[Fudzilla via MaxConsole]

According to Fudzilla, the platform will include a new memory controller to accommodate DDR3 RAM which, despite supporting two slots, will remain single-channel. Pineview, the next Atom before Cedarview, hasn’t even shipped yet — that’s expected to be the beginning of next year — so it’s interesting to glimpse this far into the future, where Atom, such as it is, will remain positioned almost exactly where it is now. Oh well! [Fudzilla]

Disclaimer: That Atom sitting on the penny up there is an older version (not that the new one will look any different).

Cats: they’re kinda dumb. They only seem smarter than dogs because they’re not so friendly, and our society judges kindness harshly. It’s a true and interesting theory! Which is why, after mice, simulating a feline-sized brain on a BlueGene/P supercomputer was next on IBM’s to-do list. But for all the kitty talk here, this project wasn’t specifically about creating a computerised house pet; it’s part of a larger, ongoing project to eventually simulate a full human brain. The cat equivalency, derived from the number of virtual neurons and synapses the simulation can manage, at 1.6 billion and 9 trillion, respectively, just gives a sense of how far along the project is: Today, despite being the biggest simulated brain ever, it’s only capable of simulating the human visual cortex, or as PopMech so delicately puts it, “the wrinkly outer layer” of the human brain.

So how long before a supercomputer can simulate (roughly — since these computer simulations don’t have the same neural patterning and learning capabilities of a real brain, among other things) an entire human cortex? Weirdly soon, says the project’s lead scientist:

To [simulate a human cortex], he’ll need to find 1000 times more computing power. At the rate that supercomputers have expanded over the last 20 years, that super-super computer could exist by 2019. “This is not just possible, it’s inevitable,” Modha says. “This will happen.”

People need to stop getting worked up about the future, honestly: Before we have to worry self-aware robot uprisings, we’re going to have to deal with decades of extremely dumb, extremely expensive fake pets. Enforced caution, I believe this is called. [Popular Mechanics]

In other words, this is the first time anyone has developed a processor that can handle any set of instructions for more than one quantum bit or “qubit”. Rapid progress when you consider that the first single-task quantum processorfirst arrived on the scene less than a year ago.

The NIST team performed 160 different processing routines on the two qubits. Although there are an infinite number of possible two-qubit programs, this set of 160 is large and diverse enough to fairly represent them, Hanneke says, making the processor “universal.” The researchers used a random number generator to select the particular routines that would be executed, so all possible programs had an equal chance of selection. The random programs avoided the possibility of bias in testing the processor in the event that some programs ran better or produced more accurate outputs than others. Each program operated accurately an average of 79 per cent of the time across 900 runs, each run lasting about 37 milliseconds. To evaluate the processor and the quality of its operation, NIST scientists compared the measured outputs of the programs to idealised, theoretical results.

The programs did not perform easily described mathematical calculations. Rather, they involved various single-qubit “rotations” and two-qubit entanglements. As an example of a rotation, if a qubit is envisioned as a dot on a sphere at the north pole for 0, at the south pole for 1, or on the equator for a balanced superposition of 0 and 1, the dot might be rotated to a different point on the sphere, perhaps from the northern to the southern hemisphere, making it more of a 1 than a 0.

Huh? Yeah, it’s a bit confusing, but you can get the basic idea by checking out our Giz Explains on the subject. Just know that programming with multiple qubits is a major turning point in creating the truly “super” computers of tomorrow. [NIST and Ars Technica]

7.08GHz. That’s the record-shattering speed an AMD Phenom II processor was overclocked to using a massive amount of liquid helium. You can watch the process in this documentary video set to a kickass soundtrack. [Crunchgear]

On top of those thermal issues, testing done by computer builder Haleron revealed that the N330 was also a killer on batteries in comparison to the, now used, two-chip N270 processor. Guess that explains why resellers have been trying to get rid of the N330 as quickly as possible. [Newswireless]

The key feature about Arrandale, versus current Core i5/i7 processors, is that it’s manufactured using a 32nm process, meaning it’ll be less power hungry. Remember the jump to the Penryn Core 2 chips a couple of years ago? Same speeds, better efficiency? Like that. The three coming out in the first half of 2010 are the Core i7-640UM (1.2GHz), Core i7-620UM (1.06GHz) and Core i5-520UM, all for “ultra thin” laptops.

We’ll also see some 32nm chips for the desktop, codenamed Clarksdale, announced in January. Mmm, chips. [Digitimes]

Bobcat’s a low power chip at Atom, like for netbooks, that can run using less than 1W of power. It’ll be manufactured using a 32nm process. It’s not hitting until 2011 though, and as Ars’ Jon Stokes points out, by then Intel will have been at 32nm with Atom for a while, and already close to going to 22nm.

Bulldozer is AMD’s new server architecture, also slated for 2011 which uses a new design with two “tightly linked cores” as the heart. You can read about it in way more detail over at Ars, though for now, Stokes says it’s still hard to tell how competitive they’re going to be with Intel’s 2011 lineup. Guess we’ll see. [Ars, Image via Michele Eve/Flickr]

To put that in perspective, 10 megawatts is the kind of energy a small hydroelectric plant produces — 20 watts is only enough juice to power up a weak light bulb. Amazingly, your physical brain runs on this minuscule amount of power, and it’s not very efficient. However, embracing this inefficiency could be the key to creating computers that mimic the human brain.

It sounds cockamamy, but it is true. Scientists have found that the brain’s 100 billion neurons are surprisingly unreliable. Their synapses fail to fire 30 per cent to 90 per cent of the time. Yet somehow the brain works. Some scientists even see neural noise as the key to human creativity. Boahen and a small group of scientists around the world hope to copy the brain’s noisy calculations and spawn a new era of energy-efficient, intelligent computing. Neurogrid is the test to see if this approach can succeed.

Most modern supercomputers are the size of a refrigerator and devour $US100,000 to $US1 million of electricity per year. Boahen’s Neurogrid will fit in a briefcase, run on the equivalent of a few D batteries, and yet, if all goes well, come close to keeping up with these Goliaths.

So far Boahen has managed to squeeze a million neurons onto his new supercomputer compared to only 45,000 last year. By 2011, he hopes to have 64 million up and running, bringing the project to the equivalent of a mouse’s brain.

Ditching reliability and efficiency in favour of organised chaos flies in the face of everything that an engineer holds dear, but the approach does make sense — and reducing the power consumption is the key to upholding Moore’s law. But how will this development change our perception of what an artificially intelligent robot might become? Instead of some cold, logical machine that can think for itself, we might end up with robots that are just as stupid and flawed as we are. Think about it. [Discover Mag via PopSci]