... but in the rest of the world, self driving cars will improve by leaps and bounds...
Depending on where in the world you are, this might be a necessity. Observing the driving habits I've seen in many countries of the far east and parts of southern europe, self driving cars *better* improve by leaps and bounds just to survive!
Back when vista was released, intel was co-releasing core2duo and core2quad desktop chips Back when win7 was released, intel was co-releasing the core-i7 chips
From intel's point of view, microsoft is 'jumping-the-gun'. Haswell isn't going to be released until summer and microsoft is gearing up for a christmas refresh, so they can't draft any Win8 refresh pull when haswell becomes available.
Nobody "draws" chips by "hand" anymore. It's all being done by a computer (there are so many design rules these days humans can't do this anymore in a realistic time frame). Reticles (the photomasks) are all fractured by computer these days because rectangles aren't really rectangles anymore at these small feature sizes (we are now past the diffraction limit so masks must be "phase-shift" masks not binary masks back in the old-days).
I don't have any specific knowledge about the A6, but what is euphamistically called hand-drawn these days is often still very automated relative to the bad-old-days when people were drawing rectangles on layers to make transitors. That was the real-hand-drawn days, but even way back then you didn't actually draw them by hand, you used a computer program to enter the coordinates for the rectangles.
Quick background: now days when typical chips go to physical design, they usually go through a system called place-and-route where pre-optimized "cells" (which have 2-4 inputs and 1-3 outputs and implement stuff like and-or-invert, or register flop) are placed down by the computer (typically using advanced heuristic algorithms) and the various inputs and outputs are connected together with many layers of wires which logically match the schematic or netlist (which is the intention of the logical design). Of course this is when physics starts to impose on the "logical" design, so often things need special fixups to make things work. Unfortunatly, the fixups and the worst case wirelengths between cells conspire to limit the performance and power of the design, but just like compiled software, it's usually good enough for most purposes. Highly leveraged regularly structured components of normal designs might have libraries, specialized compilers or even have hand intervention (e.g, rams, fifos, or register files), but not the bulk of the logic.
As far as I can tell from looking at the pictures the most likely possibility is that just that instead of letting the computer place the design completely out of small cells, some larger blocks (say like ALUs for the ARM SIMD path) were created by a designer and layout engineer who probably used a lower-level tool to put down the same small cells relative to other small cells where they think is a good place to put them and tweak the relative positioning to try to minimize the maximum wire lengths between critical parts of the block. The most common flow for doing this is mostly automated, but tweakable with human intervention (this what passed for "by-hand" these days). In addition to being designed to optimize critical paths, these larger blocks are generally desgined so that they "fit" well with other parts of the design (e.g., port order, wire pitch match, etc) to minimize wire congestion (so they can be connected with mostly straight wires, instead of those that bend). Basically looking at the patterns of whitespace in the presumed CPU, you can see the structure of these larger blocks instead of big rectangles (called partitions) which have rows of cells you get when you let a computer do place-and-route with small cells.
Just like optimizing a program, there are many levels of pain you can go through and what I described above is probably the limit these days. Say if you wanted less pain, another more automated way to get most of the same benefits is to just develop a flow that hints where to put parts of the design inside the normal rectangular placement region, and let a placement engine use those hints. The designer can just tweak the hints to get better results. Of course with this method, the routing may still have "kinks" in this case because routing is not wire-pitch-matched, but you can often get 80-90% the way there. The advantage of this lesser technique is that you don't need to spend a bunch of time developing big blocks and if there is a small mistake (of course nobody ever makes mistakes), it's much, much easier to fix the mistake w/o perturbing the whole design.
In most states, you currently have to get an emissions sticker to license your car To get this emmissions sticker, you have to actually go to a facility that measures your cars emmission. Today, with the test they record your VIN and odometer and send it to the state so you can get your license renewed.
To replace the fuel tax, the states can still require electric cars go to a place where your odometer and VIN are recorded and the state can just tack on a milage*weight surtax for electric cars your license fee which will be required to be paid before you get your license renewed. As a bonus, this will probably be less regressive than the current fuel tax (which effectively taxes older cars owned by poorer folks more than newer cars owned by wealthier folks) and better approximate a road-user-tax than a consumption tax.
Sure there will be cheating, but it will probably be good-enough (way better than the GPS proposal they had in oregon). My insurance company only requires me to phone-in my odometer to get the low-milage discount, going to a "smog" station that is regulated by the state is likely going to pass any statistical compliance tax bar.
The hard part will be to get the electric car infrastructure to allow fast charging. Once that is figured out, I doubt there will be much of a resistance.
Sigh-of-relief... My first reading of this was, maybe Nasa was reusing Aegis for some sort of swords/plowshares purpose to somehow to control the rover. I guess that isn't the case, and it's just a fancy name for Rover-software 2.0...
The reason that rang a bell with me was a bit of trivia I remembered. One of the first deployed Aegis system was the USS Yorktown. Apparently, the Aegis software was deployed on WindowNT which of course had lots of stability problems. Also the system software itself wasn't that great. For example, when someone accidentally entered some bad data, it caused a divide-by-zero error which caused the software to crash and the ship had to be towed back into port. I think Curiosity would have a hard time being "towed" back to port...
