Slashdot Mirror

From the article: "The audio was recovered by a team of scholars consisting of Lawrence Berkeley National Laboratory physicist Carl Haber, National Museum of American History curator Carlene Stephens and Library of Congress digital conversion specialist Peter Alyea, among others." See also here: http://newscenter.lbl.gov/science-shorts/2013/04/25/alexander-graham-bell/

March 2003

We don't know what the mass of a neutrino is, but we do know they're light (m
Not quite - IceCube looks for muon neutrinos and these have a mass limit of 0.19 MeV/c^2. The lowest mass constraint is actually 2 eV/c^2 for electron anti-neutrinos from tritium decay spectrum measurements.

Look, I like bashing straw man lazy programmers as much as anybody. But in scientific computing in the year 2013 - say, where you need to store 50 cubic miles of subsurface 4-dimensional seismic reflector data for 3D visualization and modeling density change over time - you run into the limits of 4 gigabytes very quickly. Never mind large-scale simulations run in TOUGH2-MP... Don't paint with such a large brush. People may piss memory on stuff that ran in less RAM back in 1996, but we're not there any more, and adventurous, relevant, and efficient uses of RAM really do exist.

Furthermore, there's a growing body of evidence that hormesis is vital for health and so a low level of exposure to radiation (ionizing and non ionizing), toxins and harmful biological entities in the environment is a good thing that promotes health.

Some fun links because I'm too lazy to find proper citations on a Saturday morning..
http://blog.sethroberts.net/2012/04/01/moderate-alcohol-consumption-associated-with-less-cirrhosis/
http://www.lewrockwell.com/miller/miller12.html
http://newsroom.ucla.edu/portal/ucla/tiny-amounts-of-ethanol-dramatically-221986.aspx
http://newscenter.lbl.gov/news-releases/2011/12/20/low-dose-radiation/

You seem not to know what the word magnitude means.
It means: by a factor of ten.
The libraries you mentioned are written in the languages you mentioned because they are used by said languages.
E.g. see colt: http://acs.lbl.gov/software/colt/
The rest of the stuff like 'multimedia' makes no much sense as this is all build in into the Java environment and if younuse it you are calling the C functions implementimg it ...

If you haven't died of a heart attack yet, plants beat us to finding a use for quantum entanglement, too.

here's an old paper describing octuple precision on the PowerPC G4

Many problems in number theory and the computational and physical sciences, espe- cially in recent times, require more floating point precision than is commonly available in fundamental computer hardware. For example, the new science of “experimental mathematics,” whereby algebraic truths are foreshadowed, even discovered numerically, requires much more than single (32-bit) or double (64-bit) precision.

That paper references Bailey's 2000 paper on Quad double algorithms, which alludes to "pure mathematics, study of mathematical constants, cryptography, and computational geometry

Except for all the actual research that goes on

Actual research results in published papers with high impact factors. NASA allocates about 1.7 billion per year to ISS, while Lawrence Berkeley National Lab uses 800 million for it's entire budget. If you want to make the claim that ISS research is cost-effective, I'd like to see your numbers.

Except for all the actual research that goes on

Actual research results in published papers with high impact factors. NASA allocates about 1.7 billion per year to ISS, while Lawrence Berkeley National Lab uses 800 million for it's entire budget. If you want to make the claim that ISS research is cost-effective, I'd like to see your numbers.

Just in the quick reading I've done, there appears to be at least a couple of different definitions involved here.

There are Biosafety Levels that discuss the level of sterilization and staff protection appropriate for the vector. Organisms with airborne infectivity (ie, Influenza) are BL 3. Interestingly, the reference that I pulled doesn't describe HIV.

For BL 3

BL3 is applicable to facilities in which work is conducted with indigenous or exotic agents that may cause serious or potentially lethal disease as a result of exposure by the inhalation route.3

Then there is Physical Containment (PC) levels:

Risk Group 2 (moderate individual risk, limited community risk) - a pathogen that can cause human, plant or animal disease, but is unlikely to be a serious hazard to laboratory workers, the community, livestock, or the environment; laboratory exposure may cause infection, but effective treatment and preventive measures are available, and the risk of spread is limited. Generally work with Risk Group 2 microorganisms shall be carried out in Physical Containment level 2 (PC2).

