multicore
Latest
ARM's latest CPUs are ready for an AI-powered future
ARM processor technology already powers many of the devices you use every day, and now the company is showing off its plans for the future with DynamIQ. Aimed squarely at pushing the artificial intelligence and machine learning systems we're expecting to see in cars, phones, gaming consoles and everything else, it's what the company claims is an evolution on the existing "big.Little" technology. Originally unveiled in 2011, that design allowed for multicore CPU designs with powerful, power-hungry chips to do the heavy lifting tethered to smaller, low-power chips that could handle background processing when a device is idle. It's why your phone can edit HD or even 4K video at one moment before sleeping throughout the night without losing all of the battery's charge. DynamIQ lays out a strategy for processors that combine cores specifically designed for whatever task is needed.
A single fiber strand could carry the world's internet traffic
Researchers in the US and Netherlands have managed to transmit data at 255Tbps across a single strand of fiber cable over a kilometer (0.6 miles), about 2,500 times faster than any commercial fiber. They used a so-called multicore cable with seven separate channels, but the hardware alone didn't account for the speed. They also squeezed 50 carriers down the seven cores, cranking each up to 5.1 Tbps using "spatial multiplexing." None of that tech alone is new, but the net result of that was 255Tbps (31.8 Terabytes per second), enough to handle the world's peak internet traffic, according to ExtremeTech. Don't expect a speed boost at home anytime soon -- there's no way to mass produce the cables yet, and current infrastructure wouldn't support it anyway. But at least you can look forward to a day when you could download all 317,060 movies in the IMDB in two and half hours.
Insert Coin: The Parallella project dreams of $99 supercomputers
In Insert Coin, we look at an exciting new tech project that requires funding before it can hit production. If you'd like to pitch a project, please send us a tip with "Insert Coin" as the subject line. Parallel computing is normally reserved for supercomputers way out of the reach of average users -- at least at the moment, anyway. Adapteva wants to challenge that with its Parallella project, designed to bring mouth-watering power to a board similar in size to the Raspberry Pi for as little as $99. It hopes to deliver up to 45GHz (in total) using its Epiphany multicore accelerators, that crucially, only chug 5 watts of juice under normal conditions. These goliath speeds currently mean high costs, which is why they need your funds to move out of the prototype stage and start cheap mass production. Specs for the board are as follows: a dual-core ARM A9 CPU running Ubuntu OS as standard, 1GB RAM, a microSD slot, two USB 2.0 ports, HDMI, Ethernet and a 16- or 64-core accelerator, with each core housing a 1GHz RISC processor, all linked "within a single shared memory architecture." An overriding theme of the Parallella project is the openness of the platform. When finalized, the full board design will be released, and each one will ship with free, open-source development tools and runtime libraries. In addition, full architecture and SDK documentation will be published online if-and-when the Kickstarter project reaches its funding goal of $750,000. That's pretty ambitious, but we're reminded of another crowd-funded venture which completely destroyed an even larger target. However, that sum will only be enough for Adapteva to produce the 16-core board, which reportedly hits 13GHz and 26 gigaflops, and is expected to set you back a measly $99. A speculative $3 million upper goal has been set for work to begin on the $199 64-core version, topping out at 45GHz and 90 gigaflops. Pledge options range from $99 to $5,000-plus, distinguished mainly by how soon you'll get your hands on one. Big spenders will also be the first to receive a 64-core board when they become available. Adapteva's Andreas Olofsson talks through the Parallella project in a video after the break, but if you're already sold on the tiny supercomputer, head over to the source link to contribute before the October 27th closing date.
