They aren't strap-on boosters, they're liquid tanks for the vectoring system for the solid-fuel core first stage. It works by injecting a strontium-based liquid into the nozzle to change the burn characteristics. It's crude by the standard of other solid-fuel launchers like Vega and Epsilon which use nozzle deflection systems but it works and it's cheap.
The solid-fuel strap-ons are considerably larger than the liquid tanks. You can see a diagram of what they look like on the Wikipedia page about the PSLV.
Engineering-wise the PSLV is a bit of a dog's breakfast -- the first stage and third stage are solid-fuel as are the strap-ons (if flown), the second and fourth stages use UDMH and N2O4, old Cold-War missile hypergolic fuels that have generally fallen out of favour with other launcher builders. The one advantage they have is storability with no cryogenic handling required on the pad, not even LOX.
ISRO have another launcher, the GSLV which is more advanced in design with a nozzle-vectoring solid-fuel core and solid strap-ons but it's got a UDMH/N2O4 second stage and a new fully-cryogenic upper stage. Interestingly enough they flew a prototype crew capsule with it on a suborbital mission last year.
This PSLV launch was a "core alone" configuration without any strap-on boosters. In its max configuration with six solid-fuel strap-ons it's good for about 3 tonnes to LEO.
I've got some NiMH AA batteries (Sanyo Eneloop) I bought in 2007 that I still use regularly. They've been through a couple of hundred charge cycles or so in cheap non-intelligent chargers of various kinds. I don't know what capacity losses they've suffered but they still do the job in flashlights and a digital camera. They're the low-self-discharge type that holds a charge for long periods and they're still doing that part of the job too, even after nearly a decade. I expect I'll still be using them for another few years.
Lithium batteries don't last that long from my experience. They need careful charging and balancing, they can't be deeply discharged without suffering degradation and they can catch fire if things go wrong. Their capacity in terms of volume and weight is impressive but it comes with a lot of extra baggage.
The battery tech that impresses me is nickel-iron. NiFe cells can be deep-discharged repeatedly without effect and they last in service for decades with some basic maintenance. They're not compact or light though and for some unknown reason they are incredibly expensive.
The worrying thing is that, as you say, there are several competing consortia which is good for competition but it will probably mean the UK will end up with a range of different reactor designs rather than one or two standard models. This will cause problems for fuel manufacture, operations, staff training, ancillary equipment procurement and so on.
The existing French reactor fleet is mostly a standardised M910 design with some tweaks here and there -- they're in the process of replacing a lot of steam generators in a fleet-wide mid-life maintenance operation and they've been able to pre-order more than forty identical steam generators to be manufactured in a rolling contract, meaning big savings and some flexibility in the replacement tempo.
In contrast the UK is planning to build a couple of EPRs, some first-of-their-kind ESBWRs, a couple of AP-1400s and maybe even some Hualong-1s, the Chinese home-sourced reactors (based on the ACPR-1000 design which has its roots in the French M910 design). That means mid-life kickers in thirty or forty years will be a dog's breakfast of replacement hardware, different for each class of reactor.
Who the fuck do you think will be paying for the electricity from this power station?
Industries, folks on low income who don't pay taxes, Irish people (via one of the interconnectors that feeds British electricity to Ireland) and regular consumers here in the UK. The construction and operating costs don't come out of taxes so "taxpayers" don't have any financial skin in the game.
The going rate in the UK for unpredictable and variable onshore wind generation, guaranteed by law, is about £95 per MWhr or a little over the asking price for new baseload nuclear. Offshore wind in the UK is guaranteed about £145 per MWhr. The folks planning to build new nuclear plants in the UK simply want price parity with the other non-carbon generators. At least one large offshore wind project was recently cancelled because £145 per MWhr wasn't thought to be enough return for the construction and operating costs and predicted profit over the project's lifespan.
The headline strike price of £92.50 per MWHr for the new nuclear plants is a maximum; it may well be reduced in the future depending on operating costs etc. and the deal runs out in 35 years time. At that point the plants will have been paid for and fresh price negotiations will ensue since those plants will still have an operating life of more than 30 years ahead of them. The initial high strike price is to ensure that if the reactors are built and deliver the non-carbon electricity they are supposed to the reactor builders won't lose out financially in the medium term.
As for taxpayers they are not involved. The funding and construction will be carried out independently of the British government and no tax monies will be used. In return the builders want a price guarantee to build generating capacity which will supply baseload electricity for most of the next hundred years or so.
I'd be stunned to discover Apple ISN'T working on a touch-screen version of OS/X. Whether that ever makes it out of the labs into the shops is another matter, more due to marketing, user satisfaction and a number of other factors and if it doesn't we'll never know. Whether it bears any resemblance to iOS internally or in the UX is another matter. MS has pushed hard for an integrated UX for Windows, Apple might decide to diverge the UX experience instead to maintain usability.
