Update at 6:45 .p.m. with additional information about QPI licensing.
Nvidia's last-minute conference announcement has turned into a bit of shocker.
Despite all the chest thumping at its gaming conference this week, the high drama of Nvision reached its denouement with a waving of the white flag. The world's largest graphics chip supplier announced support for high-end gaming graphics using Intel silicon. This has raised doubts about its clout in the gaming PC industry, based on the reaction at many hardware enthusiast Web sites and at least one PC maker.
Representative of the shock expressed after the announcement, a headline at AnandTech said: "Hell Freezes Over: Nvidia Announces Native SLI Support for the Intel X58 Chipset." Translation: Nvidia must use Intel supporting silicon to get its technology into future gaming systems--not its own.
One PC maker agrees with this sentiment. "When they were top dog they could have gotten away with this," a representative said, alluding to the Nvidia nForce 200 chip that, until the about-face Thursday, was required to enable high-end Nvidia graphics on future Intel Core i7 systems.
Speaking on the condition of anonymity, the representative said Nvidia was quickly getting boxed out by AMD's ATI graphics unit at his company. Though there are also performance reasons for this newfound preference for ATI graphics over Nvidia, in this specific case PC makers, and users alike, don't want to add a special Nvidia chip to enable graphics on high-end gaming systems, he said.
And this reaction is echoed at Anandtech and other hardware Web sites. "We heard from the very start that most motherboard manufacturers weren't going to use the nForce 200 + Intel X58 combination," according to Anandtech, referring to the i7's supporting silicon, the Intel X58 chipset.
At issue is Nvidia's Scalable Link Interface, or SLI, a critical technology for game enthusiasts who want to use more than one Nvidia graphics board to power the most demanding PC games like Crysis. Nvidia had been saying that the only one way to get to high-end game nirvana was by using its own supporting silicon.
But the Thursday announcement changed all that. Now users can configure SLI systems for Intel's upcoming Core i7 processors "natively" as Nvidia puts it. That is, without the Nvidia nForce 200 chip. "That's (the nForce 200) been the only solution. And that's been a very, very high-end solution," said Tom Peterson, director of Technical Marketing for MCP production at Nvidia.
All of this can be traced back to an earlier issue centered on whether Nvidia would make chipsets based on Intel's QuickPath Interconnect (QPI) technology to work with the i7. The company has indicated that it will not. "When we go to Bloomfield (i7 processor) we've already announced that we have no intention of building a QPI-based chipset. Because of that, we've offered only nForce 200," Peterson said.
(Correction: Nvidia maintains that it has a QPI license but has elected not to make a QPI chipset.)
This, in turn, has fueled speculation that Nvidia will get out of the chipset business. "I can see where some people would think that in the longer term, especially with the (i7)" said Dean McCarron, principal and founder of Cave Creek, Ariz.-based Mercury Research.
But McCarron thinks Nvidia will stay in the business in the near term. And this is borne out by Nvidia's indication this week about an impending announcement of a new integrated graphics chipset for Intel's current Core 2 architecture--not the i7.
Update on August 28 at 3:30 p.m. with comments on SLI and AMD-ATI
Nvidia is extending its support for Intel's upcoming Core i7 processors while it prepares to announce next-generation integrated graphics silicon.
The announcement marks an effort to expand Nvidia offerings on Intel's next high-end desktop platform, which had previously been referred to as "Bloomfield." Intel branded it Core i7 prior to the company's developer forum last week. Nvidia has already said that it has no intention to build a chipset for Intel's next-generation interconnect technology called QuickPath Interconnect or QPI, which is part of the i7 design.
Nvidia said Thursday that it will license its Scalable Link Interface (SLI) technology for Intel's Core i7 processor. Nvidia's technology will work in tandem with Intel's X58 chipset, the supporting silicon for the Core i7, which is due to ship in volume in the fourth quarter.
SLI allows systems to be configured with multiple graphics boards. So, for example, system builders and users can build systems with two, three, or four Nvidia boards.
In essence, Nvidia is offering what it calls "native" licensing of SLI to its partners and system builders. Native licensing will not require the use of Nvidia's nForce 200 bridge chip and thereby the company hopes to broaden the range of its graphics offerings on i7-based PCs.
To date, Nvidia has only offered nForce 200, "which is basically an SLI chip that acts like a PCI Express bridge. That's been the only solution and that's been a very high-end solution. We'll continue to offer this," said Tom Peterson, director of Technical Marketing for MCP production at Nvidia.
PCI, or peripheral component interface, is the most common interface inside a PC for add-in boards.
Nvidia diagram shows supported configurations for Intel "Bloomfield" Core i7 processor and X58 chipset: x8 and x16 refer to the number of PCI "lanes." Generally, the more lanes, the higher the performance.