Fortunatly, it isn't the same software, but Nasa has a propensity to tempt fate with unlucky symbols (Apollo 13, Challenger). Let's hope this isn't one of those times...
On the other hand Temperature (e.g., in Kelvin) is only marginally useful in describing the distribution of a phenomena that isn't in thermal equilibrium (say non-blackbody radiation)...
For example, people used to grade lightbulbs by their color Temperature, but that didn't say much about the quality of illumination from said lightbulb. Now they use CRI (color rendering index) for lightbulbs which give some information about the actual distribution instead of the really poor assumption that the illumination was comparable to black-body radiation distribution.
It's not clear (to me) that a galactic halo would necessarily be in thermal equilibrium, except only approximatly over a long time horizon. The real interesting observable phenomena is likely a result of this not being true (e.g. http://arxiv.org/abs/1106.4816 )
Taxpayers pay for everything that government does...
Bottom line, if you don't get health insurance yourself, we are collectively robbing taxpayers are paying for it (one way or another). Who do you think picks up the insurance when someone robs money from a bank? We do. Witness TARP.
They face a declining budget each year and have to try find ways to make it streach and what projects should cover it.
Like any far-flung political governemnt organization, not all parts of NASA are created equal. If you look at the "top" line budgeting for NASA, you'll see the following...
Pretty steady funding... Science: FY11: $4.9B, FY12: $5.0B, FY13: $4.9B (stuff like the James Web Telescope, LandSat, MSL experiments) Exploration: FY11: $3.8B, FY12: $3.7B, FY13: $3.9B (rover development, Orion, space commercialization grants) Cross Agency: FY11: $2.9B, FY12: $2.9B, FY13: $2.8B (overhead, people, buildings, pork, etc) Aeronautics: FY11: $530M, FY12: $570M, FY13: $550M (all that nextgen airtraffic control stuff)
Increasing funding... Space Tech: FY11: $450M, FY12: $570M, FY13: $700M (all the researchy stuff you are talking about)
Of course for longer term aggregate budgetary trends, you could also look at a wiki and you can see how your statement about "declining budgets" isn't really true at all http://en.wikipedia.org/wiki/Budget_of_NASA
It just seems like Nasa is getting less money since its percentage of the total federal budget has been shrinking (from 1% in the 90's to less than 0.5% today). However, in real dollars it's been about the same for a long while. Unfortunatly, the rest of the government has been getting bloated. Maybe that's the space-nerd in you feeling like Nasa is getting gyped relative to other federal spending.
Still, getting "gyped" no excuse for not getting something from the $450-700M/year we are spending on researching new Space Tech for the last 30 years. The strategy to attempt to garner sympathy for failure by pointing out others got more help is excuse of a kindergardener... Sure there are political winds to navigate, but that's the game you have to play when the tab is being picked up by a bunch of finiky taxpayers.
...the company wants to license its patents instead of heading to court.
Obviously. Who would voluntarily want to put the fate of their enterprise in the hands of 12 people who couldn't think of a valid excuse of how to get out of (civil) jury duty?
Of course you might think that mediation and arbitration would solve this problem, but if one of the sides has a little to lose, and much to gain, they often will gamble their chances with the jury trial. As a classic example, Apple vs Samsung: a billion dollars to each side is merely pocket change, not much to lose for either one. On the flip side, a couple of bankrupt memory chip companies might need to some stability and predicability just to survive. A patent troll, however, doesn't want their patents accidentally invalidated, so gambling isn't the prefered way to do business...
Just more reasons why we should advocate for vat grown meat.
Healthier, higher and more consistent quality, just as tasty and dense, and (once the tech is mature enough) cheaper than gutgrown.
It's easy to speculate that technology that doesn't exist yet will be motherhood and apple pie, but in the 60's the futuristic tech fad was better living through chemistry and that somehow we could avoid the coming Malthusian Catastrophe by developing complete artificial food. Of course looking back, it probably seemed foolish that we had the tech to make cost-effective food substitutes out of vitamin formula, but fortunatly, agri-technology allowed us to increase food production to avoid the predicted Catastrophe and we didn't just toss in the towel on food and put our fate in with the vitamin gurus...
To assume that sometime in the near future we can just grow meat and it will somehow be "better" is the same '60's utopia thinking. We don't know how to do this and it's the height of hubris to think that even when we *think* we have it figured out it that sometime in the future, people will look back (just like the vitamin thing) and realize, we really didn't have everything figured out and for now it's still net-better to eat actual food, rather than the synthetic stuff.
Vatgrown meat might be viable from a technical perspective someday, but I'm guessing most folks would initially react to it the same way they did with "pink-slime". Of course the perceived problem probably won't be the same as with "pink-slime", but I'm assuming that we will have to feed the cells in this vat grown meat somehow and I'm guessing for economic viablity reasons, that it won't be grade A materials, or grade A sourced and even if people are screaming at the top of their lungs how much healthier and higher and more consistent quality, just as tasty and dense, and cheaper the stuff might be, there will be tremendous hurdles to overcome.
For example, look at the furor over carmine (red) dye in "strawberry" frappachinos. Of course the historical alternatives to red carmine dyes is Allura Red (aka Red #40) made from petroleum, or the original Amaranth dye (which was used to make stuff like candied cherries and which was linked to cancer). This overlooks the fact that carmine dyes were infact the historical choice (pre-dating the other industrial red collorants).
