Here is all of the news we’ve collected about Intel Corporation. This mostly includes their (now defunct) Larrabee chipset, as well as their many whitepapers and benchmarks regarding parallelism and their various platforms.
You might also want to check out
While this is primarily a visualization site, occasionally we dip into other hardware aspects as they may affect the work we do. For example, we have talked about SSDs in the past, especially with regards to PCI-Express based solutions. Faster access to storage means less time waiting when loading large files, or many small files. Thus we bring you the latest rumor on an Intel chip that we have been salivating over for some time now.
The Nehalem-EX is an eight core chip based on the Nehalem architecture. The -EX tag stands for EXpandable server market. That is, this chip will go into four socket machines, which means that you will have 32 real cores on which to perform your rendering. The Nehalem-EX will be sold as high-end Xeon chips. The current high-end Xeons are the 5500 series. The new chips will occupy the 6500/7500 series range.
The X6550 will have eight cores running at 2.00 GHz, with 18 MB of L3 shared cache and two QuickPath interfaces running at 6.4 GT/s. This CPU has a TDP of 130W.
The X7550 will have eight cores running at 2.00 GHz, with 18 MB of L3 shared cache and four QuickPath interfaces running at 6.4 GT/s. This CPU has a TDP of 130W.
The L7555 will have eight cores running at 1.86 GHz, with 18 MB of L3 shared cache and four QuickPath interfaces running at 5.8 GT/s. This is a lower power version with a TDP of 90W.
The X7560 will have eight cores running at 2.26 GHz, with 18 MB of L3 shared cache and four QuickPath interfaces running at 6.4 GT/s. This CPU has a TDP of 130W.
In case you are wondering, GT/s stands for gigatransfers per second. You can think of this as billions of bits per second. This is the raw data rate without any encoding.
Personally I am looking forward to system with four sockets filled with Nehalem-EX (32 cores), with a Fermi or two in it. That way when I start rendering, the entire town will brown-out.
via : Intel to release 8-core Xeons this month | Guru3D
via : Intel to Begin Nehalem-EX Shipments This Month | Guru3D
CNNMoney has a pair of short interviews with NVidia CEO Jen-Hsun Huang about the companies plans.
In the latest installment of Connected, Fortune Senior Editor-at-Large Adam Lashinsky talks with Nvidia (NVDA) CEO Jen-Hsun Huang about his company's strategy for tablets and smartphones, and its ongoing legal battle with Intel (INTC).
A few tidbits in the first video, about Tegra:
A few notes from the second video, on the Lawsuits:
via Connected: Interview with Nvidia CEO Jen-Hsun Huang – Fortune Brainstorm Tech.
Alan Dang of Tom’s Hardware takes an in-depth look into the next decade of GPU offerings and tries to divine what we may see happen. He begins by reminiscing about the crumble of the Sound Card market (moving from the PC speaker, to the high-end Sound Card, back to integrated audio), then covers the three major players: AMD, Intel, and finally NVidia. While he bashes on NVidia for the lengthy delays of the Fermi chipsets that had the potential to destroy the company (and some say still might), he does rightfully say that NVidia is still the market leader largely due to their superior software and SDK offerings available to developers.
The bottom line is that Nvidia has a considerable lead over both Intel and AMD when it comes to high-performance parallel computing. The investments it has made in creating viable commercial tools for GPGPU are already paying off with exclusive Adobe Creative Suite 5 support and broader adoption of CUDA among scientific professionals. If the company continues its momentum and aggressively grows the GF100-based product line, it has a chance to obtain iPod-like dominance in the market and at the very least, I think Nvidia has established itself firmly in the GPGPU world. Third place will have to go to either AMD or Intel.
One thing many people leave out in these types of articles is the software aspect. It matters little if Intel’s or AMD’s chips are ‘better’, if nobody can find reliable tools to develop for them. NVidia already has compilers from multiple sources (internal, PGI, etc) and several API libraries (PhysX, ScenX, OptiX, CUDA) to make using NVidia hardware simple and uniform. This alone, combined with their renewed push into Mobile chipsets (Optimus), Embedded chipsets (Tegra), and the high-end (Fermi) should easily carry NVidia through the next decade.
Opinion: AMD, Intel, And Nvidia In The Next Ten Years : Introduction – Review Tom’s Hardware.
Intel’s Anthony Gallo shows off three augmented reality demos on an Intel Core i5 processor. For the first demo, he starts the web cam on the computer and then holds up the December issue of Esquire magazine. Robert Downey Jr. pops onto the screen. The 3-D image on the screen is changes based on the angle that the magazine is held relative to the webcam. There are four different animations that can be unlocked by the magazine. We reported about this augmented reality issue of Esquire back on October 29. For the second demo, he uses a new Avatar toy from Mattel to unlock two different animations. For the final demo, he takes a picture of a motherboard and manipulates it in 3-D on the screen.
At the recent TechEd conference, Intel was demonstrating their new Smoke Gaming engine on a Windows7 machine and a Windows Vista machine. Both machines had identical hardware (Intel i7 chipsets internally, of course) but the Windows7 machine was significantly faster.
See what a difference the new Windows 7 kernel makes compared to Windows Vista. The windows 7 kernel is optimized for use with Intel core microarchitecture not only does it perform faster it also allows you to put a few cores to sleep while maintaining peak performance.
