The dirty little secret of GPU marketing is that laptop graphics chips are never as powerful as you’d think. Think you’re going to run a game maxed out at 2K resolution with the GeForce GTX 880M in your laptop? Think again—even if it’s no trouble for a desktop-grade GTX 780. Read more…
Read this article:
Nvidia GeForce GTX 980M: A Graphical Leap For Gaming Laptops
You probably won’t have noticed the following problem, unless you happen to be the IT manager in an architecture firm or other specialist environment, but it’s been an issue nonetheless. For all our ability to virtualize compute and graphical workloads, it hasn’t so far been possible to share a single GPU core across multiple users. For example, if you’d wanted 32 people on virtual machines to access 3D plumbing and electrical drawings via AutoCAD, you’d have needed to dedicate eight expensive quad-core K1 graphics cards in your GRID server stack . Now, though, NVIDIA has managed to make virtualization work right the way through to each GPU core for users of Citrix XenDesktop 7, such that you’d only need one K1 to serve that workforce, assuming their tasks were sufficiently lightweight. Does this mean NVIDIA’s K1 sales will suddenly drop by seven eighths? We couldn’t tell ya — but probably not. Filed under: Networking , Software , NVIDIA Comments
See the original article here:
NVIDIA enables full virtualization for graphics: up to four remote users per GRID GPU
You probably won’t have noticed the following problem, unless you happen to be the IT manager in an architecture firm or other specialist environment, but it’s been an issue nonetheless. For all our ability to virtualize compute and graphical workloads, it hasn’t so far been possible to share a single GPU core across multiple users. For example, if you’d wanted 32 people on virtual machines to access 3D plumbing and electrical drawings via AutoCAD, you’d have needed to dedicate eight expensive quad-core K1 graphics cards in your GRID server stack . Now, though, NVIDIA has managed to make virtualization work right the way through to each GPU core for users of Citrix XenDesktop 7, such that you’d only need one K1 to serve that workforce, assuming their tasks were sufficiently lightweight. Does this mean NVIDIA’s K1 sales will suddenly drop by seven eighths? We couldn’t tell ya — but probably not. Filed under: Networking , Software , NVIDIA Comments
Read More:
NVIDIA enables full virtualization for graphics: up to eight remote users per GRID GPU
AMD AMD wants to talk about HSA, Heterogeneous Systems Architecture (HSA), its vision for the future of system architectures. To that end, it held a press conference last week to discuss what it’s calling “heterogeneous Uniform Memory Access” (hUMA). The company outlined what it was doing, and why, both confirming and reaffirming the things it has been saying for the last couple of years. The central HSA concept is that systems will have multiple different kinds of processor, connected together and operating as peers. The two main kinds of processor are conventional: versatile CPUs and the more specialized GPUs. Modern GPUs have enormous parallel arithmetic power, especially floating point arithmetic, but are poorly-suited to single-threaded code with lots of branches. Modern CPUs are well-suited to single-threaded code with lots of branches, but less well suited massively parallel number crunching. Splitting workloads between a CPU and a GPU, using each for the workloads it’s good at, has driven the development of general purpose GPU (GPGPU) software and development. Read 20 remaining paragraphs | Comments
More:
AMD’s “heterogeneous Uniform Memory Access” coming this year in Kaveri
In NVIDIA’s ongoing efforts to monopolize the technical-sounding graphics card market , the California-based components manufacturer today announced a fresh mobile line of GPUs aimed at notebook computing. That’s five new GPUs in total, with the GeForce GT 720M and 735M making up the “mainstream” segment, while the GT 740M, 745M, and 750M make up the “performance” portion of the lineup. All five cards include NVIDIA’s “GPU Boost 2.0” tech, which allows the GPU to alter its clock speed on-the-fly for the sake of efficiency — although this is mainly a software-level upgrade over the first iteration of Boost, and it’s still the same familiar Kepler architecture under the hood. It won’t be too long before we start seeing the newest NVIDIA mobile GPUs in notebooks at retail, as the PR says they’ll be in notebooks from “every leading manufacturer” in the coming months. Filed under: Gaming , Laptops , NVIDIA Comments
Visit site:
NVIDIA outs GeForce 700M GPUs for notebooks, boasts inclusion by ‘every leading manufacturer’
We’re here at NVIDIA’s GPU Technology Conference in San Jose, and company CEO Jen Hsun-Huang has just revealed the next step in its GPU roadmap. Called Volta, it’s scheduled to arrive after Maxwell , and will advance GPU technology with a ridiculous amount of memory bandwidth. Volta GPUs will have access to up to 1TB per second of bandwidth by stacking the DRAM on top of the GPU itself, with a silica substrate between them. Then, by cutting a hole through the silicon and connecting each layer it’s possible to move, according to Huang “all of the data from a full Blu-Ray disc through the chip in 1/50th of a second.” We aren’t exactly sure what that means for graphics, but being able to process data that quickly is bound to be a boon for gamers… whenever Volta actually arrives, of course. Filed under: Desktops , Gaming , Laptops Comments
More here:
NVIDIA reveals Volta next-gen GPU platform
NVIDIA’s GTX Titan is rumor no more , as the American computer hardware company unveiled the superpowerful graphics card this morning. With 2,688 CUDA cores, 6GB of GDDR5 RAM, and 7.1 billion transistors packed into the 10.5-inch frame, Titan’s capable of pushing 4,500 Gigaflops of raw power — NVIDIA’s pitching Titan as the means to “power the world’s first gaming supercomputers.” The company even showed off the Titan in its mightiest form, bootstrapped to two others running together (three-way SLI), which powers graphics showcase Crysis 3 running at its highest settings: a whopping 5760×1080 resolution across three monitors. Of course, a setup like that would cost you quite a pretty penny; just one GTX Titan costs $1,000, not to mention three (nor all the other hardware required to support it). Should you prefer your gaming PCs to not be of the neon-lit, triple GPU, above-$10,000 variety, NVIDIA was also showing off the Titan in a Falcon Northwest boutique PC. The company’s working with a variety of boutique PC makers to incorporate the Titan (see: Maingear ), making NVIDIA’s top of the line a teensy bit more accessible to your average joe. GTX Titan is the new top of the line for NVIDIA, effectively pushing aside the GTX 690 and setting a new benchmark for performance. Of course, with a $1,000 price tag and freedom — nay, encouragement — to tweak its nitty gritty settings, the Titan isn’t really meant for your average anyone. The PC game-playing early adopters, however? Here’s your next GPU. Hopefully you’ve got a big, empty space in your rig, as you’ll need it. The GTX Titan arrives on February 25th for $999. Gallery: NVIDIA GTX Titan Filed under: Misc , Gaming , NVIDIA Comments
See original article:
NVIDIA unveils the GTX Titan, an enormous graphics card that costs $1,000 (eyes-on)