News Posts matching "GPU"

Return to Keyword Browsing

NVIDIA Pioneers New Standard for HPC With Tesla GPUs Built on Kepler Architecture

NVIDIA today unveiled a new family of Tesla GPUs based on the revolutionary NVIDIA Kepler GPU computing architecture, which makes GPU-accelerated computing easier and more accessible for a broader range of high performance computing (HPC) scientific and technical applications.

The new NVIDIA Tesla K10 and K20 GPUs are computing accelerators built to handle the most complex HPC problems in the world. Designed with an intense focus on high performance and extreme power efficiency, Kepler is three times as efficient as its predecessor, the NVIDIA Fermi architecture, which itself established a new standard for parallel computing when introduced two years ago.

MSI Z77 Series Mainboards Lead the Global March Towards 7 GHz

The Z77 series mainboard released by MSI is equipped with Military Class III components, and has met 7 rigorous testing standards conforming to MIL-STD-810G via third-party laboratories, certifying it for the best quality and highest stability! Equipped with a Hybrid Digital Power design, it allows voltage options to be more precisely adjusted, providing a more stable power source that can achieve a higher overclocking result.

Equipped with Intel's 3rd Gen Core processor, its overclocking potential is astounding: using the Z77A-GD65 mainboard, the i7-3770K processor can successfully be overclocked from the default speed of 3.5GHz up to 7GHz, with a 200% overclocking margin! In addition, MSI's Z77 series mainboard exclusively integrates the UEFI BIOS and Windows application into one unified interface: Click BIOS II. MSI Z77 series mainboard leads global users towards an all-new overclocking realm with unlimited possibilities.

NVIDIA Releases GK107-based GT 630, GT 645 and 3 GT 640 Variants to OEMs

NVIDIA released its first desktop graphics cards based on its 28 nm GK107 silicon, in as many as three models, the GeForce GT 630, GT 640, GT 640 (second variant). Apart from these, NVIDIA appears to have launched two more models, one labeled GT 640, and another GT 645, which are based on previous generation GF116 and GF114, respectively. These new SKUs are available only to OEM partners, who will implement them in their new desktop PC designs.

Of the five new SKUs, the GT 630 and GT 640 (28 nm) are most interesting, as they are based on the GK107 silicon, with as many as 384 CUDA cores, and TDP as low as 50W. The GT 630 is a single-slot, low-profile graphics card with 384 CUDA cores, and a 128-bit wide GDDR5 memory interface, 875 MHz GPU clock speed, and 891 MHz memory clock speed. The GT 640 (28 nm) features different clock speeds. The specifications may seem confusing to the end user, but then they really don't concern people who buy retail-channel graphics cards. NVIDIA and AMD design OEM-specific SKUs completely à la carte, to OEM partners' requirements.

NVIDIA GeForce GTX 690 Slated for April 28, Launches at GeForce LAN, in Shanghai

NVIDIA is poised to launch its high-end GeForce GTX 690 dual-GPU graphics card on the 28th of April, according to a VR-Zone report. NVIDIA has chosen the 2012 NVIDIA GeForce LAN / NVIDIA Gaming Festival (NGF) held in Shanghai, China, as its launchpad. NVIDIA even put up a countdown timer on GeForce.com that counts down to the event. GeForce GTX 690 combines two 28 nm GK104 graphics processors for an SLI-on-a-stick solution. It is gunning for absolute performance leadership, enabling gamers to throw any game, at any resolution at PCs equipped with it.

PowerColor Intros Radeon HD 7750 Go! Green Passive-Cooled, Energy-Efficient VGA

PowerColor launched its second Radeon HD 7750 graphics card, a Go! Green variant. Models of this series tend to be energy-efficient (with slightly lowered voltages, reducing power draw), while maintaining reference clock speeds. Another striking feature of Go! Green variants is their silent passive-heatsink that cools the GPU and VRM. To maintain high temperature tolerance, and energy-efficiency, PowerColor tends to employ high-grade voltage regulation components on their Go! Green cards. In case of this card, it is said to utilize PowerPAK SO-8 MOSFETs, shielded chokes, and conductive-polymer capacitors.

