AMD Launches the ATI Radeon HD 4770

AMD Launches the ATI Radeon™ HD 4770, World’s First Desktop Graphics Card Utilizing Industry-Leading 40nm Process Technology

(NYSE:AMD) today launched the ATI Radeon™ HD 4770 graphics processor, a groundbreaking product that harnesses the state-of-the-art 40 nanometer manufacturing process for the desktop PC market, delivering best-in-class performance and energy efficiency at anticipated launch prices as low as $991. The ATI Radeon HD 4770 graphics card is the newest addition to the award-winning ATI RadeonTM HD 4000 series.· The new card features third-generation hardware support for the latest DirectX® 10.1 games like Battleforge, HAWX, S.T.A.L.K.E.R Clear Sky and Stormrise, delivering higher game performance and improved visual quality when compared to DirectX® 10.

· The ATI Radeon HD 4770 continues the AMD tradition of bringing industry-leading technology to market that drives a superior user experience. The ATI Radeon HD 4770 makes use of the latest GDDR5 memory technology, providing incredibly high data rates for fast game performance.
· Gamers looking to get the most value out of their graphics hardware investment can scale their performance using ATI CrossFireX™ technology, allowing a second ATI Radeon HD 4770 card to be added for a near-linear increase in performance.2
· The ATI Radeon HD 4770 graphics card allows users to enjoy HD digital content3 with incredible visual fidelity. The new card helps enhance the definition and clarity of lower resolution media, and through the second generation AMD Unified Video Decoder (UVD 2.0) technology, users can be assured of smooth video playback, crisp images, and vibrant colors on all HD content including Blu-ray discs.4 With support for HDMI, picture-in-picture (PiP) and the latest HD audio technologies like 7.1 surround sound, the new graphics card enables an impressive cinema-quality home entertainment experience.
· Celebrating the company’s 40th Anniversary, AMD continues its heritage of focusing on the design and development of new products and platforms that deliver compelling user experiences at an exceptional value. By combining new AMD Athlon™ X2 7850 processors, AMD 7-Series chipsets and ATI Radeon™ HD 4770 series graphics cards, AMD enables a PC platform that can bring HD video to life while delivering massive performance headroom and video processing acceleration.

“Consumer pressure in this economic downturn is shifting demand to lower prices and the price band around $100 has the highest demand and highest volume,” said Rick Bergman, senior vice president and general manager, Graphics Products Group, AMD.5 “AMD has responded with the launch of the ATI Radeon HD 4770, targeting this market with cutting-edge technologies like 40 nm processors, support for DirectX 10.1 gaming and GDDR5 memory. This strategy stands in stark contrast to the old “trickle down” technology approach some still use to address this important segment.”

“As a leading source of cutting-edge game development tools, Emergent has a passion for delivering realistic, visually stunning game play as well as improved performance,” said Doug Kubel, vice president, Product Development, Emergent. “By supporting DirectX 10.1 in the Gamebryo LightSpeed development system, we are able to deliver improvements in both image quality and frame rates. DirectX 10.1 graphics hardware like the ATI Radeon HD 4770 makes it possible to fully unlock the experience delivered by the combined power of DirectX10.1 and Gamebryo LightSpeed.”

Intel ready with its first 8-Core xeon processors

Santa Clara, California-based Intel is expected to detail one of its highest-anticipated server processors at the upcoming International Solid-State Circuits Conference in San Francisco between February 8 and 12, 2009. The chip maker is expected to provide an early look at its first eight-core chip, which will be part of the company's Xeon family of processors. These chips are to provide a new level of performance for server systems based on a dual-socket motherboard.

Unfortunately, at this time, there aren't any details regarding the Xeon processors to be discussed at the February 9 presentation. According to the ISSCC programme, executives of the global leading chip maker will be discussing an 8-core, 16-thread Xeon processor, which will be manufactured using a 45nm process technology. According to additional info, said processor will also be featuring 2.38 transistors and support operation of up to 6.4GT/s.

As expected, Intel is yet to officially comment on the respective Xeon processor, which it will likely do during the ISSCC event. “We are presenting 16 papers at ISSCC, but don't haveanything further to share at this point,” said Nick Jacobs, a company spokesman in Singapore.
According to the available details, Intel's presentation might just be related to the company's highly-anticipated Xeon processor that will be based on the Nehalem EP architecture. This server CPU is known to have been designed specifically for dual-socket servers and workstations. According to previous details, this processor line is due to be released sometime in early 2009.

