intel – Page 2 – Tech Today w/ Ken May

Intel to build discrete GPUs, hires AMD’s top graphics guy to do it

(credit: Intel) One might have thought that with Monday’s announcement that Intel is going to produce processors with embedded AMD GPUs that the two processor companies were on good terms. That’s looking a little less likely now. On Tuesday, AMD announced that Raja Koduri, its chief GPU architect, was to leave the company. Where was he going? That question was resolved on Wednesday : Intel. And what’s he going to do at Intel? He’s going to be the senior vice president of a new group—Core and Visual Computing—that will expand Intel’s graphics reach both into the low-end, with integrated graphics reaching into the Internet-of-Things space, and more excitingly, at the high end, with discrete GPUs. Koduri led AMD’s Radeon Technologies Group, responsible for both AMD’s discrete and integrated GPUs. Before that, he was director of graphics technology at Apple. Intel has dabbled with discrete GPUs before; its 740 GPU, released in 1998, was a standalone part using the then new AGP port. A second attempt to build a standalone GPU was the Larrabee project , but that never shipped as a GPU. In 2009 Larrabee was repositioned with Intel deciding to make it a massively multicore accelerator—the predecessor to the current Xeon Phi chips—rather than a graphics processor. Read 6 remaining paragraphs | Comments

See the original article here:
Intel to build discrete GPUs, hires AMD’s top graphics guy to do it

Intel squeezed an AMD graphics chip, RAM and CPU into one module

Intel may have unveiled its latest Core CPUs for mainstream laptops , but the company has something more advanced up its sleeves for what it calls its “enthusiast” customers. The new chip will be part of the 8th-generation Core H series of processors, and comes with discrete-level graphics cards built in, as well as its own RAM. Having all this built into the processor frees up space for other components inside a laptop, so device manufacturers can squeeze in things like larger batteries or more optimal fan designs. Intel is not sharing performance details for the new CPUs yet, but it’s promising power that will be good enough for gamers or content creators who often run taxing programs like Adobe Photoshop and Lightroom. Specifically, the new processor integrates a “semi-custom” AMD graphics chip and the second generation of Intel’s “High Bandwidth Memory (HBM2)”, which is comparable to GDDR 5 in a traditional laptop. The three typically distinct components are able to coexist on one chip because of Intel’s Embedded Multi-Die Interconnect Bridge (EMIB), which “allows heterogeneous silicon to quickly pass information in extremely close proximity.” The company also came up with a power-sharing framework that lets the GPU manage each component’s temperature, performance and energy use. This infrastructure should free up about three square inches of board space that could either be used for other components as described above, or make for thinner laptops altogether. The idea is that powerful laptops for gamers no longer have to be chunky beasts. The new Core H processor is the first consumer product to use EMIB, and will be released in the first quarter of 2018, and many laptop makers are expected to offer products powered by the chip. This is a pretty significant development that not only benefits the enthusiast audience, but could also have trickle down effects that could improve mainstream laptops (and even other devices) in the future. Source: Intel

View the original here:
Intel squeezed an AMD graphics chip, RAM and CPU into one module