As you might suspect the company in question immediatly switched to a new Lycopene (tomato) based red collorant (untested, but generally regarded as safe) to appease the masses, but I'm sure there will be many such hurdles to overcome before Vatgrown even makes it out of the test-market phase... People get bent our of shape right now, when stores douse their meat with carbon monoxide to give it that extra-red color. No telling how are they gonna react to vatgrown meat that came from a soup of chemicals...
1. Remember to smile when breaking the law... OR 2. Create startup company which intends to develop software for facial recognition of smiling people... 3. ??? 4. PROFIT!!!
...To be "faster than a Cray-2", you really would need a Beowulf cluster of iPad processors.
Cray2 -> 2GFLOPS total iPad2 -> dual-core 1GHz A9+ GPU...
The Neon coprocessor on each ARM A9 is SIMD 2x32-bit Flops/clock. On this basis, it's not to hard to believe a single iPad2 has more raw Gflops than a Cray2 (w/o needing a beowulf cluster).
On the other hand, the Cray2 was available way back in 1985, (the year that Steve Jobs got kicked out of Apple)...
It really should be seen as a sad statement of the state of American spaceflight where the first private commercial spaceflight crews were launched with equipment designed by a Communist country.
Or, you might say it was a sad state of a communist country that they had to be the first to privatize their space launch equipment...
From each according to his ability, for each according to his need. In this case, some capitalists had the ability to pay, and the (privitized Energia) needed money...
...even though unions create a low-turnover workforce interested in productivity and self-improvement because workers know that, in return, they're going to enjoy better treatment and security of employment.
Oh, if that were the case. Right now, they know they're going to enjoy better treatment and security of employment, so in many unions, it's damn the productivity and self-improvement, it's all about senority...
Like any other monopoly, at leat in the US, the unions stopped caring after they formed a monopoly (e.g. through mega-mergers like the AFL-CIO in the industrial age). If the AFL-CIO was a bank, I'm sure people would be screaming bloody murder anti-trust. Of course they got a pass. Thus started a era of decline in US manufacturing...
Of course things are starting to changing in the US, some of the biggest unions under the AFL-CIO umbrella, the SEIU, the Teamsters, and the UFCW are starting to disentangle from the AFL-CIO (officially breaking ties forming the change-to-win federation), primarily because as the old-guard unions seem to stop caring about union issues and more about (Democratic Party) politics. Of course Obama and the Democratic establishment (which relies on their backing) are trying their best to rein them back into a monolithic block. Hopefully, this whole episode will bring about better union leadership in the US, but probably only if the Democratic Party can stay out of it...
Although I have no personal knowledge about this situation, if I were to speculate on rationale for the position that Intel is taking, it's power/electrical related.
I'm gonna go out on a limb and say that they way it is probably designed, you can't just turn on both cores at full clock rate and turn on the GPU w/o creating electrical/thermal instability or melting the chip. They probably have to do lots of dynamic voltage/frequency tradeoffs to get their chip to run which considers the whole system power/heat, thermals (temperature), and electrical current battery supply. Also the parameters to control the chip probably vary greatly from chip-to-chip with manufacturing process, so the per-chip information likely is stored somewhere undocumented. Of course Intel likely has a fancy driver that does all this stuff, but it probably needs constant attention (callbacks) from the OS and information from the OS about workloads, but they...
1. Are only tuning this for Win8 (don't have time to do anything else) 2. Don't see the required callback and information support in the current linux kernel 3. Don't want to release the API information to linux kernel writer so that competitors learn their secret sauce and piggyback on their hard work
They are probably reasonably sure that running a "default" x86 linux on their chip that doesn't adapt the voltage and frequency with workload will likely be a recipe for either electrical/thermal instability or perhaps even a meltdown. Using the code phrase that "linux" work seems to be the biggest clue here...
Okay, can you tell me what the following statement means to you?
I said that you could consider integer units that emulated fp hardware to be doing flops.
1. I don't see any supercomputers emulating fp hardware on integer units... 2. Even if they did (which they don't), it would be so slow that it would be a rounding error in their flop rating.
As I (and others have posted), although there are some interesting integer problems, existing supercomputers issue those instructions to processing units that are essentially the same speed as FP units, so there's not much difference between a FLOP and the IOP number of these machines, and the scientific code that require floating point are much more interesting to the current buyers of those machines. That's why they don't rate them in a different dimension.
It would be like suggesting people rate cars by Refueling Range and 60-90 time vs MPG and 0-60 time. Sure for some applications you would like to know how far you can go on a tank of gas and how fast you can pass people on the highway, but most people buying cars today simply want to know if they can reasonably accelerate to merge on a highway and how much gas will cost them for commuting. For people that care more about the other stuff, they can look deeper into the numbers for that specific car, but they've gotta know that since most people aren't using their criteria, car manufactures aren't optimizing for them and likely won't be for the forseeable future.
Most modern supercomputers get their "flop" count from SSE3/4 and/or GPUs which are not integer, but Floating point processing machines(at least 32-bit single precision fp, but also double precision albeit at a slower rate). These machines most certainly do NOT simulate floating point with their integer units (nor cheat by calling an integer op as an approximate fp op), and they have massive amounts of dedicated hardware SIMD FP processing units to do their heavy lifting.