Th Old Fogies assembled here will remember the G5 processor Apple machines and how it seemed there would never be an Intel version of an Apple computer, and then suddenly one day there is was, ready to go on sale. It was obvious in hindsight Apple had been working on porting their OS to Intel hardware for years, just in case and when the G5 got too far behind the curve they were ready (unlike the Altivec fanbois who swore they'd never buy an Intel Apple...). I'd even bet there's iOS builds for Intel Atom platforms in the back offices just in case the ARM designs powering their tablets and phablets run out of steam.
Most servers don't need anything near 6TB of storage, let alone 6TB of (D)RAM.
Many servers do need access to that sort of storage (and a hundred times more) and it would help if those servers run the same OS on similar hardware as other servers with less demand do. The alternative is for the sort of species differentiation that hobbles High Performance Computing (HPC) because there are few standards and a lot of hand-written system code flying in close formation, different on each machine.
I expect, if anything comes of this 3D flash RAM that there will be a flurry of Good Intentions by assorted BigCorps and LittleStartups which will evolve into two or three usable hardware and OS implementations that everyone coalesces around and then we move forward from there. The intervening period will be fun to observe, from a safe distance at least.
Oh, and are you claiming that 6TB should be enough RAM for anyone? There's a Slashdot meme for that...
Seems like we're going to find out soon, 6TB of addressable non-volatile ram sounds like a game changer
A server system really needs to be able to address hundreds or thousands of terabytes of storage, not just six. That's what I meant by the server system designers having to revamp the basic concepts of a computer with RAM separate from secondary storage (HDDs or SSDs on a separate bus) to one with a "flat" storage architecture. The OS will have to change too to take account of the blurring or total elimination of the difference between addressable RAM space and secondary storage, and getting that to work reliably and efficiently will take some time and involve some serious mistakes too, I expect. Sharing storage over a network will also be a challenge, as will backup processes.
If this tech makes it into the marketplace at reasonable prices it's not going to be hanging off SAS-12 cables or any other serial links at that rate, it will be more tightly integrated with the CPU bus to deliver on the R/W and access speed improvements. Even PCI is a possible bottleneck if this 3D flash can deliver what Intel are claiming for it. Comparing its performance to DRAM is a "tell" and shows what they're thinking; this may be the fabled "non-volatile RAM" solution that's been the Holy Grail researchers have been trying to develop pretty much ever since RAM was invented. (Yes, I know there are battery-backed-up RAM solutions that claim to be non-volatile but they're only non-volatile until the battery power runs out).
There's a whole raft of other things to consider before this tech changes the IT world -- how much does it cost, how many separate fabs can produce it so there's no single-point-of-failure that could constrain supply, how much redesign of existing chipsets is required to integrate it into current server/workstation/mobile phone designs, what's the failure rate in service, power dissipation and cooling requirements etc.
Saying that the demo suggests it can be implemented into existing platforms with little difficulty. Of course as Napoleon once said, "There are lies, damned lies and rigged demos." Time will tell.
I worked as an Amazon customer service rep last year. All calls are recorded and retained, as are emails, web chats and Kindle Mayday video chats. They use the weasel-words "may be recorded" since the recording system might break.
Hundreds actually, if you're only counting atmospheric tests although some were underwater and a handful in the stratosphere. The US fired off about 220 atmospheric tests, going to underground testing after the Sunbeam series of tests in Nevada in 1962. The Soviets did a lot fewer; the US has carried out more than half of all the nuclear tests by all nations.
Actually, no China isn't putting a crapton of resources into LFTR. It's actually putting about a hundred billion bucks into building a lot of PWRs with more to come in the next ten years or so if they continue the way they're going. Chinese researchers are looking at molten-salt reactor technologies but no significant money has been spent, same with fast-spectrum reactors like the Russian BN-series designs which at least exist in the real world. They're not building any molten-salt reactors, they have no plans to build such a reactor, there are no components for such a reactor being ordered or manufactured. There is some theoretical research and computer modelling being carried out, that's all. The only experimental reactor they're actually spending money on building is a pebble-bed design, the HTR-PM.
The Oak Ridge molten-salt reactor never used thorium, ever. It ran with U-233 and later with U-235 but never thorium. I blame the Powerpoint Rangers for conflating the purely theoretical LFTR with the Oak Ridge reactor (which was only one of many possible reactor concepts being tried out back in the 1960s).
No-one wants a beta reactor, they want something that will predictably generate electricity at a reasonable cost. That's why virtually every reactor being built today, including the four new-build AP-1000s at Vogtle and Summer as well as the dozens of reactors under construction in China and elsewhere are an evolutionary development of the PWR/BWR concept. The design effort has been concentrated in ever greater cost efficiencies and safety enhancements in larger and more efficient designs generating more electricity per reactor unit.