Windows Phone 8 to support multi-core CPUs, HD resolutions, SD cards and NFC
Microsoft is on stage at the Windows Phone Developer Summit offering us a bite of what's to come in Windows Phone 8, and one of the tastiest morsels may just be the noticeably more diverse hardware it will support. The new platform won't just support dual-core processors -- it will support as many as 64 cores, should such massively parallel chips come to exist in the platform's lifetime. Also gone is that long-criticized 800 x 480 display resolution ceiling: if phone builders like, they can either opt for the increasingly common 1280 x 720 or a rarer 1280 x 768. A few subtler feature parities are coming with the upgrade, such as NFC for tags and payments as well as a long, long requested support for SD cards beyond the crude initial expansion. All told, Microsoft just brought Windows Phone right up to hardware parity with its biggest rivals, and possibly a bit beyond. To check out the latest updates from Microsoft's Windows Phone event, visit our liveblog! %Gallery-158713%
Tilera sees sense in the server wars, puts just 36 cores in its newest processor
While Tilera's forthcoming 100-core processors threaten to set off fire alarms around the world, the company has finally brought out its more sensible 36-core variant. The 1.2GHz Tile-GX36 sips just 24 watts and is designed to be especially handy with short and sharp jobs like processing internet transactions. It's a reduced instruction set (RISC) chip, so it's less power hungry and cheaper than Intel's x86 silicon. It also sports 64-bit architecture, whereas rival ARM is set to remain 32-bit until 2014. Then again, with Tilera lagging behind in terms of brand recognition and software support, a two-year head start might not be long enough.
Eyes-on the innards of Fujitsu's K supercomputer (updated)
Fujitsu's K supercomputer was on our radar before it was even completed, and naturally, we let you know when it smoked the competition and became the supercomputing speed king. So, when we had the opportunity to see a piece of K at Fujitsu's North America Technology Forum today, we couldn't pass it up. In case you forgot, K is a massive machine powered by 864 racks with 24 boards per rack housing SPARC64 CPUs. We got to see one of those boards, and Yuichiro Ajima -- who designed the inter-connection chips (ICC) on them -- was gracious enough to give us some more info on this most super of supercomputers.As you can see in the gallery above, each board has extensive plumbing to keep the SPARC silicon running at a manageable 32 - 35 degrees Celsius (90 - 95 Fahrenheit) under load. Underneath that copper cooling system lies four processors interspersed between 32 memory modules (with 2GB per module) and four ICCs lined up next to the board's rack interconnect ports. Currently, the system takes 30 megawatts to do its thing, though Ajima informed us that K's theoretical max electricity consumption is about double that -- for perspective, that means K could consume the entire output of some solar power plants. When asked if there were plans to add more racks should Fujitsu's supercomputer lose its crown, Ajima-san said that while possible, there are no plans to do so -- we'll see if that changes should a worthy opponent present itself.Update: Turns out the K's power consumption resides around 13 megawatts, with a max consumption of 16MW at its current configuration. The facility in Kobe, Japan where K resides can deliver up to 24 megawatts, so expansion is possible, but none is currently planned.
ARM unveils Cortex-A7 processor, 'big.LITTLE' computing
Fancy a glimpse of the future? That little psychedelic beauty on the right is ARM's brand new Cortex-A7 processor. Its spec sheet might not seem so colorful at first glance, because it doesn't really do things any faster than existing high-end smartphone processors. However, this UK-based chip designer isn't known for bumping its gums, so it pays to look a little deeper. For a start, the Cortex-A7 is built using a 28nm process that makes it five times smaller and more efficient than the current-gen Cortex-A8. It's also cheap enough to power sub-$100 handsets, so we could be pulling GSII-like tricks on budget phones within a couple of years. Is that it? Nope, there's more: perhaps the most important feature of the A7 is that it can be combined with much higher-power cores like the Cortex-A15 side-by-side on the same chip. This allows a super-phone or tablet to switch between two totally different processing units depending on how much power is needed at the time. ARM calls this "Big.LITTLE" computing," and a similar concept is already in use on NVIDIA's Tegra 3 (aka Kal-El) SoC, which we'll see imminently in the next Asus Transformer. However, the Tegra 3 uses five identical Cortex-A9 cores, whereas a device that mix-and-matches the A15 and A7 could potentially deliver higher highs and lower lows, giving you speed when you need it and amazing battery life when you don't. How cute is that? Full PR after the break.