Risk Group 3 (high individual risk, limited community risk) - a pathogen that usually causes serious human, plant or animal disease and may present a serious hazard to laboratory workers. It could present a risk if spread in the community or environment, but there are usually effective preventative measures or treatment available. Work with Risk Group 3 microorganisms shall be carried out in Physical Containment level 3 (PC3).

Seems confusing which doesn't surprise me.

According to a recent study by LBNL the soft cost associated with installing the panels are more than three times as high in the US compared to Germany.

http://emp.lbl.gov/sites/all/files/german-us-pv-price-ppt.pdf
Page 26: Costs that are not module costs. 4.46$/W in the US compared to 1.18$/W in Germany.

Higher cost results in lower volume.

There was another event that led to modification of the natural isotopes in North America:

http://ie.lbl.gov/paleo/paleo.html

http://abob.libs.uga.edu/bobk/nuclear.html

The discovery, published in the journal Nature, means corn...

If this research was really worthwhile, they'd have published their paper publicly instead of in some elitist magazine. This kind of behavior by scientists is exactly what late Aaron Swartz denounced. Once again important research stays hidden within the confines of paywall-locked information-vaults. Great...

By the way, Berkeley itself already published about this in November.
http://newscenter.lbl.gov/news-releases/2012/11/08/more-bang-for-the-biofuel-buck/

If Berkley already published about it, then it's not really locked behind a paywall. Besides, even if it was, most researchers have access to publications, including Nature and trade journals, etc. It is usually provided by their University. Even students get access, at least at major universities. It seems your complaint is that John Q. Public can't have free access, but then JQP isn't actually doing any research based on it and if they were, the cost of a subscription would be worth it.

The discovery, published in the journal Nature, means corn...

If this research was really worthwhile, they'd have published their paper publicly instead of in some elitist magazine. This kind of behavior by scientists is exactly what late Aaron Swartz denounced. Once again important research stays hidden within the confines of paywall-locked information-vaults. Great...

By the way, Berkeley itself already published about this in November.
http://newscenter.lbl.gov/news-releases/2012/11/08/more-bang-for-the-biofuel-buck/

Yes, lots of people protested NASA's risky space launch of a nuclear reactor but failed to stop the launch. The cops treated them just like they treated OWS. Sigh.

From the link:

The Cassini rocket will be powered by 72 pounds of plutonium -- the most ever rocketed into space. Protesters say that if the rocket explodes it could sprinkle deadly poison for hundreds of miles.

Winds can blow (plutonium) into Disney World, Universal City, into the citrus industry and destroy the economy of central Florida," said Michio Kaku, a protesting physics professor from New York. He claimed that casualties could run as high as a million people if there were an accident.

What? If you split it up into 1 million 30 milligram doses and had people directly inhale it or inject it into your blood, yeah that would do it. You could injest that much and survive (cancer risk goes up, but it is well under the LD50 of 500mg for ingestion, cyanide is more lethal) But exploding it over the ocean where people are very unlikely to encounter any at all? Maybe that is the kind of science you get form a TV physicist. Make up a scary story to get yourself headlines.

As far as the OWS quip goes, some of these people did break into a secure facility by jumping the fence. Though they deserve to be arrested it is no reason for police brutality. However the article only says that there were only arrests.

Your analogy doesn't really hold much water. A suburban is much closer to a car than a BUS, and does not actually benefit in safety from its size like a bus does:
http://www.lbl.gov/Science-Articles/Archive/EETD-SUV-Safety.html

"The safest SUV, the Suburban, has at least a 40 percent higher combined risk than the three safest midsize and large cars, the Avalon, Camry, and Accord,"

I'm just reading the test results: http://vis.lbl.gov/Events/SC08/RemoteX/index.html

NX does seem to offer some improvement. That there is room for more, for example HP's SSH is much faster than Sun's (the open source one commonly used), which could help a lot as well. NX definitely seems like an interesting approach for getting 2-10x as much effective bandwidth. It also claims to reduce some of the latency problems. It doesn't appear to be faster than VNC yet, but it would be nice if NX created a fast network transparent system.