The card sticks to reference clock speeds of 800 MHz (core) and 1125 MHz / 4.50 GHz effective (memory), packs 512 Graphics CoreNext stream processors, and 1 GB of GDDR5 memory over a 128-bit wide memory interface. The cooler is similar to the one featured on previous-generation Go! Green cards from PowerColor, it consists of a dense aluminum fin heatsink, to which heat is conveyed by copper heat pipes that draw head directly off the core. Heat dissipation happens passively. Expect the card to charge a tiny premium over other Radeon HD 7750 cards in the market.

NVIDIA Approaching Other Foundries than TSMC for 28 nm Production

NVIDIA, along with Qualcomm, is reportedly in talks with foundries other than TSMC, for manufacturing of its new 28 nm chips. Despite the fact that TSMC is ramping up its 28 nm capacity at a breakneck pace, NVIDIA is seeing a shortage of production that could affect its competitiveness. An interesting revelation here is that NVIDIA has begun sampling its GPUs on Samsung's 28 nanometer fab process. Samsung uses this process for contract-manufacturing of ARM application processors. Other foundries with proven 28 nm manufacturing capability include UMC.

TSMC 28 nm Capacity Ramp-Up Faster Than Older Processes

With launches of new-generation GPUs by NVIDIA and AMD, and new ARM application processor designs by various industry players, TSMC is under pressure to ramp up its production capacity for its new 28 nanometer note. DigiTimes research suggests that this ramp-up is going at a faster rate than older processes such as 40 nm and 65 nm nodes (when those were new). Digitimes Research analyst Nobunaga Chai claims that the 28 nm node started generating revenues in Q4, 2011, and sales ratio reached 5% in the following quarter. It is anticipated to see TSMC significantly ramp up its 28nm production capacity later in 2012, Chai believes.

HIS Launches HIS 7870 Fan 2 GB

Hightech Information System (HIS) today announced the launch of HIS 7870 Fan 2 GB. The card offers outstanding graphics quality and excellent cooling technology for enthusiastic users. With core and memory clocks at 1000 MHz and 4800 MHz, the card ensures the highest graphics enjoyment during hardcore gaming.

Equipped with HIS outstanding cooler, the card is 13 °C cooler than the reference cooler and provides a less than 28 dB quiet environment for daily tasks and movies.

The card is made of HIS high quality component. It has better OC capability with overclocking equipment and longer life span with safety design.

NVIDIA and Partners Transform Entire Content Creation Workflow at 2012 NAB Show

Content creators ranging from 2012 Oscar winners Industrial Light & Magic, a division of Lucasfilm Ltd. (Rango) and Pixomondo (Hugo) to up-and-coming independents such as Bandito Brothers (Act of Valor, Waiting for Lightning) and smaller creative services shops such as Dawnrunner Productions and Digital Spatula are increasing the speed, quality and productivity of their workflows through the adoption of GPU computing featuring NVIDIA Quadro and Tesla GPUs.

Visual effects studios, post-production houses, advertising agencies and film studios are under constant pressure to more rapidly deliver better content under tight production budgets. They use NVIDIA GPU technology to transform their content creation workflows, using parallel computing on tasks once reserved for supercomputers or racks of costly CPUs.

NVIDIA GK106 GPU Detailed

Following the launch of a couple of more GK104-based SKUs, namely GeForce GTX 670 Ti, and GeForce GTX 670, some time in May; NVIDIA will launch its third (after GK104 and GK107) Kepler architecture-based silicon, codenamed GK106. GK106 will make up sub-$200 SKUs, and succeeds the GF116, on which SKUs such as the GeForce GTX 550 Ti, are based. The GK106 will make up at least one known desktop SKU, called GeForce GTX 660.

Its specifications are listed below.

ZOTAC Unveils ZT-Z77-U1D SuperOverclock High-End Motherboard

ZOTAC unveiled a monster LGA1155 motherboard aimed at professional overclockers, and based on the new Intel Z77 Express chipset, the ZT-Z77-U1D. Pictured below, the designers' focus was evidently on giving the motherboard a very strong VRM, apart from just enough expansion and connectivity features for 2-way multi-GPU setups. To begin with, the LGA1155 socket is powered by a 27-phase VRM, which consists of AIO ferrite-core solid-state chokes, DrMOS, tantalum capacitors, and a super-ML multiphase capacitor to condition power. The VRM is controlled by a VRD12-compliant controller.