The upcoming chips will sport some of the features currently available for Intel's other Nehalem-based chips, including the use of Intel's Quick Path Interconnect (QPI), which basically replaces the front side bus, featured on Intel's previous generation of Xeon processors. This feature boosts the computer's overall performance.

AMD's Phenom II at 6.5GHz, smash 3DMark record

A team of hardcore speed freaks with access to liquid nitrogen and liquid helium created an environment for OC'ing one of AMD's latest Phenom II X4 chips in -232 degrees Celsius weather. Once the silicon was appropriately chilled, they proceeded to push the clock speed to a staggering 6.5GHz, which proved nimble enough to deliver a patently absurd 45,474 3DMark05 score.The adoring public was there to bear witness, and every nail-biting second of it is hosted up in the video just past the break.

AMD's Phenom II at 6.5GHz, smash 3DMark record

via:engadget.com

AMD finally launches Phenom II and Dragon Platform

AMD's 45nm, quad-core Phenom II has finally arrived. It's available in two flavors, the 2.8GHz X4 920 and the 3.0GHz X4 940 Black Edition, going for $235 and $275, respectively. AMD's naturally touting the ostensible cost savings of the chips . The Dragon platform includes a Phenom II proc, 4800-series Radeon HD graphics and a 790-series motherboard, all of which should combine for low power consumption -- up to 40 percent compared to Phenom processors. Where the Phenom II definitely wins is the fact that some AMD folks will be able to drop in the chip to existing motherboards, but otherwise it looks like value hounds and performance junkies aren't being served by this new chip.

AMD finally launches Phenom II

Performance

The Phenom II's integrated memory controller and HyperTransport interface give it a technical edge over competing Core 2 Quad chips, which lack those features. Intel moved to an integrated memory controller and began incorporating its own version of HyperTransport--dubbed QuickPath Interconnect--only with its Core i7 platform. The integrated memory controller and HyperTransport interface allow Phenom II processors to achieve a higher memory bandwidth than Core 2 Quad processors can, by eliminating the bottlenecks created by a frontside bus and an external controller. The arrangement, in theory, improves system performance.

Model / Processor Frequency:	AMD Phenom II Processor Model X4 940 / X4 920 / 3.0GHz, 2.8GHz
L1 Cache Sizes:	64K of L1 instruction and 64K of L1 data cache per core (512KB total L1 per processor)
L2 Cache Sizes:	512KB of L2 data cache per core (2MB total L2 per processor)
L3 Cache Size:	6MB (shared)
Memory Controller Type:	Integrated 128-bit wide memory controller, capable of being configured for dual 64-bit channels for simultaneous read/writes
Memory Controller Frequency:	Up to 1.8GHz with Dual Dynamic Power Management
Types of Memory:	Support for unregistered DIMMs up to PC2 8500 (DDR2-1066MHz)
HyperTransport 3.0:	One 16-bit/16-bit link @ up to 3600MHz full duplex
Total Processor Bandwidth:	Up to 31.5 GB/s bandwidth
Packaging:	Socket AM2+ 940-pin organic micro pin grid array (micro-PGA) (backward compatible with Socket AM2)
Fab location:	AMD's Fab 36 wafer fabrication facilities in Dresden, Germany
Process Technology:	45nm (.045-micron) DSL Silicon on Insulator (SOI)
Approximate Transistor count:	approx. ~758 million (65nm)
Approximate Die Size:	258 mm2 (45nm)
Nominal Voltage:	.0875-1.5 Volts
Max Ambient Case Temp:	62 degress Celsius
Max TDP:	125 Watts
ACP:	*to be announced after launch
Future Memory Controller Note:	Future 45nm processors versions are planned to include support for DDR3 memory

Availability and Phenom II Processor Pricing

Consumers can expect to see systems featuring Dragon platform technology from leading system builders and OEMs in the first quarter of 2009. In addition to Dragon platform technology-based offerings from AMD’s OEM partners, PC enthusiasts who enjoy building their own systems can search “Dragon” on popular e-tail sites such as newegg.com and tigerdirect.com to find special promotions on this new desktop PC platform.