I mentioned 3 cases:

a) Running an application on a machine sharing ram with the video card.
b) Running applications on a machine close enough to the video card that the latency between them is lowish and the bandwidth is plentiful and performance is irrelevant.
c) Running applications on a machine where either the latency is high or the bandwidth is limited

If because of NX (c) became acceptable then we are really down to a much deeper choice as to whether building for (a) as the primary use case more sense than building for (b) & (c). The big issue with X11 right now is that it doesn't handle (c) well. NX seems to already make (c) better but not good enough.

Generating electricity with existing RTG technology is about 5% efficient. This paper mentions phosphored white LEDs efficiencies at 55 Lm/W. This paper says "The efficiency of the color phosphors was experimentally compared within the range up to 90 Lm/W for green, up to 30 Lm/Watt for blue, and up to 35 Lm/Wt for red color at 14 kV." [In this case kV are keV since they were shooting electrons]. This site says the decay energy of a Gd-148 alpha particle is about 3.271 MeV.

Doing the math, 1 Gd-148 alpha decay is about 5.24E-13 Joules, so 1.9E+12 decays/sec would deliver 1 Watt.

Given an alpha particle power output of 1 Watt , converting it to electricity at 5% efficiency then running LEDs at 55Lm/W would result in 2.75 Lm of light. The same alpha flux directed on the phosphors would result in a minimum of 30 Lm in the blue part of the spectrum, 90 Lm in green, and 35 Lm in red.

This paper is one of my favorites - it states, "A ~0.2 kg block of pure Gd148 (~1 inch3) initially yields ~120 watts, sufficient in theory to meet the complete basal power needs of an entire human body for ~1 century...". That's an awful lot of power packed into a tiny 1 in^3 package!

Fascinating stuff.

This is the guy who fixed TCP/IP flow control when the Internet started to undergo congestion collapse in the late 1980s. (See http://ee.lbl.gov/papers/congavoid.pdf). I submit to you that he knows a HELL OF A LOT MORE about what the Internet can and cannot do than some guy who works as an account manager at a bank and puts up crap pr0n on his website (assuming you're the same as @TonyAldo and as the owner of tonyaldo.com).

That part has me confused. He's never been a non-believer in AGW.

This is misleading almost to an extreme. Muller was well known for being skeptical of the research. He in particular attacked the "Hockey Stick" (See his article from 2004: Global Warming Bombshell. His criticisms were such that he was universally regarded as a skeptic prior to BEST (see Quotes by Richard Muller, and his skepticism regarding the research was consistent. Note that Watts initially supported BEST, and that work was financed in part by the Koch brothers. Do you think that would be the case if Muller was regarded as being a "warmist"? Your popular technology article appears convincing until you read the linked sources. From those it is clear that your are right sort of, Muller never was a AGW denier per se, but the quotes read in context show that he was highly skeptical of the research which is why he was universally categorized as being one of the AGW skeptics prior to 2010. The point is that he was open minded which is why he changed his mind.

Muller's claim that he's a "recently converted skeptic", which is a flat-out lie as he has always been a warmist.

Prove it please. This article from 2004: Global Warming Bombshell shows his earlier skeptic bonifides. True, even in this article he is concerned that global warming may be real, but he is skeptical of the research and was repeatedly so (see Quotes by Richard Muller. Remember that he did get support from the Koch Brothers who are not ones who would knowingly fund a "warmist". The main difference between him and most other skeptics was that he did not reject AGW out of hand and had a degree of open mindedness and honestly that lead him to do primary independent research which lead him to change his mind, thus proving that honest global warming skeptic doesn't always have to be a oxymoron.

It's really hard to picture what is described in the article. Here's a link to an article with a diagram.

We find the bulk modulus of M-carbon to be 365+/-38 GPa, thus is one of the stiffest materials known comparable to that of cubic-BN (387+/-4 GPa) and wurtzitic BN (375+/-9 GPa). ... M-carbon also shows anisotropic compressibilities along lattice axes: the a axis is stiffest [527+/-2 GPa] and the b [271+/-1] and c [267+/-1 GPa] axes are roughly equivalent ...

It seems that the anisotropy does give a lower compressibility, but not dramatically more as in graphite (weaker plane compressibility is 2.7% of the stronger plane). It's also clear that the diamond in the diamond anvil cell used to make this is damaged by the material. The picture in the Yale News article is the damaged anvil, not the M-carbon. In SEM images, it doesn't look like graphite at all, but more like fused grains. Characterization and proof of structure is done by X-ray diffraction, a standard materials science method, using synchrotrons, which are giant particle accelerators, namely ALS at LBL and APS at Argonne.

