Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×

Submission + - Intel Ships 72-Core Knight's Landing Xeon Phi Processors (hothardware.com)

Submitted by MojoKid
MojoKid writes: It has been nearly two years since we first heard about Intel's next generation Xeon Phi "Knights Landing" processors, which are designed for High Performance Computing (HPC) applications. The processors are a big part of Intel's Scalable System Framework (SSF) and are built on using general-purpose x86 architecture and open standards. Today, Intel announced that its Xeon Phi processors are finally available to customers, nearly a year after the company's originally-quoted launch date. Intel Xeon Phi processors feature double-precision performance in excess of 3 teraflops along with 8+ teraflops of single-precision performance. All Xeon Phi processors incorporate 16GB of on-package MCDRAM memory, which Intel says is five times more power efficient as GDDR5 and offers 500GB/s of sustained memory bandwidth. MCDRAM can effectively be used as a high-speed cache or as a complimentary addition to the system DDR4 memory. Intel is targeting its Xeon Phi as a more competitive solution versus NVIDIA's dedicated Tesla GPU accelerators, citing up to a 5.2x performance advantage in visualization, up to a 2.7x in mathematical modeling, and up to a 5x increase in life sciences apps. The Xeon Phi is available in four basic configurations with 64 to 72 cores, and with processor frequencies ranging from 1.3GHz to 1.5GHz. All four support up to 384GB of DDR4 memory, but the base Xeon Phi 7210 is limited to the 2133MHz variety. Intel notes these are the company's first bootable host processors specifically designed for highly parallel workloads.

Submission + - More Details Surface On AMD 32-Core Server Chip Code Named Naples (hothardware.com)

Submitted by Anonymous Coward
An anonymous reader writes: AMD is hoping their next generation Zen processor architecture will be able to go toe-to-toe with the best that Intel has to offer and AMD is reportedly working on a high-end server variant of Zen as well, codenamed Naples. Naples would have a total of 32 cores, with a cluster of Zen cores sharing an 8MB pool of L3 cache. Total L3 shared cache is pegged at a stout 64MB and Naples will be capable of executing 64 threads while operating within a 180W power envelope. Naples reportedly will support eight independent memory channels and up to 128 PCIe Gen 3 lanes. In addition, a 16x10 GbE Ethernet controller is integrated into the chipset and Naples will use an SP3 LGA socket. The first server-based Zen processor could possibly squeak by for a late 2016 introduction, but odds are that we won't see widespread availability until 2017. At that time, you should expect Zen server processors in dual-, quad-, 16- and 32-core variants, with TDPs ranging from 35 watts to 180 watts. This is the second sighting of a 32-core AMD Zen variant. Earlier this year a CERN Engineer had details corroborating its existence in a presentation he was giving.

Submission + - Alienware Ships First Laptop With 13.3-Inch 2560X1440 OLED Display (hothardware.com)

Submitted by MojoKid
MojoKid writes: Dell's Alienware 13 gaming notebook has been popular among gamers that want a little more horsepower in a relatively light 4.5 pound 13-inch machine. However, over the past couple of years, Alienware hasn't changed-up the design much, until today that is. The company is officially making the OLED display equipped Alienware 13 available today, that they debuted back in January at CES. Initial testing and review impressions show that, as expected, that OLED display sure is nice. Although, the new Alienware 13 OLED is also representative of a full revamp, including a 6th generation Intel Skylake Core series processor and an NVMe Solid State Drive. The real kicker, however, is that Alienware's 13.3 QHD (2560X1440) OLED display offers great saturation and contrast with an extremely crisp 1ms pixel response time that delivers beautiful image quality, whether working in content creation, or in fast moving action while gaming. Viewing angles with the display are also superior to high-end IPS panels including Dell's own XPS 15 with its near-bezelless Infinity Edge panel.

Submission + - Intel Launches 10-Core Broadwell-E Core i7-6950X Extreme Edition Processor (hothardware.com)

Submitted by MojoKid
MojoKid writes: Intel has taken the wraps off their latest high-end desktop processor series today, known by the code name Broadwell-E. The flagship chip is the Core i7-6950X and it's a 10-Core / 20-thread CPU with 25MB of shared cache, a base frequency of 3GHz with boost to 3.5GHz and support for Intel's latest Turbo Boost Max 3.0 technology. Turbo Boost Max 3.0 helps boosts performance for both single and multi-threaded workloads by identifying the fastest core on the processor die at a particular moment and directing critical workloads to that core first. So, not only does the processor's frequency ramp up when needed, but workloads are also directed to the fastest possible core available. Turbo Boost Max 3.0 technology, along with the architectural advantages in Broadwell-E and the platform's support for faster memory, results in significant performance gains over Intel's previous gen Haswell-E chips, across the board. The Core i7-6950X's additional cores give it an obvious edge in multi-threaded workloads, but the processor also significantly outpaces the 8-core Core i7-5960X in single or lightly-threaded workloads as well. There are four new Broadwell-E processors due to arrive soon, including 6, 8, and 10 core chips ranging from $434 to a very pricey $1723 for the 10-core Core i7-6950X.