Of course there are many real world problems that could use parallel integer math and CPUs and GPUs are also capable of lots of SIMD integer ops as well, but that's not how supercomputers are rated these days, they are rated by the number of IEEE FP operations (mostly FMA or fused multipy-add counting as 2-ops) with at least 32-bits of precision.
The integer OPs currently don't count in the current ratings and I don't see that changing any time soon. Important scientific operations like matrix inversion, finite-element analysis, FFTs, and linear programming don't work the same with integer ops, so it is unfair to compare supercomputers by their integer ops.
Since you can call assembly directly in C, then you identify things C does poorly, and write it in assembly, then call that. It makes for remarkably efficient code, we demonstrated this in class with search and sort algorithms.
ACK!!! If there is a course in assembler and it is merely being used to bash how a specific C compiler poorly optimizes a specific piece of code, and motivating future programmers rewrite algorithms in assembler, that skill is probably a waste and is gonna tempt these students to be yet another generation of people that desire to distrust and abuse the portable/maintainable code infrastructure we built up over the years and sacrifice it in the name of percieved efficiency.
Compilers are getting better and better and often have to turn off some very important optimizations when inserting assembly language (e.g, some alias detection algorithms, strength reduction algorithms, and some loop invariance detection). Also is it is sometimes totally unclear how long certain instructions will take from an assembly point of view because of microarchitectural structural hazards (which the compiler may be optimized for). Also, often you can just recode the source and get nearly identical performance (not code, performance) with existing optimizing compilers by basically doing the desired code transformations yourself in the source code (and teach nearly the same skills).
On the other hand, such a class could be doing worthwhile stuff like teaching things structured around ABI issues and maybe analyzing a compiler peephole optmization stage from the context of an assembler; or perhaps talked about linker patching, sections and other similar issues; or maybe even talked about strategies for integrating dynamically generated code or SIMD instruction into a scalar instruction flow (something which compilers are still struggling with and are important for stuff like JIT and DSP code)...
Although I don't disagree in general that politicians don't really care too much what people want, that's a far too cynical attitude to think they like people who do what they're told. A much simpler explanation for their behaviour is that they are bought/paid-for since the easiest way to get re-elected is to spend lots of money (advertising for elections and bringing back pork for the local folks, etc). If they thought it would be easier to get re-elected by just figuring out what most of the people wanted and doing that, they would probably take that route.
Unfortunatly, figuring out what most people who vote want is hard and the only folks that tell them what to do are people that don't care what most of the people want and push their own agenda (e.g., lobbyists and national special interest groups) and carrot/stick them with lots of money. Also, even if a politician did what "most" of the people wanted, many of those people just don't vote, or if they do, they don't generally vote their own interests, they generally vote how other people tell them to vote (which may or may not corrolate with their own interests). There are only a very few "undecideds" that actually vote and can be actually swayed in significant numbers by figuring out what they want and doing that. Sadly, it's far easier to "energize the base" or convince people that will vote for you because other people (special interest groups that have spent lots of money) already told them how to vote and actually get them to vote.
As for the primaries, it's really a state-to-state issue. In Connecticut, they used to have a closed primary law, but when the Republicans wanted to allow independents to vote in their primary election, they challenged the law and SCOTUS overturned the Connecticut law which allowed the republicans to have a semi-open primary. In California, we now have a "top-two" system. Instead of having each party pick a candidate in a closed or semi-open primary, the top two vote getters in the primary, regardless of party affiliation, run-off in the general election (except for the presidential election which is semi-closed). It totally depends on which state you are in what the situation is. In AL, AK, GA, ID, IL, MA, MI, MN, MS, MO, NH, NC, ND, SC, TN, TX, VT, VA, WI, there are only open primaries, so your milage may vary... Both parties try to game the laws to their advantage...
.. and then there will be this shift to "fiat" BitCoin Certificates that look very much like BitCoin Certificates, but can be created out of thin air by certain entities that wish to control the economy where only some decreasing fraction of them are backed by real BitCoin...
So someone explain to me how these BitCoin Certificates will somehow be different from current money?
Also, peer review is not designed to catch fraud, only to catch errors in process or analysis. In the Ivy League they aren't immune to issues either...
People are just people. Just because they are working at a fancy-schmancy university, or some other folks put another gold star next to their paper (peer review is often blinded by reputation), isn't the standard we should be touting. The only arbitor that matters is nature, as one famous physicist (RPF) put it, "Nature cannot be fooled"...
Slashdot Mirror
... but in the rest of the world, self driving cars will improve by leaps and bounds...
Depending on where in the world you are, this might be a necessity. Observing the driving habits I've seen in many countries of the far east and parts of southern europe, self driving cars *better* improve by leaps and bounds just to survive!
Back when vista was released, intel was co-releasing core2duo and core2quad desktop chips
Back when win7 was released, intel was co-releasing the core-i7 chips
From intel's point of view, microsoft is 'jumping-the-gun'. Haswell isn't going to be released until summer and microsoft is gearing up for a christmas refresh, so they can't draft any Win8 refresh pull when haswell becomes available.
