can potentially fail and so should be backed up regularly to S3. Can only be attached to one running instance at a time, though of course you can run an NFS server on top of it if you want.
* Store the data on S3 directly: slow and requires special tools to access, but you can just throw data on there and forget it, as well as access it from all your instances.
Thanks -- that really helps with my storage questions. I appreciate it. :)
And there are also benefits to local generation. According to the articles, the Bloom Box is supposed to be a more efficient electric generator than a full-size power plant.
I already mentioned that. It's not about the size; it's about the generator type. And the difference isn't really that much.
It becomes even more efficient (at the site) without transmission losses.
Transmission losses are tiny. They average 7.2% in the US, and commercial power needs will have lower losses than average.
It's more nimble to changes in fuel prices (switch from natural gas to syn-gas, ethanol, etc) than a power plant
[[Citation Needed]]. Really, is someone going to take your natural gas line leading up to your building and start pushing ethanol through it? Centralized locations are much more adaptive to change than countless smaller sites that need everything distributed to them in far lesser quantities.
as well as being under your own control
Relevance to the discussion of pricing?
You also only get one markup for buying the hardware and recurring costs for maintenance, rather than both of those costs (maintenance subsidized) with an additional markup for the power companies profits.
The power company's hardware is a *lot* cheaper per watt. Orders of magnitude.
An equivalent of the Carnot limit exists for thermochemical cycles as well -- Gibbs free energy. The Second Law is not merely constrained to heat engines. H2 + O2 has a higher entropy state than H2O (esp. if the H2O is liquid or solid). So the reaction is a reduction in entropy. It must correspond with a greater increase in entropy for the equation to be balanced -- i.e., waste heat. The maximum theoretical efficiency for a fuel cell can be calculated as described here. You'll notice that it *does* depend on the operating temperature. Also note that in practice, fuel cells don't get anywhere close to their theoretical, esp. in real-world conditions where you're not running them at low loads and where they're not being fed air rather than pure, pre-compressed oxygen as one of the feedstocks, plus all of the parasitic losses.
On the other hand, they can be poisoned by sulfur.
Not necessarily. Many things favor central generation, including end-user distribution infrastructure, bulk buys, centralized maintenance, and -- here's a big one -- much longer lifespans than SOFCs.
And even if that wasn't the case, and even if we assume your scenario, the payback would be *at best* 30 years (not considering the time-value of money). To get under 30 years, you have to assume that the electricity costs rise *faster* than the NG costs, not at the same speed.
They tend to be more efficient than NG power plants... although not by much. This comes at notably greater cost, complexity, and shorter lifespan.
That's typical for ceramic fuel cells. These are very different from the PEMFCs that go into cars. They perform better and are cheaper per unit power than PEMFCs, but they generally only work for bulky, stationary installs.
Betting on natural gas prices rising slower than electricity prices seems a pretty dumb bet to me, personally.