Samsung leapfrogs Intel again with 8-nanometer chips

Samsung has qualified its 8-nanometer chip-making process for production three months ahead of schedule. It’s the same “low power plus” (LPP) process used for its current 10-nanometer silicon , not the next-gen extreme ultraviolet (EUV) lithography for its future 7-nanometer tech. That’ll yield chips that are ten percent more energy efficient and ten percent smaller than the 10-nanometer ones it’s making right now. At the same time, since the 8-nanometer chips use the same process, Samsung will be able to “rapidly ramp up, ” it said . Samsung said that the new process will be ideal for “mobile, cryptocurrency and network/server” applications. It notably worked again with Qualcomm, its 10-nanometer chip launch customer, to perfect the new tech. Rumors in Korea had it that Qualcomm would switch its 7-nanometer production to TMSC, which is reportedly slightly ahead of Samsung in developing that tech. However, Samsung confirmed with ZDNet that Qualcomm will be using its 8-nanometer process, without providing any specific details. Given that information, it seems likely that Qualcomm will build its next-gen Snapdragon chips with Samsung, using the tried-and-true LPP process instead of bleeding-edge 7-nanometer tech, which necessitates a switch to extreme ultraviolet lithography. By that time, Samsung should have its own 7-nanometer EUV process up to speed, with 6-nanometer chips set to follow after that. Anyway, Samsung Mobile is probably Qualcomm’s biggest customer with its Galaxy S8 and Note 8 phones, so it would have been pretty awkward to split off to another foundry. Though they don’t compete much in the same markets, the news puts Intel even further behind Samsung, at least in terms of chip trace sizes. Intel has yet to release any 10-nanometer chips, though it has said that when it does ( in 2018 or 2019 ), it will be ” generations ahead ” of Samsung thanks to better feature density. By then, however, Samsung might have closed that gap by being two or three actual generations ahead of Intel in terms of lithography. Samsung is expected to reveal its roadmap for 8- and 7-nanometer chips later today. Source: Samsung

Link:
Samsung leapfrogs Intel again with 8-nanometer chips

What is Intel Coffee Lake?

Intel’s 8th-generation “Coffee Lake” CPUs are now on the market. These chips come with a modest bump in CPU frequency, but the big news is that Intel is finally adding 6-core processors to its mainstream i7 and i5 lines. More cores means these chips will perform better at tasks that benefit from multithreading, such as content creation and data processing, and the increase in frequency and cores will give a boost to gaming frame rates. Intel used to release chips on a “tick-tock” cycle that saw every release alternate either a new design, or a new manufacturing process – called a “node.” A new process node, like moving from 45nm to 22nm, means smaller transistors and a faster or more power-efficient chip. But manufacturing challenges made tick-tock falter a few years ago, and now new releases are much harder to predict. “Coffee Lake” is the fourth chip Intel has released at 14nm, and the third based on the “Skylake” design from 2015. New designs and nodes are coming, but we’ll probably have to wait until they arrive in 2018 to see a big jump in performance.

View article:
What is Intel Coffee Lake?

Intel’s 8th-gen desktop CPUs boost gaming and streaming speeds

Intel has launched its 8th-generation mainstream desktop chips, calling the flagship $359 Core i7-8700K its “best gaming desktop processor ever.” The six-core, 12-thread 3.7GHz chipboosts Gears of War 4 frame rates up to 25 percent compared to 7th-gen desktop chips . It can also do 4K video editing up to 32 percent faster. The best gains are with multitasking, as Intel says gaming, streaming and recording with Player Unknown: Battlegrounds will be 45 percent fast than before. What’s more, it can be overclocked to 4.7 GHz using Intel’s Turbo Boost 2.0, and 5GHz and higher speeds are easily achieved with air or basic liquid cooling. The speed gains are even more impressive if your system is over three years old, but probably won’t push many folks with 7th-gen Intel chips into upgrading. For multimedia, it’s worth noting that Intel has implemented some new 4K tech, which will make for smoother streaming of 10-bit, HDR Ultra HD video now on services services like Netflix and many new 4K TVs. The 8th-gen Core i5 and Core i3 chips in the lineup might be more interesting for consumers. The Core i3 chips start at $117 and have quad-cores and base clock speeds up to 4 GHz, while the Core i5 models, starting at $182, are six-core units. Neither are multi-threaded, but four physical cores and four threads is a lot better than two cores and four threads, which is what the i3 lineup had before. If you have Intel’s 7th-gen Core i7 chip and want to upgrade to get every ounce of gaming or graphics speed, it’s not as simple as a chip swap. The new CPUs will require Intel’s Z370 chipset-based motherboards, so they’re not at all compatible with whatever model you have right now. That chipset has a few advantages over the last gen, like improved power delivery for 6-core chips and better support for DDR4-2666 memory, but nothing dramatic. Some pundits have noted that Intel could have made them compatible with older motherboards, but elected not to . Intel is also touting its Optane storage for gamers, but as we’ve mentioned , this won’t help you much if you already have an SSD, and not at all if you have M.2 PCIe-based storage. The mainstream 8th-gen desktop chips are a bit less interesting than the 8th-gen laptop CPUs, which offered more performance than expected , Meanwhile, Intel recently unveiled the 7th-gen X-series i9 chips, which rocked up to 18 cores , dramatically boosting performance over the top-spec 10-core i7-6950X previous-gen models. The new eighth-gen desktop chips will arrive starting on October 5th, and as mentioned, the flagship Intel Core i7-8700K will cost you $359. Source: Intel