There are a couple of new power reactor designs not based on the well-tested PWR/BWR concept but even they are evolutionary; the new Russian BN-800 fast reactor which started up last year is an offshoot of the BN-350 and BN-600 reactors built in the 1970s and the Chinese modular pebble-bed HTR-PM now being built is using technology licenced from the Germans who had the ill-fated AVR and HTHR-300 pebble-bed reactors operating in the 1980s.
The capital costs of a reactor build are high because it's an expensive piece of construction, not simply because of delays etc. Every other large project including coal-fired and natural-gas generating plants also have to spend money up front preparing plans, covering the likely environmental impacts and dealing with protests.
Nearly all modern-build Gen-IIa and Gen-III reactors like the AP-1400, the EPR, ESBWR etc. are significantly larger than the original Gen-1 and Gen-II designs, each generating well over 1GW of electricity (the EPRs when they are complete will produce 1.6GW). That takes a lot of concrete and steel for containment, bigger turbine-generator sets, a larger reactor vessel etc. Putting that all together takes longer to complete even if everything goes right first time -- the Chinese are turning out their enhanced Gen-II APR-1000s in about 6 years from breaking ground to first grid connection but they've got a tested production line in place for groundworks, components and construction.
The good long-term news is that new-build Gen-IIIs will operate for more than 60 years; the Russians just produced a reactor vessel that they claim will last in service for a century and more. This improves the financial viability of a reactor project even though they cost a chunk of money up front.
Putting all these technologies together in one design is nearly trivial.
I'm sorry but that statement makes me die a little inside. That sort of reasoning is why I call the thorium boosters Powerpoint Rangers.
Yes I didn't watch any of the thirty or forty half-hour-long Powerpoint presentations on Youtube because none of them actually show hardware in operation, they show TED talks by graduate students and Twue Beweivers with glossy brochures, simplified diagrams and wishful thinking and a lot of "and then a miracle occurs" in the middle. I've seen some of the early attempts to convince folks that thorium was the Answer to non-existent problems before I decided life was too short and went TL;DR on the subject. To engage my interest it will take working hardware, a real molten-salt mostly-thorium reactor that runs for a year or more continuously (or at least 90% uptime) with significant thermal output (10 MW would be a good start, even 7 MW like the ORNL uranium-fuelled reactor produced) and a plan to decommission such a reactor in the future after 50 or 60 years of neutron irradiation and radiochemical contamination of components in direct contact with molten fuel (not something that causes a problem with existing solid-fuel reactors, even breeders).
the people that are making it happen
Until they have funding to build and operate a complete operational test reactor for at least a few years nobody is "making it happen", they're selling dreams. Sorry.
I hate to break it to you but China is not building any kind of a molten salt reactor (MSR) never mind one where thorium makes up part of the fuel stream. They've talked about it, yes but then again a lot of Powerpoint Rangers have done so over the past ten years and more. They're not funding, pouring concrete or bending metal on an MSR, the one true sign that anyone is taking the production of a prototype reactor seriously. They have discussed building a BN-series fast-spectrum reactor as well but no funding, no metal bent, no concrete poured on that project either which is a shame as it's a very interesting concept with the ability to burn waste actinides with minimal fuel reprocessing.
India's interest in thorium is as an adjunct fuel for their heavy-water reactors mixed in with a lot of 20%-medium-enriched uranium and plutonium. There are geopolitical reasons for this decision to work on using thorium as an adjunct fuel but at the moment they're building out several PWRs, mostly Russian VVER-series designs. Their fast-breeder is a conventional sodium-cooled design, not speciafically built to use thorium. It is outside NPT and IAEA safeguards so fuelling it could be a problem.
Experimental reactor designs actually being funded, built and operated in the real world are the Chinese HTR-PM helium-cooled pebble-bed reactor and the Russian BN-800 fast-spectrum reactor, both based on existing engineering predecessors (the ill-fated German HTHR-300, the Chinese HTR-10 and the ex-Soviet sodium-cooled BN-350 and BN-600).
Just so other folks understand, no-one has built and operated a thorium-fuelled molten salt breeder reactor so the design hasn't been tested as you claim. There was a molten-salt reactor fuelled with U-233 and later on U-235 at Oak Ridge but it never used thorium in any way. The Powerpoint Rangers pushing thorium claim it would be easy to add a breeding core to such a design and there would be no serious technical challenges involved despite the very high temperatures involved, the intense radiation environment required for the breeding process and its effects on structures carrying molten fuel. That last factor was not something plutonium breeder reactors have had to put up with and their success record is not good even using fixed arrays of solid fuel.
