Here is all of the news we’ve collected about NVidia Corporation. This includes their new hardware offerings like Tegra and Fermi, as well as their software offerings such as CUDA, PhysX, and more. Feel free to browse around, and then maybe check out some of these other tags:
NVidia’s next hardware revision is code-named “Kepler”, but more than that is largely a mystery. One site claims to have gotten a copy of some Kepler specs, and if they’re true it’s pretty impressive.
As the leaked chart shown, die size of GK110 is 550mm², which is not only bigger than 365mm² of Radeon HD 7970, but also 30mm² larger than its GF110. It’s said that GK110-based GeForce GTX 680 will be up to 45% faster than AMD Radeon HD 7970. As for the price, GTX 680 is set at US$649, US$100 higher than HD 7970.
Expected to drop in April & May, they’re showing numbers around 45% over Radeons HD7970 for $650. Of course, without public confirmation it’s anybody’s guess.
FireUser.com has a user-contribution from Antonio Fontenele benchmarking a Quadro 1800FX against the AMD FirePro v5800 in a variety of tests. In an interesting twist he compares the various vendor “optimized” drivers against their counterparts provided by Autodesk (Both of these cards are Autodesk approved), and finds a startling different in performance.
In Cadalyst Systems Benchmark 2011 test, Quadro a little faster than FirePro while using AutoCAD default drivers due c2011_8.dwg file score, but it was slower than FirePro in the other files (where FirePro was 102.32% faster). However, using AutoCAD optimized drivers, Quadro earned 613 points while FirePro earned a higher score equal to 2060 points. This is means about 336.05% faster in AutoCAD 2011.
Get all the numbers in the article.
NVidia’s Cyril Crassin has a great post on their Developer Blog talking about his in-depth research into realistic lighting in computer rendering. Discussing realtime vs offline, along with many algorithms and physics effects, it’s a great-read.
The key to our approach lies in a new algorithm and data structure that allow much faster computation. Instead of working on triangles (the traditional way of rendering graphics), we’re using voxels. Each voxel is one value in a 3D grid. The voxels are stored in an octree structure (a tree structure in which each node has eight children), with voxels as the nodes on the tree. Octree structures effectively compact the information stored in large amounts of graphics data. Octree structures use less memory, making them faster and more efficient at rendering tasks such as ray tracing.
via An Inside Look At New Research On CG Global Illumination « NVIDIA.
Big news from Nvidia today as their new “Maximus” system comes to light. First announced back at SIGGRAPH, it’s now available for public use. If you’ve ever found yourself wanting to do both GPU compute and rendering at the same time (Possibly using PhysX simulations with high-end rendering, or running GPU-accelerated CFD simulations combined with visualization) and found yourself dealing with complex driver configurations or dark incantations of system configurations to get it all to work, then Maximus is exactly the system you’ve been looking for.
“To those of us who have spent their careers focused on workstations, NVIDIA Maximus represents a revolution,” said Jeff Brown, general manager, Professional Solutions Group, NVIDIA. “Previous workstation architectures forced designers and engineers to do compute-intensive work and graphics- intensive work serially and often offline. They can now do them at the same time, on the same machine, allowing professionals to explore more ideas faster and converge quickly on the best possible answers.”
Done entirely in the driver, you can now buy a good graphics-card (Fermi-based NVidia Quadro, all the way down to the $200 Quadro 600) and match it with a nice NVidia Tesla card and let the driver map your processes entirely. Right now the matching is a bit naive, putting CUDA/OpenCL processes on the Tesla and DirectX/OpenGL processes on the Quadro, but that alone can offer a huge boost in performance by allowing both rendering and compute processes to operate in parallel at maximum power.
The system is already tested and approved for an impressive list of engineering & CAD tools like ANSYS, CATIA, and MATLAB, along with other tools like Adobe Premiere and Bunkspeed. For most of these tools, they’ve already got great GPU-accelerated compute aspects, and GPU-accelerated rendering features. Unfortunately, it’s almost always been either/or and never both at the same time. With the new Maximus drivers you can run both aspects at full-speed, without special configuration or dealing with special software settings.
They’ve also got a great list of OEM Hardware partners offering full support for hardware and software, all on day 1. Another great feature is their new adaptive power consumption. It’s no secret that NVidia cards can be tough on a power supply, and adding two to a machine can be a real strain. With the new Maximus system, you (or the Hardware OEM provider) can specify voltage limits, preventing the cards from drawing too much power and burning out your power supply.
It’s a complete package all-around. Get the full press release, along with some great demonstration videos of Maximus in action, after the break.
With SC11 on the horizon, you can expect many more such announcements, but today China and NVidia are excited about the use of 2,200 Nvidia Tesla cards to run the first ever computer simulation of a complete H1N1 flu virus at atomic levels.
The CAS-IPE researchers made the simulation breakthrough by developing a molecular dynamics simulation application that takes advantage of GPU acceleration2. It was run on the Mole-8.5 GPU supercomputer, which is comprised of 288 server nodes. The system was able to simulate 770 picoseconds per day with an integration time step of 1 femtosecond for 300 million atoms or radicals1.
The new SolidWorks2012 offers some limited GPU features focused around making the visuals pop a little more than classic CAD packages. Over at SolidSmack they take it for a test-drive with one of the lower-end professional cards, the Quadro 2000, and find it works surprisingly well.