. For example, a proton's mass is 938 MeV but its two up quarks and single down quark only sum to (at most) 12.4 MeV.

Wouldn't that mean that the binding force of the nucleons were negative? And wouldn't that mean that they should fly apart?

You're thinking about atomic nuclei, which are slightly less massive than the individual neutrons and protons that comprise them.

But an individual nucleon (a proton or neutron) really does get a significant fraction of its mass not from its constituent quarks but from the strong force. For example, a proton's mass is 938 MeV but its two up quarks and single down quark only sum to (at most) 12.4 MeV.

I believe the binding of nuclei has a different sign than the binding of quarks into individual nucleons because quarks can't ever be physically separated. It's also much smaller; iron nickel has the most tightly bound nucleus, and its binding energy is only 8.8 MeV per nucleon. Though vastly larger than chemical energies, this is still less than one percent of its mass.

I think I found this here like 5 years ago and I've kept it since.
http://muller.lbl.gov/TRessays/22-ThePhysicsDiet.htm

I've emailed Richard last year by the way and he's still the weight he achieved in that article 9 years later.

FWIW: I'm an endomorph who DOES believe that some people hold weight easier, crave carbs and sugar more than others and have a lower BMR. However science is science - these things only make up a small fraction of the work. 95%+ is simply putting in the effort.

I can also confirm that adjusting diet is far, far far more rewarding than excercise for weight loss, despite other health benefits. Just as his article says.

I agree with what you post, but research now shows that very often, it's not a craving for carbs, but an actual addiction to them in terms of the way they effect brain chemistry. As such, just like quitting smoking or giving up drugs and alcohol, since there is a chemical dependency, it is not as easy as one would think. Obviously, just as many people can drink and not become alcoholics. Many can overeat and not become addicted to carbs. But for many, they do, and for them, will power often is not enough.

I think I found this here like 5 years ago and I've kept it since.
http://muller.lbl.gov/TRessays/22-ThePhysicsDiet.htm

I've emailed Richard last year by the way and he's still the weight he achieved in that article 9 years later.

FWIW: I'm an endomorph who DOES believe that some people hold weight easier, crave carbs and sugar more than others and have a lower BMR. However science is science - these things only make up a small fraction of the work. 95%+ is simply putting in the effort.

I can also confirm that adjusting diet is far, far far more rewarding than excercise for weight loss, despite other health benefits. Just as his article says.

If you are over-weight, there is only one way to lose weight... Consume fewer calories. End of story. And further, it is preferable that you consume just slightly fewer calories, over a long time-frame.

http://www.muller.lbl.gov/TRessays/22-ThePhysicsDiet.htm

In theory, exercise is an alternative, but really, unless you can force your flabby butt to go run a marathon, you won't even make a dent, and even if you do, your body will crave more calories and your increased food intake may eliminate any gains.

There are no alternatives, there are no short-cuts.

All the diet schemes out there are intended to get you to consume fewer calories, when you lack the will power to just jump head-long into it and do so directly...

Diet pills basically exist to make you feel slightly sick, so you won't want to eat so much. Secondarily, they may do the same thing as taking fiber before eating, causing your body to excrete more of the food you eat before it is digested and converted into calories.

Drinking water before a meal will occupy a portion of your stomach with calorie-free liquid, so you will feel you are "full" before you have consumed your normal number of calories of food. Diet foods work this same way... they have the same volume as normal foods, with a lower calorie count, so it takes more effort to stuff your face. This is similar to diets that reduce sugar... Less sugar means less calories, and also less insulin production may affect how soon you feel "full". The down-side of all of these is that your stomach may simply expand, over time, completely compensating for the reduced calorie intake per-volume of food. A similar problem for those who have their stomach "stapled", but aren't dedicated to maintaining a low-calorie diet.

The last option that comes to mind are single-food diets. They work by basically restricting the food you eat to a narrow subset, so you can consume as many calories as you need, but you'll basically get bored of eating the same thing over and over, and hopefully won't over-indulge on the same boring food.

But they all come back to reduced calories, just indirectly. And IMHO, the best option is just to SLIGHLY reduce serving sizes over time.