(Credit: Nvidia)The distinction between native and nForce 200 is that native SLI "allows for more common configurations," said Bryan Del Rizzo, an Nvidia spokesman.
One source at a U.S.-based PC maker said that Nvidia was losing ground to AMD-ATI by not bringing out an SLI solution that could appeal to more system builders and users, especially with Intel's Core i7 on the horizon.
"It's something that customers have been asking us for a long time and actually a big change for Nvidia," Peterson said.
Nvidia will certify partner circuit boards in its Santa Clara, Calif., certification lab, Peterson said. Certification is required to enable SLI.
On another front, Nvidia will announce a new high-performance "motherboard GPU" in the coming weeks. This will be a follow-on to its GeForce 8200 mGPU, which is an integrated graphics chipset for desktop PCs that use Intel processors.
The upcoming mGPU will compete with the Intel G45 integrated graphics chipset.
Intel has scores of futuristic, potentially game-changing research projects but Nehalem is bet-the-farm reality. The first Nehalem chips--and the first drops of what should become a giant revenue stream--will arrive later this year. So, it is not surprising that real silicon and real systems are starting to appear.
What is Nehalem exactly? The architecture will scale from two to eight processor cores, have faster chip-to-chip communication (Intel calls this technology QuickPath), do a better job of adjusting performance levels to suit power needs, and have a higher level of integration (more logic will be built directly onto the processor die).
Other salient features include more scalable memory (each processor will have its own dedicated memory), the ability to do more stuff simultaneously (up to 16 threads with simultaneous multi-threading), and new instructions to increase efficiency (called SSE4.2 instructions). Here's how Intel describes Nehalem.
A four-socket Nehalem system with integrated memory controllers
(Credit: Intel)Maximum PC appears to be the first to build a Nehalem "Bloomfield" desktop system for everyone to see.
The system uses a 2.93GHz Bloomfield processor and an Intel motherboard with an X58 chipset (which had been codenamed "Tylersburg"), which will is also due to ship in the fourth quarter.
The Bloomfield chip is larger than current Intel quad-core processors (e.g., the Q6700), according to Maximum PC. This means more fans, bigger heat sinks, and more heat to dissipate overall.
Nehalem will support faster DDR3 memory. And this points to one of Nehalem's major departures from past Intel processors: the memory controller--which talks to the DDR3 chips--is now on the processor die. Previously, this was off-chip. In short, higher levels of integration generally means higher performance.
New "overclocking features" are also offered in Nehalem, according to Maximum PC. Overclocking--running the chip faster than its rated speed--is an absolute prerequisite for gamers. Which means, of course, that initially one of Nehalem's biggest draws will be gamers.
In the more distant Nehalem future, the mobile platform has gotten a name. At this point, Intel will confirm the code name only: Calpella. But otherwise "won't comment on speculation."
Other information posted on various tech Web sites about the 2009 Nehalem mobile platform in the past few days has been in the public domain for almost a year. That is, it will have an on-die memory controller and one version of the chip will have an integrated graphics processor--which will be a first for Intel. One new twist is the timing: it may launch in the third quarter of 2009.
Update at July 15, 3:00 a.m. PDT with additional information and corrections concerning the Intel-Nvidia dispute.
Nvidia said Monday that its multichip technology will be architected to work on Intel's upcoming Nehalem chip platform.
Nvidia SLI technology supports multiple graphics boards.
(Credit: Nvidia)This announcement may help Nvidia to work around a standoff with Intel over whether Nvidia can make chipsets that work with Intel's next-generation Nehalem platform, due later this year. And also demonstrates that despite Nvidia CEO Jen-Hsun Huang' s rhetoric, Nvidia must cooperate with Intel in order to thrive.
Monday's announcement has no relation to separate licensing negotiations, according to sources familiar with the discussions. In other words, Nvidia is not announcing a chipset for Nehalem--which would require a license. It is simply a statement that Nvidia will support Nehalem with its nForce 200 Scalable Link Interconnect (SLI) chip.
However, some reports say Nvidia has reached an agreement with Intel to license Intel's Quick Path Interconnect (QPI) technology, paving the way for Nvidia to design chipsets for Nehalem.
The nForce 200 chip will work with Intel's "Bloomfield" line of Nehalem processors and the accompanying Intel chipset. SLI allows Nvidia to use multiple graphics boards in one system.
Upcoming SLI motherboards will use Nvidia nForce 200 SLI silicon, Intel Bloomfield processors, and Intel Tylersburg (X58) chipsets, Nvidia said in a statement.
"The nForce 200 SLI processor features patented SLI technology for graphics bandwidth management and multi-GPU peer-to-peer communications, both required to optimize graphics performance," Nvidia said. GPU stands for graphics processing unit.