The LGA1155 socket is wired to four DDR3 DIMM slots, supporting dual-channel DDR3-2133+ MHz memory, and two PCI-Express 3.0 x16 slots (x8/x8, when both are populated). Other expansion slots include four PCI-Express 2.0 x1, wired to the Z77 PCH. All six SATA ports from the PCH are assigned as internal ports, that's two SATA 6 Gb/s, and four SATA 3 Gb/s. Display connectivity includes DVI, HDMI, and DisplayPort. Other connectivity includes 8+2 channel HD audio, 802.11 b/g/n, Bluetooth 3.0, gigabit Ethernet, six USB 3.0 ports (two on the rear panel, four via headers), and a number of USB 2.0 ports.

Orbis Implements Multi-GPU, Too

Sony's next-generation PlayStation, reportedly codenamed "Orbis", is known to be powered by an AMD x86-64 CPU with graphics based on its Southern Islands architecture, from the older report. We're now hearing that Sony may implement a multi-GPU solution of its own. According to an IGN.com report, the CPU in question will be a custom-version of AMD's A8-3850 quad-core APU. This should serve as indication that the processor cores will be based on AMD's K10 Stars architecture, rather than K15 Bulldozer/Piledriver.

The GPU, on the other hand, will be based on the "Southern Islands" architecture, and the IGN.com report pin points it to resemble Radeon HD 7670. The HD 7670 is a re-branded HD 6670, which is based on the 40 nm "Turks" GPU. Turks uses neither Graphics CoreNext nor VLIW4, but the older VLIW5 number-crunching machinery. The most interesting piece of information here is talk of a multi-GPU configuration between this Turks-based GPU, and the GPU that's embedded into the "Llano" APU. We know that the graphics core embedded into AMD A8-3850, the Radeon HD 6550D, can work in tandem with Radeon HD 6670 to yield an AMD Hybrid CrossFireX configuration called "Radeon HD 6690D2". This could be end up being Sony's graphics weapon of choice.

Durango Implements Always-On DRM, Multi-GPU

Microsoft's next-generation Xbox console, codenamed "Durango", will implement an "always-on" DRM, according to a VG 24/7 report. Always-on is a stringent anti-piracy measure that's already in use with some PC game titles published by Ubisoft, which requires the players to be connected to the Internet when playing. Even as its proponents (read: game publishers only), claim increased sales, it faces strong opposition from the gaming community. The Christmas 2013 (tentative) launch of Durango rides on the assumption that fast, always-on broadband Internet will have proliferated to every living room.

Next up, there's talk of Durango featuring a unique multi-GPU hardware design that doesn't resemble symmetric PC multi-GPU technologies such as NVIDIA SLI or AMD CrossFire, but instead, the two GPUs will be tasked with rendering the same object redundantly. Perhaps Durango features native flicker-free stereoscopic 3D capability, and the two GPUs are merely rendering the two planes? Once again, there's talk that these GPUs will be designed by AMD. The CPU of Durango will feature "four or six" cores. Lastly, Blu-ray disc is back as the storage medium, in these latest rumors.

GIGABYTE Intros Two Entry-Level Graphics Cards

GIGABYTE introduced two new graphics card models over the last week, based on GPUs that have been in existence for a while now. The first is the GV-R645-1GI, a low-profile single-slot graphics card based on the AMD Radeon HD 6450, and the GV-N440-1GI, a cool and quiet full-height double-slot graphics card, based on NVIDIA GeForce GT 440.

The GV-R645-1GI is armed with 1 GB of DDR3 memory over a 64-bit wide memory interface. It features clock speeds of 625 MHz core, with 1100 MHz memory. It uses a tiny, radial-fin fan-heatsink to cool the GPU. Its display outputs include dual-link DVI, HDMI (gold-plated connector), and D-Sub, which is detachable. GIGABYTE tried to pull off a neat marketing trick by mentioning that the card is "PCI-E 3.0 motherboard compatible", something every PCI-Express 2.0 and 1.0 graphics card ever made, already is.