AMD FX processor brings eight cores to battle, we go eyes-on (video)
AMD fans have endured a long wait for this, while being reduced to spectators as Intel spews out an ever-increasing horde of Sandy Bridge variants and builds up the hype around its next-gen Ivy Bridge architecture. But the new FX series of processors is finally here and will be available to buy in the next few days, with the top-end FX-8150 priced at $245 in exchange for eight cores, a 3.6GHz base clock speed and easy over-clocking to 4.8GHz using the packaged Overdrive software. Your AM3+ motherboard is crying out for the upgrade, but don't succumb until you've clicked past the break -- we've got details of the full range and pricing, our initial impressions and an eyes-on video that includes a detailed chat with the guys from AMD.
Engadget Primed: are multi-core chips worth the investment?
Primed goes in-depth on the technobabble you hear on Engadget every day -- we dig deep into each topic's history and how it benefits our lives. Looking to suggest a piece of technology for us to break down? Drop us a line at primed *at* engadget *dawt* com. My, how times have changed over the last eight months. At CES 2011, we ecstatically witnessed the introduction of mobile devices with dual-core CPUs and drooled over the possibilities we'd soon have at our fingertips. Now, we look down at anything that doesn't have more than one core -- regardless of its performance. Not only are these new chips quickly becoming mainstream, Moore's Law is in full effect with our handheld devices since tri-core and quad-core systems are just over the horizon. We can't even fathom what's in the pipeline for the year 2015 and beyond (we don't think we're too far away from that 3D shark seen in Back to the Future 2). Let's not get too far ahead of ourselves here, however. After all, we first need to wrap our puny human minds around the idea of what this newfound power can do, and why it's changing the entire landscape of smartphones and tablets. In this edition of Primed, we'll focus on why multicore technology makes such a difference in the way we use our handheld devices, whether we should even consider purchasing a handset with a single-core chip inside, and why one-core tech is so 2010. Check out the whole enchilada after the break.
ARM hopes to strengthen grip on mobile PCs, take 50 percent of the market by 2015
We've already heard rumors that chip designer ARM has been trying to get its wares into the Macbook Air. While we can't add anything to that particular story, we do have further evidence that ARM is going beyond smartphones and tablets in order to target bigger form factors. The company's president, Tudor Brown, has just appeared at Computex to declare that ARM wants to conquer the "mobile PC market", where the company currently only has a 10 percent share. He's aiming for 15 percent by the end of this year, and an Intel-provoking 50 percent by 2015. "Mobile PC" is a pretty ambiguous category, but we think it's safe to assume the focus is on low- and mid-power netbooks and ultraportables. Such devices could potentially run off ARM's forthcoming multi-core chips -- like perhaps the quad-core beast inside NVIDIA's mind-blowing Kal-El processor, or the more distant Cortex-A15. It's hard to imagine these tablet-centric chips ever competing with Intel's top performers, but four years is a mighty long time in this business.
ARM predicts dual-core Cortex-A15 devices in late 2012, quad-core variants 'later on'
Smartphones and tablets, the two hottest categories of consumer devices right now, are dominated by ARM processor designs, so when the company speaks up about its product roadmap, we're inclined to listen in close. The next generation of ARM system-on-chip has been dubbed Cortex-A15 and was expected to ship in 2013, but that's now been accelerated slightly to late 2012, which is when we're told to expect actual devices on sale with A15 silicon on board. Single- and dual-core variants will get us started, before quad-core options start filtering through in 2013. ARM promises a stunning fivefold improvement in performance over current Cortex-A9 SOCs and already has NVIDIA, Samsung, ST-Ericsson, and Texas Instruments signed up as licensees for that new hotness. So now even Samsung's "desktop-class" 2GHz dual-core chip that's slated for 2012 has a reason to look over its shoulder. Happy times!