Nobody "draws" chips by "hand" anymore. It's all being done by a computer (there are so many design rules these days humans can't do this anymore in a realistic time frame). Reticles (the photomasks) are all fractured by computer these days because rectangles aren't really rectangles anymore at these small feature sizes (we are now past the diffraction limit so masks must be "phase-shift" masks not binary masks back in the old-days).
I don't have any specific knowledge about the A6, but what is euphamistically called hand-drawn these days is often still very automated relative to the bad-old-days when people were drawing rectangles on layers to make transitors. That was the real-hand-drawn days, but even way back then you didn't actually draw them by hand, you used a computer program to enter the coordinates for the rectangles.
Quick background: now days when typical chips go to physical design, they usually go through a system called place-and-route where pre-optimized "cells" (which have 2-4 inputs and 1-3 outputs and implement stuff like and-or-invert, or register flop) are placed down by the computer (typically using advanced heuristic algorithms) and the various inputs and outputs are connected together with many layers of wires which logically match the schematic or netlist (which is the intention of the logical design). Of course this is when physics starts to impose on the "logical" design, so often things need special fixups to make things work. Unfortunatly, the fixups and the worst case wirelengths between cells conspire to limit the performance and power of the design, but just like compiled software, it's usually good enough for most purposes. Highly leveraged regularly structured components of normal designs might have libraries, specialized compilers or even have hand intervention (e.g, rams, fifos, or register files), but not the bulk of the logic.
As far as I can tell from looking at the pictures the most likely possibility is that just that instead of letting the computer place the design completely out of small cells, some larger blocks (say like ALUs for the ARM SIMD path) were created by a designer and layout engineer who probably used a lower-level tool to put down the same small cells relative to other small cells where they think is a good place to put them and tweak the relative positioning to try to minimize the maximum wire lengths between critical parts of the block. The most common flow for doing this is mostly automated, but tweakable with human intervention (this what passed for "by-hand" these days). In addition to being designed to optimize critical paths, these larger blocks are generally desgined so that they "fit" well with other parts of the design (e.g., port order, wire pitch match, etc) to minimize wire congestion (so they can be connected with mostly straight wires, instead of those that bend). Basically looking at the patterns of whitespace in the presumed CPU, you can see the structure of these larger blocks instead of big rectangles (called partitions) which have rows of cells you get when you let a computer do place-and-route with small cells.
Just like optimizing a program, there are many levels of pain you can go through and what I described above is probably the limit these days. Say if you wanted less pain, another more automated way to get most of the same benefits is to just develop a flow that hints where to put parts of the design inside the normal rectangular placement region, and let a placement engine use those hints. The designer can just tweak the hints to get better results. Of course with this method, the routing may still have "kinks" in this case because routing is not wire-pitch-matched, but you can often get 80-90% the way there. The advantage of this lesser technique is that you don't need to spend a bunch of time developing big blocks and if there is a small mistake (of course nobody ever makes mistakes), it's much, much easier to fix the mistake w/o perturbing the whole design.
FWIW, it is highly unlikely that th
This will be the easy part.
In most states, you currently have to get an emissions sticker to license your car To get this emmissions sticker, you have to actually go to a facility that measures your cars emmission. Today, with the test they record your VIN and odometer and send it to the state so you can get your license renewed.
To replace the fuel tax, the states can still require electric cars go to a place where your odometer and VIN are recorded and the state can just tack on a milage*weight surtax for electric cars your license fee which will be required to be paid before you get your license renewed. As a bonus, this will probably be less regressive than the current fuel tax (which effectively taxes older cars owned by poorer folks more than newer cars owned by wealthier folks) and better approximate a road-user-tax than a consumption tax.
Sure there will be cheating, but it will probably be good-enough (way better than the GPS proposal they had in oregon). My insurance company only requires me to phone-in my odometer to get the low-milage discount, going to a "smog" station that is regulated by the state is likely going to pass any statistical compliance tax bar.
The hard part will be to get the electric car infrastructure to allow fast charging. Once that is figured out, I doubt there will be much of a resistance.
Sigh-of-relief... My first reading of this was, maybe Nasa was reusing Aegis for some sort of swords/plowshares purpose to somehow to control the rover. I guess that isn't the case, and it's just a fancy name for Rover-software 2.0...
The reason that rang a bell with me was a bit of trivia I remembered. One of the first deployed Aegis system was the USS Yorktown. Apparently, the Aegis software was deployed on WindowNT which of course had lots of stability problems. Also the system software itself wasn't that great. For example, when someone accidentally entered some bad data, it caused a divide-by-zero error which caused the software to crash and the ship had to be towed back into port. I think Curiosity would have a hard time being "towed" back to port...
Fortunatly, it isn't the same software, but Nasa has a propensity to tempt fate with unlucky symbols (Apollo 13, Challenger). Let's hope this isn't one of those times...
On the other hand Temperature (e.g., in Kelvin) is only marginally useful in describing the distribution of a phenomena that isn't in thermal equilibrium (say non-blackbody radiation)...
For example, people used to grade lightbulbs by their color Temperature, but that didn't say much about the quality of illumination from said lightbulb. Now they use CRI (color rendering index) for lightbulbs which give some information about the actual distribution instead of the really poor assumption that the illumination was comparable to black-body radiation distribution.