Sparkle Introduces its GeForce GTX 560 SE Graphics Card

Today, Sparkle announced the launch of GeForce GTX 560 SE graphics card aiming at midrange market. As the most expected superstar product for mainstream market in the first half of this year, this card features a mature 40 nm GF114 GPU, delivering the ultimate next generation gaming experience. The excellent performance not only leads over similar competing products, but also shows no way inferior to GeForce 5 series. Now is the perfect time to upgrade your graphics card at a surprisingly affordable price!

Sony PlayStation 4 Codenamed "Orbis", Runs AMD x86-64 CPU, Southern Islands GPU

Sony may have codenamed PlayStation 4 "Orbis" (IPA: /ˈor.bis/), according to a Kotaku report. The next-generation game console is slated for "holiday season, 2013." This information was sourced by Kotaku, from a reliable source with a good track-record of accuracy. Orbis is derived from the Latin word for "circle", or "to circle/orbit". The term "Orbis Vita" or "Orbis Vitae" denotes "the circle of life." The specifications of Orbis known so far, include AMD-made x86-64 CPU, and an AMD-made GPU, built on the Southern Islands (Graphics CoreNext) architecture. In all probability, it could be a unified SoC, a highly scalar Fusion chip.

Palit GeForce GTX 680 JetStream Pictured

Palit unveiled a premium custom-design GeForce GTX 680 graphics card, called "JetStream". TweakTown got to review the card (find it at the source). The card features Palit's in-house designs for both the cooler and the PCB. The cooler consists of a sporty-looking shroud that holds three PWM-controlled fans, two 80 mm (sides) and one 90 mm (center); which ventilate a large aluminum fin heatsink that spans the entire length of the card. This heatsink relies on four heat pipes that make direct contact with the GPU, to draw heat. While the main heatsink only cools the GPU actively, a secondary base-plate heatsink cools the memory and VRM. The fans in the cooler are LED-lit.

Moving on to the PCB, Palit employed a 6-phase VRM, which makes use of high-grade chokes and DrMOS (driver-MOSFET) components, to power the card. The card draws power from 6+8 pin PCIe power connectors. The card is also factory-overclocked, with the Core (base) clock pushed up to 1085 MHz (from 1006 MHz reference); and Core (boost) upped to 1150 MHz (from 1058 MHz). The memory, on the other hand, has been stepped up to 1575 MHz (6.30 GHz GDDR5 effective), up from 1502 MHz (6.00 GHz GDDR5 effective) reference. It doesn't end there, Palit assures 10% overclocking performance, 9 dB quieter operation (than reference design), and 8°C lower temperatures. The card is expected to cost US $20-30 higher than NVIDIA reference design (so, around $530).

EVE Online Players Can Use In-Game Currency to Pay For Graphics Cards

Joystiq is reporting that CCP Games, the developer behind EVE Online, has struck a deal with NVIDIA for players to use in-game currency -- PLEX -- to purchase GeForce 560 GTX graphics cards, which run almost $200 at retail. That's fake money for real items, folks. NVIDIA and CCP will start by selling 100 graphics cards (limit one per EVE account) to test the demand. After CCP announced its NVIDIA deal, the in-game price of PLEX jumped "significantly," CCP said. PLEX (Pilot License Extension) is an in-game currency that can be used to pay for 30-day EVE subscriptions. PLEX can be purchased with ISK, EVE's main currency that can be earned throughout the game, meaning some dedicated players end up with a free subscription each month. Now they may end up with a "free" graphics card as well.

NVIDIA GPUs Power Top Two Russian Supercomputers

NVIDIA today announced that the two most powerful supercomputers in Russia will use NVIDIA GPUs to address some of the world's most challenging scientific problems across a broad range of fields.

Underscoring the dramatic growth in the adoption of GPU computing across world scientific communities, the new Russia Top 50 supercomputer list released today reveals that the top two systems are accelerated by NVIDIA Tesla GPUs. These two supercomputers are housed at Lomonosov Moscow State University, which was recently named a CUDA Center of Excellence, and the Joint Supercomputer Center of the Russian Academy of Sciences (JSCC RAS). Moreover, GPUs are accelerating 12 of the country's top 50 systems -- up from seven just six months ago.

EVGA Gives Away Free High-Flow Bracket to First 200 Registered GTX 680 Users

EVGA is giving away High-Flow Brackets to the first 200 people in North America and Europe who register their EVGA GeForce GTX 680 graphics cards. The replacement expansion bracket, designed by EVGA, has more perforations than the one on NVIDIA's reference design board, its makers claim it shaves off an extra 3°C from the GPU temperature. It should be listed on the EVGA web-store for purchase separately, soon. For more information about the offer, visit this page.