Azure Confidential Computing will keep data secret, even from Microsoft

Enlarge / The Trusted Execution Environment means that, even if the application and operating system are compromised, the green code and data can’t be accessed. (credit: Microsoft ) Microsoft announced today a new feature coming to its Azure cloud platform named “Confidential Compute.” The feature will allow applications running on Azure to keep data encrypted not only when it’s at rest (in storage) or in transit (over a network) but when it’s being computed on in-memory. This ability to encrypt data when it’s in-use means that it can be kept secure even from Microsoft’s administrators, government warrants, and hackers. Confidential Computing will have two modes: one is built on virtual machines, while the other uses the SGX (“Software Guard Extensions”) feature found in Intel’s recently introduced Skylake-SP Xeon processors. Both modes will allow applications to ringfence certain parts of their code and data so that they operate in a “trusted execution environment” (TEE). Code and data that are inside a TEE cannot be inspected from outside the TEE. The virtual machine mode uses the Virtual Secure Mode (VSM) functionality of Hyper-V that was introduced in Windows 10 and Windows Server 2016. With VSM, most parts of an application will run in a regular virtual machine atop a regular operating system. The protected, TEE parts will run in a separate virtual machine containing only a basic stub operating system (enough that it can communicate with the regular VM) and only those parts of the application code that need to handle the sensitive data. Read 4 remaining paragraphs | Comments

Read the original:
Azure Confidential Computing will keep data secret, even from Microsoft

Alleged Intel i7-8700K Coffee Lake benchmarks leak online

Enlarge Alleged benchmarks for Intel’s as yet unannounced Core i7-8700K have leaked online courtesy of a tech YouTuber and an unguarded HP Omen PC at DreamHack 2017. The i7-8700K—the rumoured flagship six-core, 12-thread processor of Intel’s upcoming eighth generation Coffee Lake desktop CPUs—was put through popular benchmarking program Cinebench R15, scoring a cool 1230 points in a multithreaded test. That’s a significant leap over the previous-generation Intel Core i7-7700K (4C/8T), which typically posts a score around 950 points. That score also puts the i7-8700K neck and neck with AMD’s 6C/12T Ryzen 1600X , which we scored at 1234 points, but behind AMD’s 8C/12T Ryzen 7 1700 with its score of 1422. Meanwhile, AMD’s flagship Ryzen 7 1800X is significantly faster with a score of 1616 points. YouTuber Karl Morin was also able to run a single-threaded benchmark, which shows the i7-8700K posting a score of 196. That would make it fastest single-threaded chip around, beating the pervious generation i7-7700K’s score of around 185 points, although it’s still far from a compelling improvement in instructions per clock. Read 6 remaining paragraphs | Comments

More here:
Alleged Intel i7-8700K Coffee Lake benchmarks leak online

Linux Kernel 4.13 Officially Released

prisoninmate writes: As expected, the Linux 4.13 kernel series was made official this past weekend by none other than its creator, Linus Torvalds, which urges all Linux users to start migrating to this version as soon as possible. Work on Linux kernel 4.13 started in mid-July with the first Release Candidate (RC) milestone, which already gave us a glimpse of the new features coming to this major kernel branch. There are, of course, numerous improvements and support for new hardware through updated drivers and core components. Highlights of Linux kernel 4.13 include Intel’s Cannon Lake and Coffee Lake CPUs, support for non-blocking buffered I/O operations to improve asynchronous I/O support, support for “lifetime hints” in the block layers and the virtual filesystem, AppArmor enhancements, and better power management. There’s also AMD Raven Ridge support implemented in the AMDGPU graphics driver, which received numerous improvements, support for five-level page tables was added in the s390 architecture, and the structure randomization plugin was added as part of the build system. Read more of this story at Slashdot.