Submission + - NVIDIA GeForce GTX 1070 Launched: Titan X-Class Performance For Under $400 (hothardware.com) 1

Submitted by MojoKid
MojoKid writes: The powerful GeForce GTX 1080 got the lion's share of media attention at NVIDIA CEO Jen-Hsun Huang's official unveiling in Austin, Texas a couple of weeks back. But its little brother, the GeForce GTX 1070, was also mentioned along with its expected price point, which is far more attainable for most PC enthusiasts. The GeForce GTX 1070 would reportedly offer "Titan X class performance", which is particularly impressive, given the 1070's sub-$400 asking price. The GeForce GTX 1070 is based on the same GP104 GPU used on the GeForce GTX 1080, and as such, the GeForce GTX 1070 supports all of the new features that NVIDIA's Pascal architecture enables, including Simultaneous Multi-Projection, more advanced memory compression, and GPU Boost 3.0. However, the GPU employed in the GTX 1070 is scaled down somewhat with one of its Graphics Processing Clusters disabled, which results in a total of 1920 active CUDA cores, versus the 1080's 2560, and 120 texture units while the GTX 1080 has 160. The card was officially launched today and in the benchmarks, save for a couple of tests, the GeForce GTX 1070 outruns the GeForce GTX Titan X, and where it doesn't, the deltas separating the cards are miniscule. The GeForce GTX 1070 offers about 80 – 85% percent of the performance of the GTX 1080 and is generally faster than AMD's Radeon R9 Fury X and R9 Nano as well.

Submission + - GeForce GTX 1080 OEM Cards Now Shipping, GTX 1080 SLI Benchmarks Impressive (hothardware.com)

Submitted by MojoKid
MojoKid writes: NVIDIA launched their new flagship GeForce GTX 1080 a couple of weeks back and based on new 16nm FinFET manufacturing process technology, the beefy GPU with 8GB of GDDR5X memory on board, was claimed by NVIDA CEO Jen-Hsun Huang to offer dramatic increase in performance and efficiency. Unfortunately at the time only reference cards were available to the press but just this past Friday, NVIDIA add-in board partners like EVGA, ASUS, MSI and Gigabyte started unveiling their offerings with custom cooling and goosed-up clocks. EVGA's GeForce GTX 1080 Superclocked ACX 3.0 Edition was one of the first out of the gate and its GPU boosts to 1860MHz, with the memory humming along at 5005MHz. There were some tangible performance gains with this card versus NVIDIA's reference design but the SLI numbers with a pair of GeForce GTX 1080s are more impressive. 4K gaming at Ultra image quality settings is easily achievable. Cards from ASUS, Zotac, MSI and others have started to roll out as well.

Submission + - FBI Raids Dental Software Researcher Who Found Patient Records On Public Server

Submitted by blottsie
blottsie writes: Yet another security researcher is facing possible prosecution under the CFAA for accessing data on a publicly accessible server. The FBI on Tuesday raided Texas-based dental software security researcher Justin Shafer, who found the protected health records of 22,000 patients stored on an anonymous FTP.

“This is a troubling development. I hope the government doesn't think that accessing unsecured files on a public FTP server counts as an unauthorized access under the CFAA,” Orin Kerr, a George Washington University law professor and CFAA scholar told the Daily Dot. “If that turns out to be the government's theory—which we don't know yet, as we only have the warrant so far—it will be a significant overreach that raises the same issues as were briefed but not resolved in [Andrew 'weev' Auernheimer's] case. I'll be watching this closely.”

Submission + - Intel Skull Canyon NUC Tested: Skylake And Iris Pro Graphics In Tiny 1" Thick PC (hothardware.com)

Submitted by Deathspawner
Deathspawner writes: Intel first teased their forthcoming NUC (Next Unit of Computing) mini PC, codenamed Skull Canyon, back at CES in January. However, systems have only just started shipping earlier this month. Styled with a new, thinner (but longer) all black chassis and Intel's classic Skull-branded logo, this NUC is targeted squarely at enthusiasts. The Intel Skull Canyon NUC6i7KYK not only boasts a Skylake quad-core CPU that boosts to 3.5GHz, but also Intel's fastest Iris Pro Graphics 580 integrated graphics core with 128MB of on-chip eDRAM (embedded DRAM). With the Skylake platform, Skull Canyon also sports DDR4-2133MHz memory, up to two M.2 NVMe Solid State Drives, four USB 3 ports, a Thunderbolt 3 port and a built-in SD card reader. It's a fair amount of computing horsepower in a roughly 8-inch by 4-inch, by 1-inch thick form factor. NUCs have been fairly popular in the market due to their size and convenience, though they've historically not be designed for higher-end workloads and gaming. However, in testing, this performance-built NUC, with its DX12 compatible graphic core, showed the ability to run most current-gen game titles at up to 1080p resolutions with medium image quality. The barebones kit is a bit on the pricey side at $650 street, but it's definitely one of the faster tiny PCs on the market.