Researchers tout self-repairing multi-core processors
The race for ever-tinier computer chips is on, and barring physical limitations, doesn't seem to be slowing anytime soon -- but with chips, as with humans, the smaller they get, the more fragile they become. A team of researchers called CRISP (Cutting edge Reconfigurable ICs for Stream Processing) is working to create a self-repairing multi-core processor that would allow on-chip components to keep on shrinking, while combating concerns over accelerated degradation. Basically, the team's conceptualized a chip that allows for 100 percent functionality, even with faulty components. With multiple cores sharing tasks, and a run-time resource manager doling out those tasks, the chip can continue to degrade without ever compromising its intended functions -- a process CRISP calls graceful degradation. Once one core fails, the on-chip manager assigns its task to another core, continuing on in this fashion for the complete lifetime of the chip. Of course the technology is still in its infancy, but if CRISP's chips comes to fruition, we could see virtually indestructible processors that make 14nm look bulky by comparison.
LG Optimus 2X review
The world cried out for a dual-core smartphone and LG and NVIDIA answered the call. Actually, the world only ever dreamt about multicore mobile architectures up until late last year, but sometimes that's all it takes to get those zany engineers engineering. So here we are, in early February 2011, beholding the world's first smartphone built around a dual-core processor, the Optimus 2X. This is a landmark handset in more ways than one, however, as its presence on the market signals LG's first sincere foray into the Android high end. Although the company delivered two thoroughly competent devices for the platform with the Optimus S and T in 2010, they were the very definition of mid-range smartphones and the truth is that Samsung, HTC and Motorola were left to fight among themselves for the most demanding Android users' hard-earned rubles. So now that LG's joined their ranks, was the wait worth it? %Gallery-115835%
Sony's next-gen PSP (NGP) has a quad-core ARM Cortex-A9 processor, quad-core GPU as well
You know that crazy next-gen PSP (NGP) with multiple touchpads, dual analog sticks, and quadrupled resolution that Sony just trotted out? Yeah, it's got a quad-core Cortex-A9 and a quad-core Imagination Technologies PowerVR SGX543MP4+ GPU doing the grunt work within. We've never seen a handheld this powerful. Then again, considering the darn thing won't be launching until this holiday season, maybe quad-core parts will be the least Sony will need in order to match up to the "super phones" coming up this year. We're just wondering how long any of these souped-up portables will last on a charge. Full spec sheet after the break.
NVIDIA Tegra 3, equipped with 1.5GHz quad-core madness, teased by a familiar slide
How aggressive can NVIDIA get? That's the question puzzling our brainboxes right now as we gaze upon the complete version of the slide that let us know about a potential Tegra 2 3D chip over the weekend. It's not every day you hear of a 1.5GHz quad-core mobile SOC, but our discovery of corroborating evidence for the T25 module sitting alongside it makes us more willing to credit the possibility of a Blu-ray-crunching, 13,800 MIPS-capable, multicore Cortex-A9 Tegra 3. Moreover, the roadmap of production samples in Q4 of 2010 fits perfectly with NVIDIA's claim that Tegra 3 was "almost done" in September of that year. The ULP designation on this listing stands for Ultra Low Power in NVIDIA parlance, which would indicate an aggressively tuned power management system -- the only way we can envision a quad-core anything operating within a tablet. Fall 2011 is when we should know for sure.
Fujitsu K supercomputer will do 10 petaflops in 2012, eat Crays for breakfast
10's a nice round number, isn't it? Round, yes, but also wildly impressive when you put the word "petaflops" behind it as Fujitsu has done with its upcoming K supercomputer, which will be able to crunch through 10 quadrillion operations every second. Compare that to the current champ of processing farms, Cray's Jaguar, which can handle only (only!) 1.75 petaflops of workload and you'll know that we're talking about a seminal leap in performance. Japan's Riken Research Institute is the fortunate addressee on the crates of ultrafast SPARC64 VIIIfx processors that Fujitsu is now shipping out and the current plan is to have everything up and running by 2012. In total, there'll be 80,000 CPUs, each possessing 8 cores running at 2.2GHz, which will be housed within 800 racks. So yes, there'll be a machine somewhere on the Japanese isle with 640,000 processing cores at its disposal. Feeling safe?