In addition to Dragon platform technology, individual AMD Phenom II X4 processors will be available from AMD channel e-tailers at launch:

AMD Phenom II X4 940 Black Edition processor – (3.0GHz) – $275 MSRP

AMD Phenom II X4 920 processor – (2.8GHz) – $235 MSRP

Intel announced completion of 32nm Process Development Phase

Intel announced that it had completed the development phase of its next-generation manufacturing process that further shrinks chip circuitry to 32 nanometers.

The company said that it is on track for production readiness of this future generation using even more energy-efficient, denser and higher performing transistors in the fourth quarter of 2009.Intel will provide technical details around the 32nm process technology along with several other topics during presentations at the International Electron Devices meeting (IEDM) next week in San Francisco.

Finishing the development phase for the company's 32nm process technology and production readiness in this timeframe means that Intel remains on pace with its ambitious product and manufacturing cadence referred to as the company's "tick-tock" strategy.That plan revolves around introducing an entirely new processor microarchitecture alternating with a cutting edge manufacturing process about every 12 months.

The Intel 32nm paper and presentation describe a logic technology that incorporates second-generation high-k + metal gate technology, 193nm immersion lithography for critical patterning layers and enhanced transistor strain techniques. These featuresenhanced transistor strain techniques. These features enhance the performance and energy efficiency of Intel processors. Intel's manufacturing process has the highest transistor performance and the highest transistor density of any reported 32nm technology in the industry, the company claims.

"Our manufacturing prowess and resulting products have helped us widen our lead in computing performance and battery life for Intel-based laptops, servers and desktops," said Mark Bohr, Intel Senior Fellow and director of process architecture and integration. "As we've shown this year, the manufacturing strategy and execution have also given us the ability to create entirely new product lines for MIDs, CE equipment, embedded computers and netbooks."

Other Intel IEDM papers will describe a low power system on chip version of Intel's 45nm process, transistors based on compound semiconductors, substrate engineering to improve performance of 45nm transistors, integrating chemical mechanical polish for the 45nm node and beyond; and, integrating an array of silicon photonics modulators. Intel will also participate in a short course on 22nm CMOS Technology.

Shanghai Chips from AMD now available

AMD has released its new Shanghai platform, signaling a move to 45nm. The first chips out the door are quad-core Opteron parts, which AMD claims will deliver up to 35 percent more performance and up to a 35 percent decrease in power consumption when idle.Acording to CNet, a desktop platform (codenamed Dragon) consisting of 45nm Shanghai desktop CPUs along with AMD 700 series chipsets and ATI Radeon HD 4000 graphics will also see the light of day in the same time frame and attempt to compete with Intel's Core i7 platform.

"This enhanced AMD Opteron processor represents the most dramatic performance and performance-per-watt increases for AMD products since the introduction of the world's first x86 dual-core processors nearly four years ago," Randy Allen, AMD senior VP for Computing Solutions Group, said in a statement. "Simply put, the quad-core AMD Opteron is the right technology at the right time."Shanghai, which is essentially a refresh of Barcelona and not an entirely new architecture, supports DDR2-800 memory and comes with a tweaked Direct Connect Architecture. The current batch of 75-watt Shanghai chips will be followed up by a launch of 55-watt Opteron and an SE 105-watt part in Q1 2009.

Shanghai Chips from AMD now available

Broad Software Ecosystem Support from AMD

AMD today announced ecosystem support by leading x86 operating system (OS) and virtualization partners for the new Quad-Core AMD Opteron processor, codenamed “Shanghai.” Built on 45nm process technology, the enhanced processor will offer innovations for AMD software partners like Microsoft, Red Hat, Sun Microsystems and VMware specifically designed to improve OS and virtualization performance. Strategic improvements in comparison to 65nm Quad-Core AMD Opteron processors include a larger L3 memory cache, and features designed to further optimize application performance in virtualized environments. The new AMD Opteron processor is uniquely architected to handle more memory-intensive applications while also enabling customers of all sizes to operate a flexible and scalable IT infrastructure.

“AMD is seeing rapid industry growth and demand for cutting edge technologies such as cloud computing and virtualization,” said Earl Stahl, vice president, Software Development, AMD. “As a result of our ongoing collaboration with software partners, AMD customers can soon easily upgrade to the new Quad-Core AMD Opteron processor. Enterprise customers can leverage the ongoing enhanced performance and features like AMD Virtualization (AMD-V) technology that AMD continues to deliver, while protecting the investments made to their current software infrastructure.”With more companies turning to virtualization to help improve data center efficiency and costs, AMD designed the new AMD Opteron processor to deliver a robust platform that can manage customers’ most demanding workloads, such as web hosting and complex database management, helping to maximize the business benefits of virtualization.