Submission + - Toshiba OCZ RD400 Series NVMe SSD Launched, Offers Over 2.6GB/Sec (hothardware.com)

Submitted by MojoKid
MojoKid writes: Toshiba OCZ launched a new series of NVMe solid state drives today, targeted squarely at the high-performance enthusiast computing market. The Toshiba OCZ RD400 NVMe SSD will be available in a variety of capacities and form factors. The drives will be sold as PCIe add-in cards or standalone gum-stick drives, with capacities ranging from 128GB on up to 1TB. Peak sequential bandwidth ranges from 2.2GB/s on the 128GB drive on up to 2.6GB/s on the larger capacities. Writes on the 128GB drive top out at 620MB/s, 1.15GB/s on the 256GB drive, 1600MB/s on the 512MB drive and 1550MB/s on the 1TB drive. RD400 drives feature Toshiba 15nm MLC NAND flash memory and a Toshiba-made controller. Unfortunately, no details were given on the controller, but it is a native PCIe 3.1 device that's NVM Express Revision 1.1b compliant. In testing the 512GB and 1TB models, IOMeter shows the drives finishing about in the middle of the pack versus some other high-end NVMe / M.2 SSDs. In all other tests, however, the Toshiba OCZ RD 400 series drives finished at or near the top of the charts, hitting their peak 2.6GB/sec read and 1.6GB/sec write throughput numbers.

Submission + - NVIDIA GeForce GTX 1080 Launched, Benchmarks Show Great Perf Per Watt (hothardware.com)

Submitted by MojoKid
MojoKid writes: A couple of weeks ago, NVIDIA CEO Jen Hsun-Huang unveiled the company's new high-end graphics cards, the GeForce GTX 1080 and GeForce GTX 1070. Huang claimed the cards represented a true generational leap in performance and efficiency, thanks to NVIDIA's new Pascal GPU architecture. Yesterday, the official launch of the GeForce GTX1080 took place, with full specs revealed and the benchmarks results are in. The GPU at the heart of the GeForce GTX 1080 Founder's Edition has a base clock of 1607MHz and a boost clock of 1733MHz, though boost clocks actually shot a little higher right out of the box in some game tests (one test card went up to 1823MHz occasionally, without overclocking). The GDDR5X memory on the card is clocked at 5GHz for an effective data rate of 10Gbps. At its reference clocks, the GTX 1080 offers up to 320GB/s of memory bandwidth and a peak texture fillrate of 257.1 GigaTexels/s, all within a 180 watt power envelope and the card only needs a single, 8-pin power feed. In the benchmarks GeForce GTX 1080 is roughly 20 -25 percent faster than a Titan X and 10 – 15 percent faster than a factory overclocked GeForce GTX 980 Ti. It was also significantly faster than AMD's Radeon R9 Fury X, in both DirectX 11 and Direct X 12 game titles.

Submission + - Tesla Model S Owner Claims Vehicle Went Rogue Causing An Accident By Itself (hothardware.com)

Submitted by MojoKid
MojoKid writes: A Tesla Model S owner is laying blame on the company and its product for an accident involving his pricey EV and a parked trailer. Jared Overton claims that on April 29th, he parked his Model S on the side of the road and ran some errands. He was parked behind a trailer at the time. A worker from the business he was visiting greeted him outside after which he went inside the establishment. Roughly five minutes later, he came out to find his Model S slammed into the trailer in front of it. How exactly did his Model S start up on its own and roll several feet down the road crashing into another parked vehicle? Good question. Overton was not happy about the accident, which smashed the car's windshield, so he decided to contact Tesla to tell them that his vehicle had "gone rogue." Tesla responded and cited owner error. According to the vehicle's logs, Overton had put the vehicle in Summon mode right before the exiting the vehicle, which is activated by "a double-press of the gear selector stalk button, shifting from Drive to Park and requesting Summon activation." Those are understandably deliberate actions that must be taken to invoke Summon, so either Overton didn't remember doing all of that (unlikely) or his Model S simply spazzed out (possible).

