October 13th, 2015 | Edited by Zoran Stosic | hardware
The new storage technology aims to be 1,000 times faster than flash, and will compete with similar tech from Intel and Micron.HP and SanDisk are plotting new storage technology that could be 1,000 times faster than flash memory, though they’ll face some competition along the way.
Details on the new technology are scarce, but the goal is to create a “universal memory” that serves as both long-term storage and RAM, the Wall Street Journal reports. The goal is to commercialize this technology some time between 2018 and 2020.
Why this matters: Today’s computers offer RAM and storage separately, because the former is much more expensive and purges its data when the machine powers down. Programs and files are stored on flash memory, but during use they’ll load some data into RAM for faster short-term access. A single type of memory for both short- and long-term storage could boost a PC’s performance dramatically.
Intel, Micron, and a manufacturing shift
HP and SanDisk aren’t the only ones with eyes on next-generation memory. Earlier this year, Intel and Micron announced a partnership on a new type of storage, touting the same “1,000 times” performance improvements. Originally dubbed 3D XPoint, the technology will commercially be known as Intel Optane, and is scheduled to hit the market in 2016—well ahead of HP and SanDisk.
The two companies aren’t just looking at servers, which are often the first stop for new storage technology. Intel aims to have Optane in laptops next year as well, and is introducing storage controllers, interfaces, and interconnects to help make it happen. HP and SanDisk are also collaborating on computer systems that use their storage technology.
The Wall Street Journal’s report sheds some light on why the companies may be entering such fierce competition in the first place: Neither Intel/Micron nor HP/SanDisk plan to license their technology, which means you won’t see it on products from Seagate, Samsung or other storage vendors.
The licensing model typically ends with vendors racing to the bottom on pricing, but apparently that won’t be the case with this technology. In other words, expect to pay a premium for those faster speeds, unless (or until) another entity comes up with something similar that all vendors can use.
October 10th, 2015 | Edited by Zoran Stosic | hardware
Certified cables will carry labels assuring the consumer that they’ve been tested to meet the HDMI 2.0 spec.
One HDMI cable is as good as another, right? Wrong. The old saying “an HDMI cable will either work or it won’t, because digital is all or nothing” is a myth. A poor-quality HDMI cable can deliver a degraded signal, resulting in a snowy picture or worse. A crappy HDMI cable, especially a long one, can also cause problems that you can’t see: radiating enough electromagnetic interference (EMI) to cause problems on your Wi-Fi network.
Having said that, bad HDMI cables are pretty hard to find, at least when asked to carry 1080p video just a few feet. It can be a different story when you enter the realm of 60-frames-per-second 4K video with high dynamic range, high-resolution multi-channel audio, and perhaps even ethernet. According to the standard, an HDMI 2.0 cable should be capable of delivering “ultra-reliable performance at the full 18Gbps bandwidth.”
To that end, HDMI Licensing LLC—the group responsible for developing and maintaining the HDMI standard that’s used on nearly every TV, PC, monitor, projector, Blu-ray player, A/V receiver, and media streamer shipping today—has announced a new cable certification program. Instead of hoping for the best—or paying ridiculous prices for cables made by companies with marketing budgets that dwarf what they spend on manufacturing—you could just shop for HDMI cables labeled “HDMI Premium Certified Cable.”
October 8th, 2015 | Edited by Zoran Stosic | hardware
The SanDisk Industrial XT iNAND embedded flash drive was introduced on Oct. 5, 2015.The SanDisk Industrial components are designed to work harder, last longer and withstand harsh conditions.
Consumers rely on flash memory cards in phones, tablets and cameras every day. But put those same cards in security cams, cellular base stations or the electrical grid and you’ll have a problem.
Industrial devices need flash that can work harder and withstand more extreme temperatures than consumer gear, and they’ll be operating out in the field years after a typical phone or camera card has been replaced. So SanDisk is introducing a line of components built for the Internet of Things.
IoT is expected to put thousands of sensors, meters, robots and machines into the field with growing needs to process and store data.
The SanDisk Industrial line includes cards for the familiar SD, microSD and eMMC (embedded MultiMediaCard) standards, but built to tougher specifications.
For example, the SanDisk Industrial XT SD Cards and XT iNAND embedded flash drives announced Monday are rated to work in temperatures as low as -40 degrees Celsius (-40 Fahrenheit), compared with -25 Celsius for a typical consumer SD card.