Thorium (specifically Th-232) doesn't fission. It has to be bred up into U-233 by absorbing a neutron which can then fission by being hit by another neutron releasing energy. Fission of U-233 releases an average of about 2.2 neutrons to carry out further breeding and fission. This breeding-fission process is a bit knife-edge compared to regular PWRs, BWRs and other uranium-fuelled reactors where only one neutron is required to fission a U-235 nucleus and produce 2+ more.
What recent developments in "thorium fission" can you point us at? There's a lot of Powerpoint presentations and glossy brochures being waved around by folks looking to make a buck from research funding and subsidies but no-one is bending metal and pouring concrete right now on anything based on thorium as a primary source of nuclear energy. There ARE some experiments with thorium going on; PWR-style fuel pellets with thorium mixed in with uranium and plutonium are being exposed in a test reactor in Norway at the moment and there's an experimental Chinese pebble-bed reactor which can use some thorium in the fuel pebbles but that's about it.
As for modularity, it's not a new thing in regular uranium reactors -- see the units used in submarines, icebreakers and large aircraft carriers for the past fifty years and more as a worked example. The Russians are building a "power barge" carrying a ship reactor to produce about 40MW of electricity for coastal communities in Siberia and the Chinese are pouring concrete on a commercial modular power reactor (about 105MWe) but it's going to be a pebble-bed design fuelled entirely by uranium and plutonium to begin with. It might use a small amount of thorium in the future but that's a long way off and its commercial viability is still to be proven. Previous attempts to commercialise pebble-bed reactors capable of using some thorium such as the German THTR-300 were not a success.
I recently put Cinnamon Mint (17?) on an old server I got cheap, it worked like a champ. Then I added a graphics card (an nVidia PCI-bus FX5200, to fit in one of the PCI slots since the server mobo doesn't have any PCI-e x16 slots). Mint refused to boot up into a windowed environment, instead it dumped me to a command line prompt and told me something about restarting MDM, whatever that is.
I replaced the Mint installation on the server with Windows 7. It works like a champ and it doesn't complain about the video card. Any setting and configurations for resolution etc. are done from a windowed GUI with clicky boxes and drop-downs rather than unhelpful command-line options. nVidia don't support the ancient FX5200 under Win7 but the legacy 64-bit Vista drivers went on without a quibble and run perfectly.
I can't recommend Mint if you want to ever change anything from your default installation. Windows is much better in that case, IME.
Charlie Stross wrote a short story, "Dechlorinating the Moderator" a while back about a convention of hobbyist particle physics geeks using stuff like this to produce Higgs bosons in a hotel's banqueting suite.
Even smarter, install two fake charging ports next to each other. One has an "Out of Order" sign on it. The other one says "FREE CHARGING!".
Wire them up so the "FREE" charger discharges the battery of anyone who plugs into it while feeding the power to the "Out of Order" charger your own electric car is plugged into.
A lot of iron and steel foundries in the UK had a particular model of spectrometer built in the 1960s for carrying out analyses of metal samples. This spectrometer had an option of a hard-wired electric typewriter (not an IBM Selectric, a Remington or similar) that would print out the results of an analysis in a simple table. This was before Centronics printers were readily available. Back in the early 80s I worked with a consulting metallurgist to add a box to tap into the signals from the analyser to the typewriter and present them to a parallel-port interface so they could be read by a desktop computer. From memory the drive signals were at 24V with some weird pulse combinations for shifts, code pages etc. that we decoded by trial and error -- the company making the analysers wouldn't hand out the information and we couldn't mess with their internals since they were covered by long-term maintenance agreements (decades and more, these were very expensive bits of kit). We got the interface to work and the foundries were very happy to pay us to get the information out in digital for eventually supporting ISO9000 chain-of-custody certification which was necessary given that some of the places that used these analysers were making single large castings for paper-making machinery worth a million bucks a pop.
How in the name of all stupid plot devices does each and every space suit, vehicle, structure and other large chunk of habitat equipment not have its own, independent up-link to the multiple Earth-Mars radio relays we already have in orbit around that planet?
Worked example -- the Apollo moonsuits had a short-range VHF link back to the Lunar Lander. From there the radio comms was via an S-band microwave link back to Earth. There was no direct link from the suits to the Command Module in orbit, even when it was above the horizon for the astronauts on the Lunar surface.
The digitising system used in the Surface Pro 1 and 2 was a Wacom-based design, it's now an nTrig in the Surface Pro 3. Both of them have pens that are powered by near-field from the digitising surface so they don't need separate power or charging at all. Since Apple are relying on their passive capacitive digitiser to work with the Pencil they can't power it that way.
Wacom provide a range of specialist stylii for artists such as an airbrush model, I'm not sure if nTrig do.