Future systems "can be powered by one, two, or even three Nvidia GeForce GPUs, including the new...GeForce GTX 280 and GTX 260 GPUs," according to Nvidia.
Nvidia included statements from system suppliers in the Monday release. "It's great to see that Nvidia opted to enable SLI on the future Intel Bloomfield platform," said Rahul Sood, CTO Voodoo Business Unit, HP. "Make love not war I say...and Nvidia's (enabling) of Intel chipsets to support SLI will make our jobs much easier."
Nvidia claims that nForce 200 SLI silicon with Intel's new Bloomfield processor and Tylersburg chipset core logic chipset will deliver up to a 2.8X performance boost over traditional single graphics card platforms.
Motherboards and PC systems that will use the Nvidia nForce 200 SLI chip, Nvidia GeForce GPUs, and Nvidia SLI technology will be available from companies such as Acer, ASUS, Dell, Falcon Northwest, Legend, and Velocity Micro.
While the marquee processor theme at IDF Shanghai is "milliwatts to petaflops," Intel is also set to offer a vision of universal connectivity.
The main theme for the event, which starts Wednesday, Beijing time, refers to "very, very big to very, very small and low power," according to Pat Gelsinger, senior vice president and co-general manager of Intel's digital enterprise group, speaking in a video.
(See: Intel rolls out five new Atom processors.)
"Milliwatts" refers to chips such as Atom, a tiny low-power, low-cost processor destined for ultramobile devices and low-cost desktops typically running either Linux or Windows XP. The first Atom chips will launch in June.
"Petaflops" refers to high-performance computing--what used to be called supercomputing. ("Peta" is quadrillion, or a thousand trillion; "flop" is floating-point operation.) Intel is targeting petaflop supercomputers that would compete with the fastest supercomputer in the world: IBM's Blue Gene/P machines.
Though more technology and product details will certainly emerge in the next two days in Shanghai, the main chip themes are already out there. Gelsinger spelled them out at briefing earlier this month.
The specs for Intel's Dunnington processor
(Credit: Intel)The chip buzzwords are: Tukwila, a new quad-core chip with 2 billion transistors, a whopping 30MB of cache, and a new interconnect technology called QuickPath; Dunnington, a six-core chip for multiprocessor computers that can support four or more processors (in this case, each with six cores); Nehalem, a follow-on to the current "Penryn" processors, it is a new 45-nanometer chip microarchitecture due in the fourth quarter that scales up to eight cores; and Larrabee, a visual-computing architecture that uses many cores ("many" usually means many more than a typical quad-core computer).
In addition to Atom, the processor spotlight will likely fall on Nehalem and Larrabee. Nehalem is a relatively known quantity; Larrabee, a relatively unknown quantity. So interest should focus on the latter.
Nehalem boasts increased parallelism, better branch prediction (to move instructions more quickly through the instruction pipeline), and an on-chip memory controller for increased memory performance--what Intel calls "memory latency reduction." Something, by the way, Advanced Micro Devices already has in its chips.
Larrabee is a graphics processor scheduled for the 2009-2010 time frame. It will include a new vector instruction set to improve the performance of graphics and video applications. Larrabee will be compatible with Intel's popular x86 instruction set, theoretically making life easier for software developers.
On another front, Intel is evangelizing universal connectivity, always a problematic proposition, simply because it invariably promises more (sometimes much more) than it can deliver. Intel puts it this way: "Imagine a day when a single device small enough to fit in your pocket...knows your tendencies and preferences and can adapt and optimize its interfaces to match what you are doing at any point any time...Imagine a day when this device...can dynamically become a hybrid combination of other computing and multimedia devices in close proximity." You get the picture. Intel calls this "Carry Small, Live Large."
On a slightly more practical level, the Cliffside technology is being demonstrated from the Mobile Products Group; it enables a single Wi-Fi adapter to function like two independent Wi-Fi adapters. The hope is that this technology could sync your MP3 and video files without a USB cable, directly and wirelessly connecting your notebook to your TV to view HD movies. More here.
There is also a demonstration of wireless device discovery and setup. This demonstration shows how to detect and connect to nearby wireless displays, using the familiar FnF7 (Function F7 key combination).
The following is a partial list of the sessions at this week's Intel Developer Forum in China, which runs April 2 to 3. Topics set to be covered at IDF Shanghai include Netbook, Nettop, Bloomfield (Nehalem), solid-state drives, QuickAssist (accelerators), system-on-chip (Tolapai), and USB 3.0.
The items below are taken directly from Intel's own material:
Intel Atom Processor for the Desktop
The Essential Building Block for Purpose Built, Basic Desktop Computing Devices: Intel's strategy for basic desktop computing devices. See how the new Intel Atom processor based desktop platforms provide the world's best solution for basic computing, content consumption, and thin client.