Continue reading here:
Linux Kernel 4.13 Officially Released

Intel’s eighth-gen CPUs will be more powerful than we thought

Earlier this year, we didn’t expect much from Intel’s upcoming eighth-generation Core processors. But at Computex in May, the company surprised us all by revealing they’ll be 30 percent faster than last year’s chips. That alone would have been notable, but now Intel is making an even bolder claim: They’re actually 40 percent speedier. At least, that’s the case for one of its laptop chips when running through the Sysmark benchmark. And when it comes to five-year-old computers — the kind Intel expects the majority of consumers to upgrade from — they’re twice as fast at productivity tasks. So how did Intel manage this feat? For the first time, it managed to cram four CPU cores into its U-series chips, which have typically been intended for ultraportable laptops and hybrid devices. To go quad-core with the company’s previous chips, you’d have to step up to its H-series CPUs, which are designed for more performance-focused notebooks. Additionally, Intel managed to squeeze another 500MHz of Turbo Boost performance into the chips, allowing them to reach speeds up to 4.2GHz. Whereas last year’s seventh-generation CPUs were focused on improving 4K performance, the upcoming offerings will feature far more raw performance. The eighth-generation Core family for ultraportables ranges from the i5-8250U, with speeds between 1.6GHz and 3.4GHz, and the i7-8650U, clocking from 1.9GHz to 4.2GHz. They’re all quad-core chips, but as usual, you can expect better performance from the i7 lineup. As for other notable stats, Intel claims the 8250U can encode a 4K video 14.7 times faster than a five-year-old PC running a comparable chip. In real-world terms, the eighth-gen chip converted the 4K clip in three minutes, compared to 45 minutes with its third-gen sibling. Unsurprisingly, then, the older your computer is, the bigger performance jump you’ll notice. Despite having more power under the hood, Intel says the new CPUs won’t be a step backwards when it comes to battery life. Unfortunately, there’s nothing really pushing battery life forward either. Its current benchmarks indicate that laptops running these chips should feature around 10 hours of battery life when playing 4K video. That’s pretty much the same as before. To be fair, though, last year’s chips also reclaimed some battery life with more efficient 4K performance. Intel is basically following through with its typical game plan: Release a new batch of chips for ultraportables, and then ramp things up on the desktop end. With quad-core performance, we might finally be able to see ultraportable notebooks that can go toe to toe with last year’s beefier machines. And the company continues to look ahead as well. During a private briefing, I got a chance to see a few device prototypes that looked nothing like PCs we have today. I can’t say much about them, but overall, they made it clear that Intel isn’t just planning to coast on traditional PCs forever. You can expect the first batch of laptops with eighth-generation Intel CPUs in September. Desktop models will follow later in the fall. Looking ahead even further into next year, the company expects to release its first 10-nanometer chips as part of this generation as well.

More:
Intel’s eighth-gen CPUs will be more powerful than we thought

Intel Unveils One-Petabyte Storage Servers For Data Centers

Slashdot reader #9, 219 Guy Smiley shared this report on a new breed of high-density flash storage. The Inquirer reports: Intel has unveiled a brand new form factor for solid state disc drives (SSDs)… Intel Optane’s new “ruler” format will allow up to a petabyte of storage on a single 1U server rack… By using 3D-NAND, the ruler crams in even more data and will provide more stability with less chance of catastrophic failure with data loss. The company has promised that the Ruler will have more bandwidth, input/output operations per second and lower latency than SAS… As part of the announcement, Intel also announced a range of “hard drive replacement” SSDs — the S4500 and S4600 0 which are said to have the highest density 32-layer 3D NAND on the market, and are specifically aimed at data centres that want to move to solid state simply and if necessary, in stages. Read more of this story at Slashdot.

Read the original post:
Intel Unveils One-Petabyte Storage Servers For Data Centers