Slashdot Mirror
They aren't strap-on boosters, they're liquid tanks for the vectoring system for the solid-fuel core first stage. It works by injecting a strontium-based liquid into the nozzle to change the burn characteristics. It's crude by the standard of other solid-fuel launchers like Vega and Epsilon which use nozzle deflection systems but it works and it's cheap.
The solid-fuel strap-ons are considerably larger than the liquid tanks. You can see a diagram of what they look like on the Wikipedia page about the PSLV.
Engineering-wise the PSLV is a bit of a dog's breakfast -- the first stage and third stage are solid-fuel as are the strap-ons (if flown), the second and fourth stages use UDMH and N2O4, old Cold-War missile hypergolic fuels that have generally fallen out of favour with other launcher builders. The one advantage they have is storability with no cryogenic handling required on the pad, not even LOX.
ISRO have another launcher, the GSLV which is more advanced in design with a nozzle-vectoring solid-fuel core and solid strap-ons but it's got a UDMH/N2O4 second stage and a new fully-cryogenic upper stage. Interestingly enough they flew a prototype crew capsule with it on a suborbital mission last year.
This PSLV launch was a "core alone" configuration without any strap-on boosters. In its max configuration with six solid-fuel strap-ons it's good for about 3 tonnes to LEO.
I've got some NiMH AA batteries (Sanyo Eneloop) I bought in 2007 that I still use regularly. They've been through a couple of hundred charge cycles or so in cheap non-intelligent chargers of various kinds. I don't know what capacity losses they've suffered but they still do the job in flashlights and a digital camera. They're the low-self-discharge type that holds a charge for long periods and they're still doing that part of the job too, even after nearly a decade. I expect I'll still be using them for another few years.
Lithium batteries don't last that long from my experience. They need careful charging and balancing, they can't be deeply discharged without suffering degradation and they can catch fire if things go wrong. Their capacity in terms of volume and weight is impressive but it comes with a lot of extra baggage.
The battery tech that impresses me is nickel-iron. NiFe cells can be deep-discharged repeatedly without effect and they last in service for decades with some basic maintenance. They're not compact or light though and for some unknown reason they are incredibly expensive.
The worrying thing is that, as you say, there are several competing consortia which is good for competition but it will probably mean the UK will end up with a range of different reactor designs rather than one or two standard models. This will cause problems for fuel manufacture, operations, staff training, ancillary equipment procurement and so on.
The existing French reactor fleet is mostly a standardised M910 design with some tweaks here and there -- they're in the process of replacing a lot of steam generators in a fleet-wide mid-life maintenance operation and they've been able to pre-order more than forty identical steam generators to be manufactured in a rolling contract, meaning big savings and some flexibility in the replacement tempo.
In contrast the UK is planning to build a couple of EPRs, some first-of-their-kind ESBWRs, a couple of AP-1400s and maybe even some Hualong-1s, the Chinese home-sourced reactors (based on the ACPR-1000 design which has its roots in the French M910 design). That means mid-life kickers in thirty or forty years will be a dog's breakfast of replacement hardware, different for each class of reactor.
Who the fuck do you think will be paying for the electricity from this power station?
Industries, folks on low income who don't pay taxes, Irish people (via one of the interconnectors that feeds British electricity to Ireland) and regular consumers here in the UK. The construction and operating costs don't come out of taxes so "taxpayers" don't have any financial skin in the game.
The going rate in the UK for unpredictable and variable onshore wind generation, guaranteed by law, is about £95 per MWhr or a little over the asking price for new baseload nuclear. Offshore wind in the UK is guaranteed about £145 per MWhr. The folks planning to build new nuclear plants in the UK simply want price parity with the other non-carbon generators. At least one large offshore wind project was recently cancelled because £145 per MWhr wasn't thought to be enough return for the construction and operating costs and predicted profit over the project's lifespan.
The headline strike price of £92.50 per MWHr for the new nuclear plants is a maximum; it may well be reduced in the future depending on operating costs etc. and the deal runs out in 35 years time. At that point the plants will have been paid for and fresh price negotiations will ensue since those plants will still have an operating life of more than 30 years ahead of them. The initial high strike price is to ensure that if the reactors are built and deliver the non-carbon electricity they are supposed to the reactor builders won't lose out financially in the medium term.
As for taxpayers they are not involved. The funding and construction will be carried out independently of the British government and no tax monies will be used. In return the builders want a price guarantee to build generating capacity which will supply baseload electricity for most of the next hundred years or so.
I'd be stunned to discover Apple ISN'T working on a touch-screen version of OS/X. Whether that ever makes it out of the labs into the shops is another matter, more due to marketing, user satisfaction and a number of other factors and if it doesn't we'll never know. Whether it bears any resemblance to iOS internally or in the UX is another matter. MS has pushed hard for an integrated UX for Windows, Apple might decide to diverge the UX experience instead to maintain usability.