You can eat all the same greasy and wonderful fast food and junk food you want, you just have to eat LESS of it. Get the single burger instead of the double burger... Or get the burger you want, and just don't get the sides (french fries, onion rings, etc)... Or get all the same food, and just drink water with it, instead of soda/juice/milk. All of which will reduce your calorie intake, and over time, your weight. And since you're still able to eat all the same food, I believe this is, by FAR, the easiest diet to stick to.

I personally lost about 60lbs with this strategy... First it was smaller versions of the same fast food, and/or fewer side dishes (which saved me a lot of money, too... "super-sizing" FEELS like a good value for the money, but in fact spending less is ALWAYS the better deal).

Then it turned into rotating between a few single-course cheap and simple meals at home... rice, potatoes, pasta, ramen, sandwiches, etc. with just the occasional splurging on a quick run for greasy fast food when I had a craving for it. And that became less and less frequent, too.

As a fringe benefit, as you stay just slightly hungry, your energy level goes up, and you have the URGE to exercise (you don't have to force yourself to do so). And as you lose weight, exercise is also easier, and endurance goes up. But most importantly, the added muscle mass will serve you well in maintaining your weight... If you go back to eating poorly, whether for a short or medium term, the muscle will do a great job of helping your body consume the excess calories, turning it into more muscle, or heat, instead of fat.

Now if you'll excuse me, I need a cookie.

I'm somewhat more hopeful than you, based on advances in x-ray optics.

For typical x-ray photons (e.g. 10 keV), the refractive index is 0.99999 (delta = 1E-5). Even though this is very close to 1, we've figured out how to make practical lenses. For instance Compound Refractive Lenses use a sequence of refracting interfaces to accumulate the small refractive effect. Capillary optics can be used to confine x-ray beams. A Fresnel lens design can be used to decrease the thickness of the lens, giving you more refractive power per unit length of the total optic. In fact, you can use a Fresnel zone plate design, which focuses the beam due to diffraction (another variant is a Laue lens which focuses due to Bragg diffraction, e.g. multilayer Laue lenses are now being used for ultrahigh focusing of x-rays). Clever people have even designed lenses that simultaneously exploit refractive and diffractive focusing (kinoform lenses).

All this to say that with some ingenuity, the rather small refractive index differences available for x-rays have been turned into decent amounts of focusing in x-ray optics. We have x-rays optics now with focal lengths on the order of meters. It's not trivial to do, but it can be done. It sounds like this present work is suggesting that for gamma-rays the refractive index differences will be on the order of 1E-7, which is only two orders-of-magnitude worse than for x-rays. So, with some additional effort and ingenuity, I could see the development of workable gamma-ray optics. I'm not saying it will be easy (we're still talking about tens or hundreds of meters for the overall camera)... but for certain demanding applications it might be worth doing.

It's too small to be useful in tradition lens applications. Even at lower X-ray/Gamma energies when the refractive index is much bigger ( see: http://henke.lbl.gov/optical_constants/getdb2.html) you can't really use make reasonable lens.

It would be far far easier to increase the "commonly" available divergent sources' intensity than attempt to recover losses due to divergence with such a weakly focusing system. Hell, you could likely achieve a much bigger intensity increase by moving the source closer.

Can anyone even think of the medical imaging system they are envisaging. I assume they mean a phase-contrast approach but that needs a intense 700kV source with a small spot size and thin structured grating that absorb 700kV photons (which I don't think exist thou it's maybe just about within the capacity of man to make them).

The only application I can think of for this is for X-ray telescopes and astrophysics where you can't increase the source but you can have massively long imaging systems. However I will bet that their are not many interesting astrophysical events that emit only 700kV+ photons and the attenuation of the massive silicon lens may well counteract the benefit of the focusing.

From what I understand, this is not absolutely definitive, but cancer researchers at Lawrence Berkeley Lab published a paper where they used imaging of cellular responses to radiation damage to show that at low levels, it appears that cells repair DNA damage due to radiation very effectively.

Seriously, follow that link, and learn.

You are probably alluding to discovery that bi-layer or tri-layer graphene stacks can be induced to have a tunable bandgap... http://www.lbl.gov/msd/assets/docs/highlights/09-9FengWang_bilayer_graphene.pdf

if you ever meet a perfect version of yourself constructed by aliens from radio transmissions, don't shake their hand.