It's not clear (to me) that a galactic halo would necessarily be in thermal equilibrium, except only approximatly over a long time horizon. The real interesting observable phenomena is likely a result of this not being true (e.g. http://arxiv.org/abs/1106.4816 )
And you don't think taxpayers pay for prisons?
Taxpayers pay for everything that government does...
Bottom line, if you don't get health insurance yourself, we are collectively robbing taxpayers are paying for it (one way or another). Who do you think picks up the insurance when someone robs money from a bank? We do. Witness TARP.
They face a declining budget each year and have to try find ways to make it streach and what projects should cover it.
Like any far-flung political governemnt organization, not all parts of NASA are created equal.
If you look at the "top" line budgeting for NASA, you'll see the following...
Decreasing funding...
Operations: FY11: $5.1B, FY12: $4.2B, FY13: $4.0B (e.g., shuttle, ISS)
Pretty steady funding...
Science: FY11: $4.9B, FY12: $5.0B, FY13: $4.9B (stuff like the James Web Telescope, LandSat, MSL experiments)
Exploration: FY11: $3.8B, FY12: $3.7B, FY13: $3.9B (rover development, Orion, space commercialization grants)
Cross Agency: FY11: $2.9B, FY12: $2.9B, FY13: $2.8B (overhead, people, buildings, pork, etc)
Aeronautics: FY11: $530M, FY12: $570M, FY13: $550M (all that nextgen airtraffic control stuff)
Increasing funding...
Space Tech: FY11: $450M, FY12: $570M, FY13: $700M (all the researchy stuff you are talking about)
Of course for longer term aggregate budgetary trends, you could also look at a wiki and you can see how your statement about "declining budgets" isn't really true at all http://en.wikipedia.org/wiki/Budget_of_NASA
It just seems like Nasa is getting less money since its percentage of the total federal budget has been shrinking (from 1% in the 90's to less than 0.5% today). However, in real dollars it's been about the same for a long while. Unfortunatly, the rest of the government has been getting bloated. Maybe that's the space-nerd in you feeling like Nasa is getting gyped relative to other federal spending.
Still, getting "gyped" no excuse for not getting something from the $450-700M/year we are spending on researching new Space Tech for the last 30 years. The strategy to attempt to garner sympathy for failure by pointing out others got more help is excuse of a kindergardener... Sure there are political winds to navigate, but that's the game you have to play when the tab is being picked up by a bunch of finiky taxpayers.
s/bank/taxpayer/g...
Fixed that for you ;^)
...the company wants to license its patents instead of heading to court.
Obviously. Who would voluntarily want to put the fate of their enterprise in the hands of 12 people who couldn't think of a valid excuse of how to get out of (civil) jury duty?
Of course you might think that mediation and arbitration would solve this problem, but if one of the sides has a little to lose, and much to gain, they often will gamble their chances with the jury trial. As a classic example, Apple vs Samsung: a billion dollars to each side is merely pocket change, not much to lose for either one. On the flip side, a couple of bankrupt memory chip companies might need to some stability and predicability just to survive. A patent troll, however, doesn't want their patents accidentally invalidated, so gambling isn't the prefered way to do business...
Just more reasons why we should advocate for vat grown meat.
Healthier, higher and more consistent quality, just as tasty and dense, and (once the tech is mature enough) cheaper than gutgrown.
It's easy to speculate that technology that doesn't exist yet will be motherhood and apple pie, but in the 60's the futuristic tech fad was better living through chemistry and that somehow we could avoid the coming Malthusian Catastrophe by developing complete artificial food. Of course looking back, it probably seemed foolish that we had the tech to make cost-effective food substitutes out of vitamin formula, but fortunatly, agri-technology allowed us to increase food production to avoid the predicted Catastrophe and we didn't just toss in the towel on food and put our fate in with the vitamin gurus...
To assume that sometime in the near future we can just grow meat and it will somehow be "better" is the same '60's utopia thinking. We don't know how to do this and it's the height of hubris to think that even when we *think* we have it figured out it that sometime in the future, people will look back (just like the vitamin thing) and realize, we really didn't have everything figured out and for now it's still net-better to eat actual food, rather than the synthetic stuff.
Vatgrown meat might be viable from a technical perspective someday, but I'm guessing most folks would initially react to it the same way they did with "pink-slime". Of course the perceived problem probably won't be the same as with "pink-slime", but I'm assuming that we will have to feed the cells in this vat grown meat somehow and I'm guessing for economic viablity reasons, that it won't be grade A materials, or grade A sourced and even if people are screaming at the top of their lungs how much healthier and higher and more consistent quality, just as tasty and dense, and cheaper the stuff might be, there will be tremendous hurdles to overcome.
For example, look at the furor over carmine (red) dye in "strawberry" frappachinos. Of course the historical alternatives to red carmine dyes is Allura Red (aka Red #40) made from petroleum, or the original Amaranth dye (which was used to make stuff like candied cherries and which was linked to cancer). This overlooks the fact that carmine dyes were infact the historical choice (pre-dating the other industrial red collorants).