The demonstrator attributes this to Windows 7′s better handling of multiple threads and improved understanding of the Nehalem architecture.
See the demonstration after the break.
Intel integrated graphics chips have never been known for being excellent performers. Sure, they are good enough to surf the web, but not much beyond that. That makes them good for low cost machines that simply work, which makes them great for family members who need reliabile machines. Bit-Tech has posted an article on the Intel GMA HD Graphics Performance.
If you’re serious about your movie watching you’ll be more pleased that the HD-audio passthrough now works, you can get your picture-in-picture and deep colour resolutions, and when you’re not using it, the idle power use of the Core i5-661 and Core i3-530 is very low.
Likewise, Anandtech has posted an article called the Intel Core i3 530 Review. They too are looking at the graphics performance on the new Intel integrated graphics. From the article:
So the theory is that these graphics cores are easily overclockable. I fired up our testbed and adjusted the GPU clock. It’s a single BIOS option and without any changes to voltage or cooling I managed to get our Core i3 530’s GPU running at 1200MHz. That’s a 64% overclock!
That 64% overclock yielded a 52% increase in performance in World of Warcraft. That is not saying too much, as performance went from 12.5 frames per second up to 19 frames per second.
via Bit-Tech : Intel GMA HD Graphics: Is It Any Good?
via Anandtech : Overclocking Intel’s HD Graphics – It Works…Very Well
A pair of interesting legal turns have lead to a pair of interesting financial turns for the two giants of the computer graphics industry. First off, AMD published the quarterly report to wrap up 2009, and found themselves a surprise, they’re in the black for the first time in 13 quarters? How did this miracle happen? They got a big $1.25Billion from Intel for that legal settlement. But that’s not all:
To put things in perspective, last year AMD lost $1.4 billion just on Q4’08, and $3.1B for all of 2008, so this is a massive turnaround for the company. At this point both of their core divisions are turning a small profit, and the company’s reduced exposure to the foundry business has greatly improved their bottom line. AMD took a loss of $99M in Q4 from their share of Global Foundries – so if they were able to drop the foundry business entirely, they would have likely turned a true GAAP profit. Although AMD has several problems, at the moment it’s the foundry business that continues to be doing the most harm to them.
So while AMD is reveling in what will probably be a short-term gain (the Intel payment was a one-time event, afterall), NVidia is getting hammered on the other end of the legal spectrum. Just yesterday, a judge ruled that NVidia is in violation of some of those Rambus patents that have been haunting them for the last few years.
A judge with the U.S. International Trade Commission in Washington said today that Nvidia is violating three patents owned by Rambus. Two other patents are invalid, said Judge Theodore Essex. His decision, which is subject to review by the full commission, may result in a ban on imports of Nvidia chips and products that use them, including some computers made by Hewlett-Packard Co.
No doubt NVidia will settle or pay a fine to keep availability of product here in the States, which will lead to a huge hit on their financials for Q1 of 2010. Hopefully the Fermi and 3D Gaming will be big enough to offset the payout.
via AnandTech: AMD’s Financials.
Video cameras have become integrated into many devices. Intel is looking to take advantage of these video cameras for gesture based interaction. Intel’s Tony Chun describes mobile augmented reality for develping accelerators to be used in smart phones for image recognition and speech recognition. You can hit the link below for a two minute video describing the technology. From Intel’s website:
As video cameras are integrated into more devices from laptops to phones, computer vision capabilities have become increasingly attractive to enable applications such as gesture-based user interfaces and augmented reality. Intel has demonstrated a functional, reconfigurable hardware accelerator to enable advanced vision capabilities on mobile devices. This research from Intel Labs, St. Petersburg explores the automated design of reconfigurable accelerators based on tools-aided application analysis targeting computationally-intensive media workloads such as the SURF object recognition algorithm.
via Intel : Computer Vision Accelerator
Dr. Micheal J. Gourlay has published a 5-part series on the Intel Software Network about Fluid Simulation for Video Games. From his own description:
This article, the fifth in a series, describes the profiling and optimization of a fluid simulation, presented in the third and fourth articles. The first article summarized fluid dynamics; the second surveyed fluid simulation techniques; and the third and fourth presented a vortex-particle fluid simulation with two-way fluid-body interactions, which runs in real time. This article exploits yet another feature of Intel® Threading Building Blocks (Intel® TBB) to spread more work across multiple threads. This article describes a process for profiling CPU usage and uses that information to optimize and further parallelize the code so that it runs faster.
A great read covering simulation and animation, along with some complex CFD and parallelization strategies. Read all five parts:
In Intel’s booth at CES, you can see two giant double-HD screens showing a realtime visualization of web news stories.
The entire setup is apparently powered by a single i7 processor, and it’s easy to see how you could shrink an application like this down for use on oh, say, a tablet of some sort. Is it an efficient way to sort through information? No, no it is not. But it sure is engaging, and that’s gotta count for something.
See video of the setup on Gizmodo’s site.
Intel’s Double-HD MultiTouch Wall Is Incredible – Intel – Gizmodo.
VizWorld.com Got an idea for a story? Drop us a line at [email protected] or other means.
Comments