For those who don't know, this refers to a story in the Feynman Lectures on Physics. Here's my version; I've taken some liberties.

Imagine you're on the phone with an alien who speaks English, except they don't know what "left" and "right" mean. You want to explain it to them so they know which tentacle they should use to shake the right-handed President's hand if they should ever meet. The alien can be anywhere in the universe, so you can't refer to stellar positions or similar, leading you to devise an experiment for them to perform.

Your initial attempts use gravity, electricity, and magnets, but you notice each experiment comes out essentially the same if you swap "right" and "left"--for instance, you could give the alien instructions for making a clock in hopes of defining "right" using clockwise rotation, except if the alien made the clock exactly backwards by reversing the notion of "left" and "right", they wouldn't be able to tell. A particle physicist happens by and tells you about a magical experiment involving the weak nuclear force that *does* distinguish left and right inasmuch as the experiment fails if the alien screws up "right" and "left" and succeeds otherwise. (For the curious, some more details here and here.) Great, problem solved.

"But wait!" the physicist says. "The alien needs to use regular matter instead of antimatter in the experiment. The results will be reversed otherwise! Come to think of it, I have no idea how to tell them the difference between matter and antimatter. If you ever meet them and they start trying to shake your left hand, RUN, since the alien will be made of antimatter!"

I just want to provide a little context to this announcement. As shown in the article, this is a Baryon, made up of 3 quarks. With 6 possible types of quarks, and 3 spots, this makes for many possible combinations of Baryons, a lot that have been found. Here is a current list of baryons:

PDG Baryon List

The proton and neutron are the p and n in the top left. The new Cascade (Xi_b) will be in the bottom right, in the "Bottom quark" section.

So this is neat and all, but hyped up a bit because its the LHC. A couple of these new Baryon (and also Mesons) are confirmed every year.

That's really, really nearby... if you define 'nearby' like the Nearby Supernova Factory I'm in - 400 million to 1 billion light years.

This one is only 15.4 million light years away. So close it could practically order pizza.

How does building another refinery result in a step toward energy independence? And how, exactly is this post 'insightful' ?

More importantly, are you sure this is the right thing to do? At $12 per watt ($14 billion, 2.2GW). That's several times more than the installed cost of solar, according to Lawrence Berkeley National Laboratory. And that doesn't include the cost of running the plant, disposing of radioactive waste that is a hazard for billions of years. Even if storage for this waste is dirt cheap, it would certainly add up over 4.5 x 10^9 years.

The numbers are in and it turns out SUVs are NOT safer.

Sure, they're safer in a head-on collision but that's one of the least frequent types of accident. In most other types of accident the SUVs tend to flip over and kill everybody inside. Ref: http://eetd.lbl.gov/ea/teepa/pdf/TRB_Safety_1-03.pdf

Ironically, the feeling of 'safety' also causes more SUV accidents because the drivers take more risks.

seeing that she escaped death by inches only by flat-spotting all four tires hitting the brakes

ABS may have reduced the stopping distance. If you're flat-spotting tyres, you don't have ABS.

Aside from that, SUVs are safer than sports, compacts, and subcompacts, but not by a multiple of 5. On the flip side, compacts and subcompacts are much safer if you happen to be the other driver.

See page 8

The best all round choice seems to be luxury imports, which are least risky for everyone. But they are expensive ; minivans seem to be a good compromise - they're not as forgiving on the other driver as mid to large size cars, but they are much safer for you. SUVs seem to be nearly twice as dangerous as a large size car to the other vehicle and slightly less safe for you than a mid to large size car, which I presume is what your wife was driving if she was getting 25 mpg.

Did you have a citation for that "5 times safer" figure?

Sorry for the repeat, but here is a better link, very informative: http://www.lbl.gov/solar/ipfiles/plenary/chu_Solar_to_Chem_Energy_3-28-05.ppt

Here you go: http://www.lbl.gov/solar/ipfiles/plenary/chu_Solar_to_Chem_Energy_3-28-05.ppt
I see your point. I don't understand why what I wrote is somehow a contradiction. If we build a nuclear plant every day there will be a huge nuclear waste and proliferation problem. This is unsolved.