Th Old Fogies assembled here will remember the G5 processor Apple machines and how it seemed there would never be an Intel version of an Apple computer, and then suddenly one day there is was, ready to go on sale. It was obvious in hindsight Apple had been working on porting their OS to Intel hardware for years, just in case and when the G5 got too far behind the curve they were ready (unlike the Altivec fanbois who swore they'd never buy an Intel Apple...). I'd even bet there's iOS builds for Intel Atom platforms in the back offices just in case the ARM designs powering their tablets and phablets run out of steam.
Most servers don't need anything near 6TB of storage, let alone 6TB of (D)RAM.
Many servers do need access to that sort of storage (and a hundred times more) and it would help if those servers run the same OS on similar hardware as other servers with less demand do. The alternative is for the sort of species differentiation that hobbles High Performance Computing (HPC) because there are few standards and a lot of hand-written system code flying in close formation, different on each machine.
I expect, if anything comes of this 3D flash RAM that there will be a flurry of Good Intentions by assorted BigCorps and LittleStartups which will evolve into two or three usable hardware and OS implementations that everyone coalesces around and then we move forward from there. The intervening period will be fun to observe, from a safe distance at least.
Oh, and are you claiming that 6TB should be enough RAM for anyone? There's a Slashdot meme for that...
Seems like we're going to find out soon, 6TB of addressable non-volatile ram sounds like a game changer
A server system really needs to be able to address hundreds or thousands of terabytes of storage, not just six. That's what I meant by the server system designers having to revamp the basic concepts of a computer with RAM separate from secondary storage (HDDs or SSDs on a separate bus) to one with a "flat" storage architecture. The OS will have to change too to take account of the blurring or total elimination of the difference between addressable RAM space and secondary storage, and getting that to work reliably and efficiently will take some time and involve some serious mistakes too, I expect. Sharing storage over a network will also be a challenge, as will backup processes.
If this tech makes it into the marketplace at reasonable prices it's not going to be hanging off SAS-12 cables or any other serial links at that rate, it will be more tightly integrated with the CPU bus to deliver on the R/W and access speed improvements. Even PCI is a possible bottleneck if this 3D flash can deliver what Intel are claiming for it. Comparing its performance to DRAM is a "tell" and shows what they're thinking; this may be the fabled "non-volatile RAM" solution that's been the Holy Grail researchers have been trying to develop pretty much ever since RAM was invented. (Yes, I know there are battery-backed-up RAM solutions that claim to be non-volatile but they're only non-volatile until the battery power runs out).
There's a whole raft of other things to consider before this tech changes the IT world -- how much does it cost, how many separate fabs can produce it so there's no single-point-of-failure that could constrain supply, how much redesign of existing chipsets is required to integrate it into current server/workstation/mobile phone designs, what's the failure rate in service, power dissipation and cooling requirements etc.
Saying that the demo suggests it can be implemented into existing platforms with little difficulty. Of course as Napoleon once said, "There are lies, damned lies and rigged demos." Time will tell.
I worked as an Amazon customer service rep last year. All calls are recorded and retained, as are emails, web chats and Kindle Mayday video chats. They use the weasel-words "may be recorded" since the recording system might break.
Hundreds actually, if you're only counting atmospheric tests although some were underwater and a handful in the stratosphere. The US fired off about 220 atmospheric tests, going to underground testing after the Sunbeam series of tests in Nevada in 1962. The Soviets did a lot fewer; the US has carried out more than half of all the nuclear tests by all nations.
Actually, no China isn't putting a crapton of resources into LFTR. It's actually putting about a hundred billion bucks into building a lot of PWRs with more to come in the next ten years or so if they continue the way they're going. Chinese researchers are looking at molten-salt reactor technologies but no significant money has been spent, same with fast-spectrum reactors like the Russian BN-series designs which at least exist in the real world. They're not building any molten-salt reactors, they have no plans to build such a reactor, there are no components for such a reactor being ordered or manufactured. There is some theoretical research and computer modelling being carried out, that's all. The only experimental reactor they're actually spending money on building is a pebble-bed design, the HTR-PM.
The Oak Ridge molten-salt reactor never used thorium, ever. It ran with U-233 and later with U-235 but never thorium. I blame the Powerpoint Rangers for conflating the purely theoretical LFTR with the Oak Ridge reactor (which was only one of many possible reactor concepts being tried out back in the 1960s).
No-one wants a beta reactor, they want something that will predictably generate electricity at a reasonable cost. That's why virtually every reactor being built today, including the four new-build AP-1000s at Vogtle and Summer as well as the dozens of reactors under construction in China and elsewhere are an evolutionary development of the PWR/BWR concept. The design effort has been concentrated in ever greater cost efficiencies and safety enhancements in larger and more efficient designs generating more electricity per reactor unit.