The image below is a screenshot of a data set from NVIDIA shown in SolidWorks with RealView graphics on. RealView graphics utilize the GPU to render the semi-realistic graphics on the SolidWorks screen. The other window is the PhotoView 360 Preview window. PhotoView 360 is 100% CPU-based rendering and doesn’t task the GPU, so for PV360 rendering your benefits come along with more multi-threaded CPU cores. The SolidWorks models are all CPU as well. The GPU does little for processing the model, so more GPU’s won’t gain you any added performance.
Of course, it sounds like SolidWorks isn’t really pushing the GPU that hard, focusing only on some nice rendering features in the realtime viewport. High-end renderings are still classic CPU raytracing, and none of the software seems to use any GPGPU features, so the lower-end cards make for a nice inexpensive way to add some more “oomph” to your workstation.
Of course, NVidia is quick to point out the growing trend of designers using post-rendering tools like KeyShot and BunkSpeed to do their high-end renderings, which will definitely take advantage of higher-end Quadros.
via Can SolidWorks 2012 Spit the NVIDIA Quadro 2000 GPU Juice? – SolidSmack.com.
NVidia has just announced the newest version of their 3D Vision products, attempting to squash the common complaints of darkness and faint imagery from active displays. The new tech boasts “LightBoost” technology, doubling the typical brightness of the displays through special monitors and a new generation of classes.
“NVIDIA 3D LightBoost technology makes 3D games, movies and photos more stunning and life-like than ever before,” said David Wung, senior director of product management of Open Platform Business (OPBG) Group for ASUS Computer International. ”With 3D LightBoost and our new full-HD monitors, colors are richer, textures and subtle image details virtually jump off the screen, and the overall quality of the experience is something to behold. We are thrilled to be the first desktop display manufacturer to bring this new level of 3D visual quality to our customers with the ASUS VG278H.”
The glasses are still IR based, and available from retailers for $149. Hopefully this means new Vision Pro glasses are on the way!
via NVIDIA 3D Vision Vaults to New Dimension With Next-Gen 3D Glasses and Monitors – NVIDIA Newsroom.
ORNL is making a play to take back the #1 spot on the Top 500 with another upgrade to their Jaguar frankenputer, I mean supercomputer. Already part Cray XT4 and part XT5, the new “wing” will be XK6 systems running with NVidia GPU’s. The initial parts will be Tesla M2090′s, but then later parts will be based on the new Kepler design. The final system is designed to hold 18,000 GPU’s.
“Oak Ridge’s decision to base Titan on Tesla GPUs underscores the growing belief that GPU-based heterogeneous computing is the best approach to reach exascale computing levels within the next decade,” said Steve Scott, chief technology officer of Tesla products at NVIDIA, referring to computing performance levels of 1,000 petaflops. ”The Tesla GPUs will provide over 85 percent of the peak performance of Titan. You simply can’t get this level of performance in a power- and cost-efficient way with CPUs alone.”
I’ve personally come to believe that adding GPU’s to HPC’s is a short-term solution to a bigger problem. Until we can come up with some operating systems and better tools for programming these behemoths, the major hindrance will be the development cycle.
Click for fullsize
A new case study from NVidia covers the creation of an impressive 3D Display wall 25 JVC monitors driven by 13 NVidia Quadroplex systems. The result is an amazing synchronized display driven by a handful of workstations, offering up 52 million pixels of scientific data in a beautiful stereoscopic interactive display.
“Most of what people see on the display is the output of an interactive application. It’s not pre-rendered but rather interactively drawn on the screen,” he explained. “For a protein crystal structure, for example, it’s just a PDB file converted into a mesh, and this software knows how to render it. For volumetric data like an MRI [magnetic resonance imaging], it’s a Z-stack of images. What this means is that instead of a clinician having to cycle through a series of single grayscale images one at a time, if we write the right tools, people can visualize the MRI in stereo 3D as a continuous surface and see things like lesions more clearly.”
I’ve seen similar displays built on a smaller scale, but this is quite possibly the largest and highest resolution 3D display built to date. Now that it’s up and running, more and more schools are coming to them to try it out.
“It’s really one of those things where the sky’s the limit,” said McCrory. “We have astronomers doing incredible work with simulating the evolution of star systems. The Business School has shown an interest in visualizing economic data to show trends. We have requests coming from every school.”
Get the full case study after the break.
NVidia has really put a lot of emphasis on their upcoming ARM Tegra system codenamed “Kal-El”, and has slowly been leaking technical details over the last few months. The latest tidbit came out earlier today: Details of the “fifth core” of the new design.
“Our next-generation Tegra processor, codenamed “Project Kal-El,” is widely known as the world’s first quad-core mobile chip. Today, we’re unveiling Kal-El’s little-known fifth core in two new whitepapers that detail its “Variable SMP” architecture.” said NVIDIA’s Matt Wuebbling on a blog post on the company’s website. “This extra core – which we call the “companion core” – runs at a lower frequency and operates at exceptionally low power. During less power-hungry tasks like web reading, music playback and video playback, Kal-El completely powers down its four performance-tuned cores and instead uses its fifth companion core. For higher performance tasks, Kal-El disables its companion core and turns on its four performance cores, one at a time, as the work load increases.”
So essentially they’ve added a 5th “idle” core to the system, one that operates at super-low power and has just enough horsepower to keep the system running and able to power up more powerful cores as needed. This will be great for systems with a high idle time (like their mentioned websurfing and email) where the overwhelming bulk of time is just spent waiting for the user to do something.
via NVIDIA Reveals Kal-El’s Fifth “Companion” CPU Core – HotHardware.
VizWorld.com Got an idea for a story? Drop us a line at [email protected] or other means.
Comments