Marvell unveils 1.5GHz triple-core application processor, all current smartphones look on in envy
Marvell's decided to whip out the "game changer" tag for its latest slice of silicon, but when you read the spec sheet that accompanies it, you might be willing to forgive it. Just this once. The new Armada 628 application processor delivers three cores, two of which crank along at 1.5GHz, and enough graphical prowess to churn 200 million triangles a second. You might remember we were once impressed by the Hummingbird's 90 million -- yeah, not so much anymore. The 628 is capable of 1080p 3D video and graphics (meaning it can sustain two simultaneous 1080p streams, one for each eye) and pledges to have an "ultra" low power profile: more than 10 hours of 1080p video or 140 hours of music playback are on offer. If that's not enough, it's also the first mobile SOC to include USB 3.0 support, adding yet another speed crown to its bulging resume. Now if it can also be SuperSpeedy in coming to market, that'd be just swell.
AMD's Bobcat and Bulldozer, 2011 flagship CPU cores, detailed today
One of these days AMD is gonna have to stop talking about its Atom-killing Bobcat and Xeon-ending Bulldozer cores and finally release them. But, until that happy moment arrives in 2011 (fingers crossed), we'll have to content ourselves with more presentation slides. First up, the Bobcat core is AMD's long overdue play for the netbook/ultrathin market. Pitched as having 90 percent of the performance of current-gen, K8-based mainstream chips, AMD's new mobility core will require "less than half the area and a fraction of the power" of its predecessors. That sounds like just the recipe to make the company relevant in laptop purchasing decisions again, while a touted ability for the core to run on less than one watt of power (by lowering operating frequencies and voltages, and therefore performance) could see it appear in even smaller form factors, such as MIDs. The Bobcat's now all set to become the centerpiece of the Ontario APU -- AMD's first Fusion chip, ahead of Llano -- which will be ramping up production late this year, in time for an early 2011 arrival. The Bulldozer also has a future in the Fusion line, but it's earliest role will be as a standalone CPU product for servers and high-end consumer markets. The crafty thing about its architecture is that every one Bulldozer module will be counted as two cores. This is because AMD has split its internal processing pipelines into two (while sharing as many internal components as possible), resulting in a sort of multicore-within-the-core arrangement. The way the company puts it, it's multithreading done right. Interlagos is the codename of the first Opteron chips to sport this new core, showing up at some point next year in a 16-core arrangement (that's 8 Bulldozers, if you're keeping score at home) and promising 50 percent better performance than the current Magny-Cours flagship. Big words, AMD. Now let's see you stick to a schedule for once.%Gallery-100088%
Intel plans to stuff more than 8 cores, extra speed into 2011 server chips
Yeah yeah, "more cores and faster speeds," you've heard it all before right? That'd be our reaction too if we weren't talking about the successor to the Nehalem-Ex, Intel's most gruesomely overpowered chip to date. Launched under the Xeon 7500 branding in March, it represents Intel's single biggest generational leap so far, and with its eight cores, sixteen threads, and 24MB of shared onboard cache, you could probably see why. Time waits for no CPU though, and Intel's planned 32nm Westmere-Ex successor will move things forward with an unspecified increase in both core count (speculated to be jumping up to 12) and operating frequencies, while keeping within the same power envelope. Given the current 2.26GHz default speed and 2.66GHz Turbo Boost option of the 7500, that means we're probably looking at a 2.4GHz to 2.5GHz 12-core, hyper-threaded processor, scheduled to land at some point next year. Time to make some apps that can use all that parallel processing power, nay?
Intel's 48-core processor destined for science, ships to universities soon
If you've been hankering to get your hands on that stamp-sized 48-core processor Intel introduced last year, you'd better brush off your doctorate -- the chipmaker says it will send samples of the CPU to researchers and academic institutions by the end of Q2. Clocked between 1.66GHz and 1.83GHz like Intel's Atom netbook chips, the 48 cores won't boost your framerates in Crysis -- rather, they're intended for linear algebra, fluid dynamics and server work -- but what we wouldn't give to try. Oh well -- suppose we'll just have to make do with puny 8- and 12-core chips for now.