There are a couple of new power reactor designs not based on the well-tested PWR/BWR concept but even they are evolutionary; the new Russian BN-800 fast reactor which started up last year is an offshoot of the BN-350 and BN-600 reactors built in the 1970s and the Chinese modular pebble-bed HTR-PM now being built is using technology licenced from the Germans who had the ill-fated AVR and HTHR-300 pebble-bed reactors operating in the 1980s.
The capital costs of a reactor build are high because it's an expensive piece of construction, not simply because of delays etc. Every other large project including coal-fired and natural-gas generating plants also have to spend money up front preparing plans, covering the likely environmental impacts and dealing with protests.
Nearly all modern-build Gen-IIa and Gen-III reactors like the AP-1400, the EPR, ESBWR etc. are significantly larger than the original Gen-1 and Gen-II designs, each generating well over 1GW of electricity (the EPRs when they are complete will produce 1.6GW). That takes a lot of concrete and steel for containment, bigger turbine-generator sets, a larger reactor vessel etc. Putting that all together takes longer to complete even if everything goes right first time -- the Chinese are turning out their enhanced Gen-II APR-1000s in about 6 years from breaking ground to first grid connection but they've got a tested production line in place for groundworks, components and construction.
The good long-term news is that new-build Gen-IIIs will operate for more than 60 years; the Russians just produced a reactor vessel that they claim will last in service for a century and more. This improves the financial viability of a reactor project even though they cost a chunk of money up front.
Putting all these technologies together in one design is nearly trivial.
I'm sorry but that statement makes me die a little inside. That sort of reasoning is why I call the thorium boosters Powerpoint Rangers.
Yes I didn't watch any of the thirty or forty half-hour-long Powerpoint presentations on Youtube because none of them actually show hardware in operation, they show TED talks by graduate students and Twue Beweivers with glossy brochures, simplified diagrams and wishful thinking and a lot of "and then a miracle occurs" in the middle. I've seen some of the early attempts to convince folks that thorium was the Answer to non-existent problems before I decided life was too short and went TL;DR on the subject. To engage my interest it will take working hardware, a real molten-salt mostly-thorium reactor that runs for a year or more continuously (or at least 90% uptime) with significant thermal output (10 MW would be a good start, even 7 MW like the ORNL uranium-fuelled reactor produced) and a plan to decommission such a reactor in the future after 50 or 60 years of neutron irradiation and radiochemical contamination of components in direct contact with molten fuel (not something that causes a problem with existing solid-fuel reactors, even breeders).
the people that are making it happen
Until they have funding to build and operate a complete operational test reactor for at least a few years nobody is "making it happen", they're selling dreams. Sorry.
I hate to break it to you but China is not building any kind of a molten salt reactor (MSR) never mind one where thorium makes up part of the fuel stream. They've talked about it, yes but then again a lot of Powerpoint Rangers have done so over the past ten years and more. They're not funding, pouring concrete or bending metal on an MSR, the one true sign that anyone is taking the production of a prototype reactor seriously. They have discussed building a BN-series fast-spectrum reactor as well but no funding, no metal bent, no concrete poured on that project either which is a shame as it's a very interesting concept with the ability to burn waste actinides with minimal fuel reprocessing.
India's interest in thorium is as an adjunct fuel for their heavy-water reactors mixed in with a lot of 20%-medium-enriched uranium and plutonium. There are geopolitical reasons for this decision to work on using thorium as an adjunct fuel but at the moment they're building out several PWRs, mostly Russian VVER-series designs. Their fast-breeder is a conventional sodium-cooled design, not speciafically built to use thorium. It is outside NPT and IAEA safeguards so fuelling it could be a problem.
Experimental reactor designs actually being funded, built and operated in the real world are the Chinese HTR-PM helium-cooled pebble-bed reactor and the Russian BN-800 fast-spectrum reactor, both based on existing engineering predecessors (the ill-fated German HTHR-300, the Chinese HTR-10 and the ex-Soviet sodium-cooled BN-350 and BN-600).
Just so other folks understand, no-one has built and operated a thorium-fuelled molten salt breeder reactor so the design hasn't been tested as you claim. There was a molten-salt reactor fuelled with U-233 and later on U-235 at Oak Ridge but it never used thorium in any way. The Powerpoint Rangers pushing thorium claim it would be easy to add a breeding core to such a design and there would be no serious technical challenges involved despite the very high temperatures involved, the intense radiation environment required for the breeding process and its effects on structures carrying molten fuel. That last factor was not something plutonium breeder reactors have had to put up with and their success record is not good even using fixed arrays of solid fuel.