I recently became convinced by an argument made by Lawrence Berkeley Lab scientists that solar is the only power source that we have that really makes sense for powering human needs in the future. Check it out here http://www.lbl.gov/solar/

I'm doing my PhD in HPC. From my perspective GPUs do indeed offer a lot of GFLOPS but it's often impossible to max them out. Especially for stencil codes (read: virtually all physical simulation codes) this is hardly possible because of the low operational intensity of stencils. CPUs achieve much higher efficiencies here because they can do cache blocking. The caches on e.g. Fermi are much too small to do that well. So no: in this case a GPU won't necessarily yield you a higher performance or efficiency

As for the pricing: in scientific computing we typically care for double precision. For that you have to by Nvidia Tesla cards, which cost several thousand dollars, each. I don't know the prices of the Sparc chips, but I doubt they'll be higher.

No, I'm not. I'm saying that's what the GP and the GGP do. A sceptic is someone who demands better evidence, a denialist is someone who changes the terms after evidence was provided. You can look at Muller's actual claims back then:

Last month's article by McIntyre and McKitrick raised pertinent questions. They had been given access (by Mann) to details of the work that were not publicly available. Independent analysis and (when possible) independent data sets are ultimately the arbiter of truth. This is precisely the way that science should, and usually does, proceed. That's why Nobel Prizes are often awarded one to three decades after the work was completed—to avoid mistakes. Truth is not easy to find, but a slow process is the only one that works reliably.

It was unfortunate that many scientists endorsed the hockey stick before it could be subjected to the tedious review of time. Ironically, it appears that these scientists skipped the vetting precisely because the results were so important.

Let me be clear. My own reading of the literature and study of paleoclimate suggests strongly that carbon dioxide from burning of fossil fuels will prove to be the greatest pollutant of human history. It is likely to have severe and detrimental effects on global climate. I would love to believe that the results of Mann et al. are correct, and that the last few years have been the warmest in a millennium.

Love to believe? My own words make me shudder. They trigger my scientist's instinct for caution. When a conclusion is attractive, I am tempted to lower my standards, to do shoddy work. But that is not the way to truth. When the conclusions are attractive, we must be extra cautious.

That's actual scientific scepticism. Being biased against the consensus is something entirely different. Selecting to believe only what people who share your biases tell you is not scepticism. Attacking someone for lying due to which label the media has chosen to stick onto them is not scepticism, it's libel.

Who's talking about disk I/O? I'm talking about network bandwidth, which is required for synchronization, e.g. to update ghost zones in stencil codes. The required bandwidth is proportional to the computational power of the nodes. Latency can actually be hidden, too, by overlapping computation and communication -- at least for the afore mentioned stencil codes, which represent the largest fraction of simulation codes out there. That said, disk I/O is still vital, at least if you want to actually see what your super computer has computed. Of course peak performance is never actually achieved, but a 10 PFLOPS machine is useless if production codes all run at 10 TFLOPS. That's why Blue Waters did target 1 PFLOPS application performance -- that meant more to them than 10 PFLOPS Linpack throughput.

I'm really sorry, but... you are ignorant, quite literally. You've ignored what people have written to you about neutrino flavor mixing. You're willfully ignorant, I'm afraid, and that certainly makes you appear arrogant.

Your theory (if I'm able to understand) is that this effect is due to coherent flavor-exchanging forward neutrino scattering. Others have replied, myself included, with some degree of clarity as to why this is not the case. The "why" of neutrino oscillation is because their mass eigenstates, which are fixed over time, are not identical to their flavor eigenstates. Therefore, their flavors have an increasing probability of varying over time.

A bit more theoretical: There is a mixing angle in Hilbert space between the mass and flavor eigenstates. (Hope I remembered that correctly. My theory-fu doesn't go much deeper these days. It's been a while.) If this sounds like gobbledegook, get yourself a good elementary book on QED and electroweak interactions and settle in for a new, perhaps difficult, understanding.

Better yet, this looks like a definitive on-point source:

http://pdg.lbl.gov/2005/reviews/numixrpp.pdf

from LBL by a FNAL author. Enjoy!

(The one up on the hill.)

Early data on stuff from Keck, Gemini or Subaru is rare, yeah, unless you have a bunch of Target-of-Opportunity time or can persuade people to take a few shots for you during their programs. But even on Mauna Kea, there are lesser (but still "huge" to most people) scopes where time's easier to get, so when your survey pipeline (from KAIT or PTF or QUEST or whatever you want) throws you a new target, you don't have to wait more than a couple nights before going after it.