Intel Basic Mobile Platform with Intel Atom Processor: Growing New Markets
Overview of growth opportunities for internet-centric computing devices or netbook market. Introduction to 2008 Basic Mobile Platform with Intel Atom Processor Architecture. Understand low cost system BOM (Bill Of Materials) & implications for netbooks. How netbooks are different from Mobile Internet devices, Ultra Mobile PCs, and Entry Notebooks. Opportunities for OEMs/ODM/Software community.
The Intel-Powered Classmate PC: Innovating for the Future
An introduction to technology for the next billion users and how to designing a product for first time user in environments that are not ideal. This includes the implementation of specific technical features in the Second and Next Generation Intel-powered classmate PC including: - PATA MLC Flash - Rugged and high temperature tolerant parts/design - Fan-less thermal system design.
Passive and Fanless Thermal Enabling for Affordable Internet-Centric Nettop
Introduction of Nettop, internet-centric computing devices roadmap and associated thermal solution space. Value proposition of passive/fanless cooling - Cost, Noise, new usage.
High End Desktop Platform, Next Generation (Nehalem) Processor
Background on the Desktop version of our Next Generation Intel Microarchitecture (Nehalem) Processor. Key electrical and system design guidelines to consider in designing a platform with the new Intel desktop processor (Bloomfield). Information on the new Quick Path Interconnect (QPI) & the Integrated Bus Controller on the new Intel desktop processor (Bloomfield) processor. Updates on other key new technologies of this High End Desktop platform and the advantages they bring to developers
Microsoft Windows Rally Technologies
The Windows Rally technologies provide manufacturers of network-connected devices with an architecture that enables effortless setup, more secure and manageable connectivity to other devices and computers, and rich end-user experiences. Using Windows Rally technologies enables hardware and software developers to focus development resources on product differentiation rather than connectivity fundamentals. For our mutual consumers, the result will be a dramatic reduction in complexity. Connecting a network device to a PC will be effortless, reliable, and more secure.
SATA Solid-State Drives: Not All Drives are Created Equal
A technical overview of NAND-based high-performance SATA solid-state drive (SSD) technology. A comparative review between Intel's upcoming SATA SSD products and other SSD and conventional hard disk drive alternatives, focusing on the areas of performance, power and reliability
DDR3, the Optimal Memory for Notebooks
Understand the benefits of DDR3 memory. Understand the DDR3 eco-system readiness and outlook for Mobile. Learn how memory impacts critical performance and power benchmarks and what these benchmarks mean to end-users.
Extreme Mobile Gaming: Design Considerations for High Performance Notebooks
Introduction to Extreme Edition Gaming Notebooks. Details on Extreme Edition Notebook hardware & software offerings. Updates on available Thermal technologies. Methods to tune platform performance
Inside Intel Next Generation Nehalem Microarchitecture
Intel's Nehalem microarchitecture is the next major microarchitecture update from Intel aligning to it's "Tick-Tock" cadence for processors. This session will provide architectural insight into this new microarchitecture that will start shipping in the second half of 2008. You will learn the details behind the key microarchitecture features including: Enhancements to the out of order execution engine. Enhancement to the Platform bandwidth. Enhancements to the cache subsystem. Extension to the instruction set with SSE4.2. And more...
Intel QuickAssist Technology Components
Understanding of the scope of Intel QuickAssist Acceleration Technology and Components. Hardware and software architecture of Intel QuickAssist Technology FSB-FPGA Accelerators and AAL. Learn about the Integrated Accelerator & Intel's new System On Chip Product (Tolapai). Hardware and software architecture framework of Tolapai
Moblin.org - Open Source Development for the Intel Powered Mobile Internet Device
Moblin.org, the umbrella open source project hosts the core software stack and technologies for Intel powered Mobile Internet Device (MID). This course will go into technical details on Moblin.org, for how you could setup platform-specific environment and build applications based on Moblin.org sub-projects and technologies.
SuperSpeed USB - Fast Sync-N-Go for Mobile Devices
SuperSpeed Market overview and Industry timeline updates. Update on SuperSpeed USB (USB 3.0) specification architecture details. Guidance on industry opportunities for 2008/2009
Intel Architecture Based SoC (System-on-Chip)
Join us to hear about the vision for Intel Architecture-based SoCs and the unique benefits that they bring to the industry. This Technology Insight will focus on a generic SoC on-die architecture and its impact on SoC-based platforms for embedded communications, mobile devices, and consumer electronics. The talk will also outline Intel's expectations of the industry ecosystem in the areas of Electronic System Level (ESL) language, tools, and new methodologies.
- prev
- 1
- next