ISV Ecosystem Support

By collaborating with leading OS and virtual infrastructure vendors to help ensure applications are optimized for the new Quad-Core AMD Opteron processor, AMD is providing customers with greater opportunity to utilize the software solutions that best meet the needs of their businesses.

“With the adoption of Microsoft Windows Server 2008 Hyper-V to address customers’ infrastructure virtualization needs, it’s exciting to see the new Quad-Core AMD Opteron processor coming to market to help improve the performance and scalability of virtualized environments and help provide power savings benefits for customers,” said Mike Schutz, director of product management, Windows Server virtualization at Microsoft. “We look forward to continuing to work closely with AMD to help customers utilize virtualization technologies and enable businesses of all sizes to make the most of their IT investments.”

“Red Hat is committed to bringing performance, reliability and security to our customers as they expand their use of open source virtualization throughout the enterprise,” said Brian Stevens, CTO at Red Hat. “As evidenced by our recent collaboration with AMD on live migration technology, Red Hat continues to deliver business value to customers via high-performance, standards-based virtualization solutions for heterogeneous environments.”

“Sun is committed to working with AMD to include optimization and tuning advancements in the Sun xVM virtualization software portfolio in order to meet the desktop and server virtualization requirements of our customers,” said Jim McHugh, vice president of marketing, Datacenter, Sun Microsystems. “Sun xVM Server, a datacenter-grade server virtualization solution, runs on 45 nm Quad-Core AMD Opteron™ processor-based systems and offers customers the ability to virtualize and manage heterogeneous workloads, including Windows, Linux and Solaris™ operating systems, as guests. Sun xVM Server is an easy-to-use, open source hypervisor that enables a simplified and secure environment for our customers.”

“The VMware platform helps AMD customers to minimize capital and operating costs, increase availability of mission-critical applications, and promote energy efficiencies,” said Shekar Ayyar, vice president of infrastructure alliances, VMware. “The advanced virtualization features in the new Quad-Core AMD Opteron processors, such as AMD Extended Migration and AMD-V with Rapid Virtualization Indexing, can extend those benefits to optimize application performance in VMware virtualized environments.”

AMD also recognizes support for its forthcoming and current Quad-Core AMD Opteron processors from software and virtualization partners such as Parallels, Transitive and Virtual Iron. For more information about the software support of the Quad-Core AMD Opteron processor. source

Intel Core i7 Nehalem launching on November 17

Intel provided information about its next-generation micro-architecture (Nehalem) with the Core i7 processors at a media event on Wednesday in Athens, Greece, prior to the worldwide official release of the system on November 17.Intel Core i7 Nehalem launching on November 17

The event ran under the headline of "the fastest processor on the planet". The two major characteristics of the new i7 architecture are the triple-channel integrated memory controllers and the new interface called Quick Path Interconnect (QPI). Other highlights include the monolithic die to consolidate all four cores in a single piece of silicon, and the shared L3 cache. Further improvements include the use of the Hyper-threading as well as the Turbo Boost technologies. The Core i7 processors are also optimized to operate with Intel's X58 chipset.The Corei7 series will feature three quad-core models: The 2.66 GHz Core i7 920 will aim for the sub-$1000 class of PCs, the 2.93 GHz Core i7 940 for more expensive PCs in the $1500-$2000 range and the 3.2 GHz 965 Extreme Edition at enthusiast PCs.

The prices (1K pieces) of these CPUs are $999 for the Extreme Edition, $562 for the 940 and $284 for the 920.The X58 Express chipset covers both the northbridge/southbridge operations. The X58 IOH supports up to 36 PCIe 2.0 lanes. Depending on the processor model, QPI may run at either 6.4 GT/sec (Core i7-965) or 6.4 GT/sec (Core i7-940, i7-920) to communicate with the X58 Express chipset in each direction. This is twice the bandwidth of Intel's currently available X48 chipset, plus the full-duplex transport instead of the half-duplex. AMD's HyperTransport 3.0 still has an advantage herewith up to 41.6GB/s of bi-directional bandwidth.However, the X58 Express does not support the legacy PS/2 IDE, serial or parallel ports.