As you might suspect the company in question immediatly switched to a new Lycopene (tomato) based red collorant (untested, but generally regarded as safe) to appease the masses, but I'm sure there will be many such hurdles to overcome before Vatgrown even makes it out of the test-market phase... People get bent our of shape right now, when stores douse their meat with carbon monoxide to give it that extra-red color. No telling how are they gonna react to vatgrown meat that came from a soup of chemicals...
1. Remember to smile when breaking the law... OR
2. Create startup company which intends to develop software for facial recognition of smiling people...
3. ???
4. PROFIT!!!
Somehow, when I read this headline about romania, the first thing that came to mind was the q-bomb...
...To be "faster than a Cray-2", you really would need a Beowulf cluster of iPad processors.
Cray2 -> 2GFLOPS total
iPad2 -> dual-core 1GHz A9+ GPU...
The Neon coprocessor on each ARM A9 is SIMD 2x32-bit Flops/clock. On this basis, it's not to hard to believe a single iPad2 has more raw Gflops than a Cray2 (w/o needing a beowulf cluster).
On the other hand, the Cray2 was available way back in 1985, (the year that Steve Jobs got kicked out of Apple)...
It really should be seen as a sad statement of the state of American spaceflight where the first private commercial spaceflight crews were launched with equipment designed by a Communist country.
Or, you might say it was a sad state of a communist country that they had to be the first to privatize their space launch equipment...
From each according to his ability, for each according to his need. In this case, some capitalists had the ability to pay, and the (privitized Energia) needed money...
...even though unions create a low-turnover workforce interested in productivity and self-improvement because workers know that, in return, they're going to enjoy better treatment and security of employment.
Oh, if that were the case. Right now, they know they're going to enjoy better treatment and security of employment, so in many unions, it's damn the productivity and self-improvement, it's all about senority...
Like any other monopoly, at leat in the US, the unions stopped caring after they formed a monopoly (e.g. through mega-mergers like the AFL-CIO in the industrial age). If the AFL-CIO was a bank, I'm sure people would be screaming bloody murder anti-trust. Of course they got a pass. Thus started a era of decline in US manufacturing...
Of course things are starting to changing in the US, some of the biggest unions under the AFL-CIO umbrella, the SEIU, the Teamsters, and the UFCW are starting to disentangle from the AFL-CIO (officially breaking ties forming the change-to-win federation), primarily because as the old-guard unions seem to stop caring about union issues and more about (Democratic Party) politics. Of course Obama and the Democratic establishment (which relies on their backing) are trying their best to rein them back into a monolithic block. Hopefully, this whole episode will bring about better union leadership in the US, but probably only if the Democratic Party can stay out of it...
Although I have no personal knowledge about this situation, if I were to speculate on rationale for the position that Intel is taking, it's power/electrical related.
I'm gonna go out on a limb and say that they way it is probably designed, you can't just turn on both cores at full clock rate and turn on the GPU w/o creating electrical/thermal instability or melting the chip. They probably have to do lots of dynamic voltage/frequency tradeoffs to get their chip to run which considers the whole system power/heat, thermals (temperature), and electrical current battery supply. Also the parameters to control the chip probably vary greatly from chip-to-chip with manufacturing process, so the per-chip information likely is stored somewhere undocumented. Of course Intel likely has a fancy driver that does all this stuff, but it probably needs constant attention (callbacks) from the OS and information from the OS about workloads, but they...
1. Are only tuning this for Win8 (don't have time to do anything else)
2. Don't see the required callback and information support in the current linux kernel
3. Don't want to release the API information to linux kernel writer so that competitors learn their secret sauce and piggyback on their hard work
They are probably reasonably sure that running a "default" x86 linux on their chip that doesn't adapt the voltage and frequency with workload will likely be a recipe for either electrical/thermal instability or perhaps even a meltdown. Using the code phrase that "linux" work seems to be the biggest clue here...
Okay, can you tell me what the following statement means to you?
I said that you could consider integer units that emulated fp hardware to be doing flops.
1. I don't see any supercomputers emulating fp hardware on integer units...
2. Even if they did (which they don't), it would be so slow that it would be a rounding error in their flop rating.
As I (and others have posted), although there are some interesting integer problems, existing supercomputers issue those instructions to processing units that are essentially the same speed as FP units, so there's not much difference between a FLOP and the IOP number of these machines, and the scientific code that require floating point are much more interesting to the current buyers of those machines. That's why they don't rate them in a different dimension.
It would be like suggesting people rate cars by Refueling Range and 60-90 time vs MPG and 0-60 time. Sure for some applications you would like to know how far you can go on a tank of gas and how fast you can pass people on the highway, but most people buying cars today simply want to know if they can reasonably accelerate to merge on a highway and how much gas will cost them for commuting. For people that care more about the other stuff, they can look deeper into the numbers for that specific car, but they've gotta know that since most people aren't using their criteria, car manufactures aren't optimizing for them and likely won't be for the forseeable future.
I'm pretty sure you are mistaken on this point.
Most modern supercomputers get their "flop" count from SSE3/4 and/or GPUs which are not integer, but Floating point processing machines(at least 32-bit single precision fp, but also double precision albeit at a slower rate). These machines most certainly do NOT simulate floating point with their integer units (nor cheat by calling an integer op as an approximate fp op), and they have massive amounts of dedicated hardware SIMD FP processing units to do their heavy lifting.