Submission + - NVIDIA Unveils GeForce GTX 1080 And GTX 1070, Faster Than Titan X For A Lot Less (hothardware.com)

Submitted by MojoKid
MojoKid writes: Tonight at an event in Austin, Texas, NVIDIA unveiled its next-generation Pascal-based GeForce graphics cards, known as the GeForce GTX 1080 and GeForce GTX 1070. NVIDIA's Pascal architecture is based on 16nm FinFET technology, similar to that of NVIDIA's high end data center Tesla P100 processing engine, but the GeForce cards are targeted at the consumer gaming market. NVIDIA's GP104 GPU at the heart of the new GeForce cards is comprised of some 8 billion transistors and features a 256-bit memory interface with 8GB of Micron GDDR5X graphics memory on the GeForce GTX 1080. The GTX 1070, however, employs standard GDDR5. The core clock speed of the GeForce GTX 1080 hit 2.1GHz at one point during the demonstration, though GTX 1070 clocks were not disclosed. NVIDIA CEO Jen-Hsun Huang claimed the new GeForce GTX 1080 is faster than a pair of GeForce GTX 980 cards in SLI and faster than the company's very expensive Titan X graphics card but at half the price. The new GeForce GTX 1080 will be offered in two versions, a standard card with an MSRP of $599 or a highly-overlcockable Founders Edition for $699. The standard GTX 1070 will arrive at $379, while a Founders Edition will be priced at $449. Availability for the GTX 1080 is slated for May 27th and the GTX 1070 for June 10.

Submission + - NVIDIA Unveils Pascal-Based GeForce GTX 1080 And GTX 1070 Graphics Cards (hothardware.com)

Submitted by MojoKid
MojoKid writes: Tonight at an event in Austin, Texas, NVIDIA unveiled its highly anticipated, next-generation Pascal-based GeForce graphics cards, known as the GeForce GTX 1080 and GeForce GTX 1070. NVIDIA's Pascal architecture is based on 16nm FinFET technology, similar to that of NVIDIA's high end data center Tesla P100 processing engine, but the GeForce cards are targeted at the consumer gaming market. NVIDIA's GP104 GPU at the heart of the new GeForce cards is comprised of some 8 billion transistors and features a 256-bit memory interface with 8GB of Micron GDDR5X graphics memory-- commonly referred to as G5X--on the GeForce GTX 1080. The GTX 1070, however, employs standard GDDR5. The core clock speed of the GeForce GTX 1080 hit 2.1GHz at one point during the demonstrations, though GTX 1070 clocks were not disclosed. NVIDIA CEO Jen-Hsun Huang claimed the new GeForce GTX 1080 is faster than a pair of GeForce GTX 980 cards in SLI and faster than the company's very expensive Titan X graphics card. The new GeForce GTX 1080 will be offered in two versions, a standard card with an MSRP of $599 or a highly-overlcockable Founders Edition for $699. The standard GTX 1070 will arrive at $379, while a Founders Edition will be priced at $449. Availability for the GTX 1080 is slated for May 27th and the GTX 1070 for June 10.

Submission + - AMD Radeon Technologies Group Crimson Software Driver Effort Starting To Pay Off (hothardware.com)

Submitted by MojoKid
MojoKid writes: Modern graphics drivers have evolved well beyond the device driver alone to user interfaces, libraries, tools, and applications that are all packaged up to what we now refer to as a "driver." The software has essentially morphed into a mini graphics OS. AMD recently retooled their graphics driver stack, dubbed Radeon Software Crimson Edition, and things appear to be moving in the right direction in terms of stability and performance optimizations. The software team at AMD's Radeon Technologies Group has delivered 3 WHQL certified drivers in the first quarter of 2016 (as compared to none in the same period of last year). AMD is also claiming that stability with the company's drivers have been improved as well, and over 100 fixes and customer end-user complaints have been resolved. The company is also claiming that recent releases have steadily improved general performance and VR quality. With current RTG software, graphics cards dating back to the Radeon R9 290X era meet or exceed Valve's and HTC's requirements for a good VR experience in the HTC Vive. AMD historically has struggled with graphics driver stability at times but currently seems to be in a good position thanks to the efforts of their recently formed Radeon Technologies Group.

Submission + - AMD Boosts FirePro W9100 With 32GB GDDR5 Memory To Stoke VR Creators (hothardware.com)

Submitted by MojoKid
MojoKid writes: With such a big emphasis on virtual reality and the development platforms backing the VR movement, it should come as no surprise that hardware OEMs are quickly embracing the need for more computational and rendering power with the need to work with larger datasets. To the end, AMD has announced that a new member is joining its FirePro W9100 family, a card that steps up to the plate with 32GB of GDDR5 frame buffer memory. AMD says that the FirePro W9100 32GB, which doubles up the amount of memory that was previously available, was specifically designed for content creators. This release also allows AMD to once again leapfrog NVIDIA, which last month introduced the Quadro M6000 with 24GB of RAM. This new FirePro is a professional GPU based on AMD's Hawaii architecture with 2816 stream processors, 176 texture units, a 930GHz core clock, and a 275W TDP.

Slashdot Top Deals

Function reject.

Close