Thorium fission can scale.
Thorium (specifically Th-232) doesn't fission. It has to be bred up into U-233 by absorbing a neutron which can then fission by being hit by another neutron releasing energy. Fission of U-233 releases an average of about 2.2 neutrons to carry out further breeding and fission. This breeding-fission process is a bit knife-edge compared to regular PWRs, BWRs and other uranium-fuelled reactors where only one neutron is required to fission a U-235 nucleus and produce 2+ more.
What recent developments in "thorium fission" can you point us at? There's a lot of Powerpoint presentations and glossy brochures being waved around by folks looking to make a buck from research funding and subsidies but no-one is bending metal and pouring concrete right now on anything based on thorium as a primary source of nuclear energy. There ARE some experiments with thorium going on; PWR-style fuel pellets with thorium mixed in with uranium and plutonium are being exposed in a test reactor in Norway at the moment and there's an experimental Chinese pebble-bed reactor which can use some thorium in the fuel pebbles but that's about it.
As for modularity, it's not a new thing in regular uranium reactors -- see the units used in submarines, icebreakers and large aircraft carriers for the past fifty years and more as a worked example. The Russians are building a "power barge" carrying a ship reactor to produce about 40MW of electricity for coastal communities in Siberia and the Chinese are pouring concrete on a commercial modular power reactor (about 105MWe) but it's going to be a pebble-bed design fuelled entirely by uranium and plutonium to begin with. It might use a small amount of thorium in the future but that's a long way off and its commercial viability is still to be proven. Previous attempts to commercialise pebble-bed reactors capable of using some thorium such as the German THTR-300 were not a success.
I recently put Cinnamon Mint (17?) on an old server I got cheap, it worked like a champ. Then I added a graphics card (an nVidia PCI-bus FX5200, to fit in one of the PCI slots since the server mobo doesn't have any PCI-e x16 slots). Mint refused to boot up into a windowed environment, instead it dumped me to a command line prompt and told me something about restarting MDM, whatever that is.
I replaced the Mint installation on the server with Windows 7. It works like a champ and it doesn't complain about the video card. Any setting and configurations for resolution etc. are done from a windowed GUI with clicky boxes and drop-downs rather than unhelpful command-line options. nVidia don't support the ancient FX5200 under Win7 but the legacy 64-bit Vista drivers went on without a quibble and run perfectly.
I can't recommend Mint if you want to ever change anything from your default installation. Windows is much better in that case, IME.
Charlie Stross wrote a short story, "Dechlorinating the Moderator" a while back about a convention of hobbyist particle physics geeks using stuff like this to produce Higgs bosons in a hotel's banqueting suite.
Even smarter, install two fake charging ports next to each other. One has an "Out of Order" sign on it. The other one says "FREE CHARGING!".
Wire them up so the "FREE" charger discharges the battery of anyone who plugs into it while feeding the power to the "Out of Order" charger your own electric car is plugged into.
A lot of iron and steel foundries in the UK had a particular model of spectrometer built in the 1960s for carrying out analyses of metal samples. This spectrometer had an option of a hard-wired electric typewriter (not an IBM Selectric, a Remington or similar) that would print out the results of an analysis in a simple table. This was before Centronics printers were readily available. Back in the early 80s I worked with a consulting metallurgist to add a box to tap into the signals from the analyser to the typewriter and present them to a parallel-port interface so they could be read by a desktop computer. From memory the drive signals were at 24V with some weird pulse combinations for shifts, code pages etc. that we decoded by trial and error -- the company making the analysers wouldn't hand out the information and we couldn't mess with their internals since they were covered by long-term maintenance agreements (decades and more, these were very expensive bits of kit). We got the interface to work and the foundries were very happy to pay us to get the information out in digital for eventually supporting ISO9000 chain-of-custody certification which was necessary given that some of the places that used these analysers were making single large castings for paper-making machinery worth a million bucks a pop.
How in the name of all stupid plot devices does each and every space suit, vehicle, structure and other large chunk of habitat equipment not have its own, independent up-link to the multiple Earth-Mars radio relays we already have in orbit around that planet?
Worked example -- the Apollo moonsuits had a short-range VHF link back to the Lunar Lander. From there the radio comms was via an S-band microwave link back to Earth. There was no direct link from the suits to the Command Module in orbit, even when it was above the horizon for the astronauts on the Lunar surface.
The digitising system used in the Surface Pro 1 and 2 was a Wacom-based design, it's now an nTrig in the Surface Pro 3. Both of them have pens that are powered by near-field from the digitising surface so they don't need separate power or charging at all. Since Apple are relying on their passive capacitive digitiser to work with the Pencil they can't power it that way.
Wacom provide a range of specialist stylii for artists such as an airbrush model, I'm not sure if nTrig do.