Of course there are many real world problems that could use parallel integer math and CPUs and GPUs are also capable of lots of SIMD integer ops as well, but that's not how supercomputers are rated these days, they are rated by the number of IEEE FP operations (mostly FMA or fused multipy-add counting as 2-ops) with at least 32-bits of precision.
The integer OPs currently don't count in the current ratings and I don't see that changing any time soon. Important scientific operations like matrix inversion, finite-element analysis, FFTs, and linear programming don't work the same with integer ops, so it is unfair to compare supercomputers by their integer ops.
Since you can call assembly directly in C, then you identify things C does poorly, and write it in assembly, then call that. It makes for remarkably efficient code, we demonstrated this in class with search and sort algorithms.
ACK!!! If there is a course in assembler and it is merely being used to bash how a specific C compiler poorly optimizes a specific piece of code, and motivating future programmers rewrite algorithms in assembler, that skill is probably a waste and is gonna tempt these students to be yet another generation of people that desire to distrust and abuse the portable/maintainable code infrastructure we built up over the years and sacrifice it in the name of percieved efficiency.
Compilers are getting better and better and often have to turn off some very important optimizations when inserting assembly language (e.g, some alias detection algorithms, strength reduction algorithms, and some loop invariance detection). Also is it is sometimes totally unclear how long certain instructions will take from an assembly point of view because of microarchitectural structural hazards (which the compiler may be optimized for). Also, often you can just recode the source and get nearly identical performance (not code, performance) with existing optimizing compilers by basically doing the desired code transformations yourself in the source code (and teach nearly the same skills).
On the other hand, such a class could be doing worthwhile stuff like teaching things structured around ABI issues and maybe analyzing a compiler peephole optmization stage from the context of an assembler; or perhaps talked about linker patching, sections and other similar issues; or maybe even talked about strategies for integrating dynamically generated code or SIMD instruction into a scalar instruction flow (something which compilers are still struggling with and are important for stuff like JIT and DSP code)...
Just sayn'
Although I don't disagree in general that politicians don't really care too much what people want, that's a far too cynical attitude to think they like people who do what they're told. A much simpler explanation for their behaviour is that they are bought/paid-for since the easiest way to get re-elected is to spend lots of money (advertising for elections and bringing back pork for the local folks, etc). If they thought it would be easier to get re-elected by just figuring out what most of the people wanted and doing that, they would probably take that route.
Unfortunatly, figuring out what most people who vote want is hard and the only folks that tell them what to do are people that don't care what most of the people want and push their own agenda (e.g., lobbyists and national special interest groups) and carrot/stick them with lots of money. Also, even if a politician did what "most" of the people wanted, many of those people just don't vote, or if they do, they don't generally vote their own interests, they generally vote how other people tell them to vote (which may or may not corrolate with their own interests). There are only a very few "undecideds" that actually vote and can be actually swayed in significant numbers by figuring out what they want and doing that. Sadly, it's far easier to "energize the base" or convince people that will vote for you because other people (special interest groups that have spent lots of money) already told them how to vote and actually get them to vote.
As for the primaries, it's really a state-to-state issue. In Connecticut, they used to have a closed primary law, but when the Republicans wanted to allow independents to vote in their primary election, they challenged the law and SCOTUS overturned the Connecticut law which allowed the republicans to have a semi-open primary. In California, we now have a "top-two" system. Instead of having each party pick a candidate in a closed or semi-open primary, the top two vote getters in the primary, regardless of party affiliation, run-off in the general election (except for the presidential election which is semi-closed). It totally depends on which state you are in what the situation is. In AL, AK, GA, ID, IL, MA, MI, MN, MS, MO, NH, NC, ND, SC, TN, TX, VT, VA, WI, there are only open primaries, so your milage may vary... Both parties try to game the laws to their advantage...
.. and then there will be this shift to "fiat" BitCoin Certificates that look very much like BitCoin Certificates, but can be created out of thin air by certain entities that wish to control the economy where only some decreasing fraction of them are backed by real BitCoin...
So someone explain to me how these BitCoin Certificates will somehow be different from current money?
handbrake, parking brake?
primary and secondary brake shoes?
primary and secondary brake pistons and fluid reservoirs?
seatbelt and airbags?
Yeah, because slashdot always only carries peer reviewed research from top notch Ivy League universities.
Oh wait a second ... these papers are actually peer-reviewed results from Ivy League research universities.
Peer-review isn't immue to issues...
http://science.slashdot.org/story/12/04/06/139231/majority-of-landmark-cancer-studies-cannot-be-replicated
Also, peer review is not designed to catch fraud, only to catch errors in process or analysis.
In the Ivy League they aren't immune to issues either...
http://en.wikipedia.org/wiki/Marc_Hauser
Also (okay, these aren't researchers, but perhaps this is more relatable...)
http://www.nytimes.com/2012/08/31/education/harvard-says-125-students-may-have-cheated-on-exam.html
People are just people. Just because they are working at a fancy-schmancy university, or some other folks put another gold star next to their paper (peer review is often blinded by reputation), isn't the standard we should be touting. The only arbitor that matters is nature, as one famous physicist (RPF) put it, "Nature cannot be fooled"...
s/bamboo stick/ruler/ -> old parochial school approach...