We have somewhere around 40% of the time on the 2.2-meter on Mauna Kea, which is our usual tool for going after SNe, although of course some of the bigger names in the collaboration (Perlmutter, Aldering) get time on Keck as well.

If there hadn't been a lightning strike at the 2.2-m during last weekend's snow-and-lightning storm, I would be observing the M51 supernova this evening (and not for the first time). Pesky lightning! :(

People in the larger vehicle of a 2 vehicle crash, tend to have longer lifespans post-incident.

I refer you to Wenzel, T.; Ross, M. (2003), Are SUVs Safer than Cars? An Analysis of Risk by Vehicle Type and Model. Short answer: not necessarily. (Possible reason: SUVs don't have crumple zones as effective as those of smaller vehicles, so the occupants are subjected to higher acceleration.)

In other words: yes, this "Duh" science is necessary, because sometimes your clear and obvious rules are wrong.

Hmm, nice to see that JPEG XR actually supports some decent HDR pixel formats (OpenEXR 16-bit float RGB, and the Radiance shared-exported format) -- IIRC, its progenitors at MS were notably inferior in this area.

Now if they can only get the patent situation sorted to everybody's satisfaction, and some good free libraries get written, it might actually be a contender!

Ok sure, go ahead and run a 8k x 8k Linpack, tell me how that goes and how non-limited you are.

I guess if 8kx8k Linpack refers to matrix multiplication, I'm pretty sure that the Cell will perform at close to its 200GFLOP/s performance. Matrix multiplication can be really well broken down into block-matrix multiplications that nicely fit in the 256KB of available SRAM. Data transfer cost per block grows O(N^2), while FLOPs grow O(N^3). With 64x64 block matrixes you have to transfer 2*64*64 floats while having to compute 32 times as many multiply&adds. With 8 SPUs sharing the single XDR RAM you'll only have bandwidth to transfer 1/4 float per cycle and core. However, corresponding to that transfer rate, the core has to compute 32*1/4=8 madds per cycle, which happens to be twice the theoretical peak performance of the core. So no bandwidth problem at all. You'll be able to increase block size to 96x96 if you want to reduce bandwidth per FLOP even further. this paper claims close to peak performance for 2304x2304 sized matrices.

Sorry man, the Cell is fine for some things but the idea that it doesn't face the same realistic limits other hardware does is silly. You can talk all you like about high speed stuff on the cache, but that applies only for things that'll fit in there. When you have larger problem sets that have to go back and forth to main memory a lot, I'm afraid it isn't so fast.

You'll have these kind of losses on any architecture, including GPUs. What i'm saying is that the Cell is designed so that you'll almost always be able to reach close to 100% utilization of the available peak FLOP/s. That is something that you won't ever normally achieve on a GPU. On the other hand, GPUs have a much higher peak FLOP/s so loosing half of that due to bandwidth problem seems to be more acceptible.

Regardless my point was simply how unrealistic it is to call the thing a supercomputer. If a couple hundred GFLOPS makes a supercomputer then my GPU is a supercomputer.

Don't downplay what GPUs can do computationally either. They are the kings of Folding, yes ahead of the PS3. So long as your problem meets some requirements (highly parallel, single precision FP, fits in to GPU memory, not a lot of branching and when it branches everything branches the same direction) they scream. Is that all things? No, certainly not, you can find things they drag ass on. However the same happens with the Cell when compared to something like a Core i7. For some things, due to the SPEs the Cell is faster, however for others, due to constraints of the PPE it is slower.

Well it's not the constraints on the PPE that make the Cell slow. The Cell is just as fast as 200GPLOP/s can be. Modern GPUs are much faster than that, even if they cannot utilize all their horsepower. The downside is that GPUs rely on very different programming&compiler paradigms, where Cell was designed to mostly use multi-core, shared memory pradigms and standard C compilers with SIMD extensions, everything nicely managed by a standard MMU-utilizing operating system.

It is an interesting architecture and useful for some things, but it is not particularly impressive compared to other modern processors. Doesn't mean it is worthless, just that it is not "OMG this is so fast!".

It was that fast when it came out. Unfortunately development of the successors was cancelled by IBM, probably due to GPGPU programming eating its lunch.