The industrial cards can also write more data before they have to be replaced: as much as 128TB, far more than is typical for a consumer-grade part, said Martin Booth, director of SanDisk Industrial and SanDisk Automotive. This kind of endurance is what’s needed in IoT devices like remote video cameras that will capture video around the clock for as long as five years, he said. Otherwise they would have to be replaced more frequently, a costly proposition if the owner needs to send out a truck and a technician.
SanDisk achieved the gains by holding back on things that make flash for a phone or camera attractive to consumers, like maximum write speeds and densely packed memory cells that drive down cost per bit by half every year or two, Booth said. The heavy-duty flash is made with an older 19-nanometer manufacturing process, not the 15nm process that’s the cutting edge today at SanDisk.
Another feature, Enhanced Power Immunity, will help prevent data loss in case of power failure. It uses special firmware for recovering data if the power is cut off, something ordinary flash cards may not be able to do if, for example, the user pulls a card out of a PC while it’s still transferring data.
The new parts range in size from 4GB or 8GB up to 64GB and will cost more than comparable consumer-grade products, but less than twice as much, Booth said. They are available now and will only be sold to equipment manufacturers.
October 6th, 2015 | Edited by Zoran Stosic | hardware
Microsoft’s digital assistant arrives on the Xbox One in 2016, but Xbox One Experience Preview testers will get to try it later this year.
On Friday evening, Microsoft’s Larry Hryb—better known as Major Nelson—took to his blog to make a couple announcements. First, a new update to the Xbox Beta app for Windows 10 is on its way. We’ll get to that in a moment, but the big news is that Cortana, Microsoft’s digital assistant technology, will make its way to the Xbox in 2016.
According to Major Nelson, beta testers will get a chance to try Cortana on their Xbox “later this year” when Microsoft releases it through the Xbox One experience preview program. This will allow the company to “ ensure that the experience is tuned for gamers” ahead of its public launch next year. Major Nelson didn’t provide any further details as to when in 2016 you can expect the official version of Cortana for Xbox, however.
The story behind the story: Cortana debuted in Windows Phone 8.1, and Microsoft also built it into Windows 10, so adding it to the Xbox One seems to be the most logical next step. We still don’t know what Cortana for Xbox will be able to do, but it will likely build off of the Xbox One’s existing support for spoken commands via the Kinect controller.
Xbox updates coming for Windows 10
In the meantime, while you’re waiting for Cortana for Xbox One, you may as well check out the latest Xbox Beta app for Windows 10. This new update includes support for a new group text chat feature, the ability to reply directly to notifications, and some new social features, such as an easier way to share in-game achievements and game clips, among other things. The new Xbox Beta app for Windows 10 is available for download now; a general public release date for these features has yet to be announced.
September 24th, 2015 | Edited by Zoran Stosic | hardware
Intel corrects a report about the demise of its socketed Broadwell CPU, but how long it’ll live remains anyone’s guess.
Intel’s Broadwell-C desktop processor is alive and well—for now, at least.
A report on Thursday by ITWorld claimed that Intel was discontinuing the socketed Broadwell CPU after just one month on the market. But Intel quickly corrected that story,telling AnandTech that it will continue to manufacture and sell Broadwell-C. (Disclosure: Both PCWorld and ITWorld are owned by International Data Group.)
Broadwell-C is a unique chip in Intel’s lineup for a couple of reasons: First, it’s the only Broadwell chip for desktop tower PCs, and it arrived much later than usual in Intel’s product cycle. (Intel originally planned to skip socketed Broadwell entirely, a move the company now regrets.)
More importantly, Broadwell-C is Intel’s only socketed desktop chip with 128MB of embedded DRAM. This on-package memory allows for impressive gaming performance with just integrated graphics, and also provides a nice boost when paired with discrete graphics. In the past, Intel has reserved this configuration for laptops and mini-desktops where the CPU is soldered to the motherboard.
The reported demise of Broadwell-C was apparently just a mix-up, but made waves in enthusiast tech forums such as Slashdot and various subreddits. ITWorld has since corrected and amended its story, noting that it is in fact the next-generation Skylake-C that has been cancelled. Apparently Intel just doesn’t see enough market demand for that embedded DRAM setup. ITWorld also speculates that increased costs and lower yields could be to blame.
Why this matters: For many users, this may all be a moot point with the arrival ofSkylake CPUs, but it could also be an opportunity to pick up a decent last-generation CPU for less cash as long as Intel keeps making them. While opting for a cheaper processor and a low-end graphics card probably makes the most sense in traditional PC setups, Broadwell-C could be a compelling option if you want to perform entry-level gaming or other graphics-heavy tasks in a rig with an ultra-small form factor case where discrete graphics can’t fit well.