Embedded Graphics: Imagination Tech Supplies More GPU IP than all Others Combined

Jon Peddie Research (JPR), the industry's research and consulting firm for graphics and multimedia, announced estimated mobile devices graphics chip shipments for 2011. Mobile devices include smartphones, tablets, cameras, vehicles, and game consoles.

The market for smart mobile devices is soaring. Mobile devices are not necessarily replacing the PC, but they are outselling PCs. Increasingly mobile devices have a graphics processor and those processors are integrated in the device's system on a chip (SoC) application processor.

Gainward Announces its GeForce GTX 680

Palit-owned AIB maker Gainward has done as expected today by unveiling a branded GeForce GTX 680 graphics card. Seen below, the Kepler-powered card features support for DirectX 11(.1), CUDA, PhysX, 3D Vision Surround, and for 3-way SLI.

Like all the other stock GTX 680s announced today, Gainward's offering has 1536 CUDA Cores, a base GPU clock of 1006 MHz, a 256-bit memory interface, 2 GB of GDDR5 VRAM @ 6000 MHz, a PCI-Express 3.0 bus, two 6-pin PCIe power connectors, and dual-DVI, HDMI and DisplayPort outputs. The card is now available priced at €498.95.

AVADirect Starts Selling NVIDIA GTX 680 Graphics Processing Unit

Today AVADirect, a leading custom computer manufacturer, announced that they are now selling the new NVIDIA GTX 680 graphic processing unit. The GTX 680 is based off of the previous 580 graphics processing unit that NVIDIA has been re-inventing for the past year. They produced ground-breaking performance benchmarks, and reached a whole new level of capabilities that were not tapped by the graphics community, until now. The GTX 680 promises to reach new heights and create a serious impact in gaming and 3D rendering performance.

NVIDIA GPU Boost Technology is similar to the Turbo Boost function seen in Intel Sandy Bridge processors. The graphics processing unit will increase the clock speed based on current demand allocated to the graphics card. If you're immersed in a game, and you come across a scenario that takes more graphics processing power, the Boost Technology will increase the core clock thus keeping gameplay as smooth as possible. TXAA Technology, developed by Nvidia, is a new form of anti-aliasing that combines core functions of FXAA to create a balance between performance and image quality. Adaptive Vertical Sync enables per your current frame rate to provide the most fluent gameplay throughout many games available. PCI Express 3.0 has recently been adopted by Intel's X79 chipset and promised increased bandwidth for more performance. By using this technology through the bus of the graphics card, performance will be heavily improved, over a PCI Express 2.0 expansion slot.

GIGABYTE Out With its GeForce GTX 680

GIGABYTE took its own sweet time launching its GeForce GTX 680, but no harm done. Carrying the product code GV-N680D5-2GD-B, GIGABYTE's graphics card sticks to NVIDIA reference board design, decorated with a slick-looking abstract art. The card also sticks to NVIDIA reference clock speeds of 1006 MHz (core), 1058 MHz (boost), and 1502 MHz / 6.00 GHz effective GDDR5 memory. It is armed with 2 GB of memory sitting across a 256-bit wide memory interface. Based on the 28 nm GK104 GPU, NVIDIA GeForce GTX 680 is powered by 1536 CUDA cores, and a revolutionary, energy-efficient architecture. GIGABYTE GeForce GTX 680 is priced at US $499.

EK Unveils its GeForce GTX 680 Full Cover Water Blocks

Not to be out-gunned by Aqua Computer, Ljubljana-based EK Water Blocks has also announced its own liquid cooling offerings for NVIDIA's GeForce GTX 680 graphics card. The EK-FC680 GTX series full cover (GPU, RAM, VRM) water blocks are designed to fit standard GTX 680 (and 670 apparently) cards and feature G1/4" threaded fittings can be used on both sides.

There are four blocks listed by EK, two priced at €84.95 that have an electrolytic copper base, and two with an electroless nickel (EN) plated base which cost €94.96. The blocks can have either a POM Acetal or an acrylic cover.
Return to Keyword Browsing