Turbo mode

Under the "TurboMode", the utilization of the number of active cores is used to adjust the frequency. For example, if only two cores are active, then there will be much more headroom within the thermal and power envelope and the CPU will be clocked higher. So the "turbo" frequencies of each core depend on the number of cores active. We could say that this is mainly a power saving feature.

Turbo Mode boasts single threaded applications as well as the performance of multi-threaded applications, according to Intel. Below is a graph that Intel showed with the Core i7 Extreme CPU to process in single-core and quad-core configurations, with Turbo Mode enabled. The graph shows the percent performance increase with Turbo Mode enable.
Technorati Tags: core i7, intel i7, chipset, core i7, gadgets, hardware, Intel, Nehalem Chipset, processor

Intel shows working 80-core processor

Last year, Intel announced a project to build a processor that would pack no less than 80 cores. While the idea may seem rather futuristic to some, the chip maker is reported to have already demonstrated the CPU at the Gitex Dubai trade show. Reports say that the Santa Clara based company wasn't shy at all at showcasing the processor, with the demonstration being made in the general public area outside of the trade show.

Intel shows working 80-core processor

As we all know, Intel has other multi-core projects in the pipeline, and Larrabee is one of them. However, the showcased processor was not Larrabee, but another CPU. The company is said to have started working on this CPU long before Larrabee came into play. The showcased CPU features 80 small cores, and is able to deliver 1 Teraflops of computing power. The 80-core chip is said to need only 78.35W of power, and that it features a 3.13GHz clock speed.

According to Fudzilla, Intel also detailed several other features of the CPU. The technology it comes with allows for almost all of the cores to be totally shut down when not needed. Also, it seems that the CPU can downclock all the way to 780MHz in such situations, and that the peak Teraflops performance will drop to 0.01. The CPU would only need 6.45W to compute the 4tile, 4x4 matrix mult withcomm equation.The research processor was built under the 65nm process and the company expects it to be ready for the market in the next five or seven years. It seems that Intel will come out with this eighty-core CPU as a part of the future line of sixteen, thirty two, sixty four CPUs, right before a ninety six-core version. The chip manufacturer considers the multi-core as being the near future for the CPU industry, although such an approach might sound rather idealistic.

Since the current multi-core processors are already facing problems concerning the lack of adequate software able to scale on a large number of cores, Intel should also consider conducting a research project in this area as well, so as to allow users to benefit from all the eighty featured cores.

Technorati Tags:intel, cpu, 80-core, processor, gpu, hardware

AMD 45nm CPU Lineup Details to be revealed

At long last, power users have a plethora of performance numbers to ponder now that Intel has lifted its NDA on Core i7 benchmarks. But even though the first batch of benches show the new architecture living up to the hype, AMD isn't packing its bags and going home. On the contrary, the rival chip maker has a slew of 45nm chips coming out, starting this month.

Citing sources at un-named motherboard makers, DigiTimes reports AMD will launch a pair of 45nm quad-core desktop CPUs (Deneb) designed for AM2+ systems this month. The Phenom X4 20550 will come clocked at 3GHz and the 20350 at 2.8GHz. A series of 45nm triple-core chips are also on the way, though these won't start shipping until Q1 2009. These chips include the 14x00, 12x00, and 1xx00e series. On the high end, AMD plans to launch six 45nm quad-core Deneb chips and four entry-level Propus chips in the same time frame.All the new releases could potentially have AMD competing with its upcoming dual-core Athlon X2 processors. To prevent this scenario from playing out, AMD will delay shipping its 45nm AM3-based dual-core parts (Regor) until Q3 2009.

The new processors will also lead to price cuts within AMD's existing product line as the chip maker looks to clear its inventory. If you're an AMD-loyal, keep your eyes peeled for some tantalizing deals on Phenom processors this holiday shopping season and beyond

Rumours about Intel launching Montevina Platform for notebooks

Intel is going to update its Montevina notebook PC platform in April, 2009 with the introduction of the Montevina Refresh platform, according to a DIGITIMES report, which cites unnamed sources within Intel. The launch of the platform will be accompanied by two new processors, the Core 2 Duo T9900 and P8800.


Intel also plans to unveil its GM47 chipset for high-end notebooks in first quarter of next year. Entry-level and small form factor (SFF) PC will also not be over looked, as Intel will launch the GL43 and GS40 chipsets in July or August.A deluge of new processors for the Centrino 2 platform are soon going to be made available by the world’s leading chip maker. Also, the GM55 chipset for Intel’s upcoming 6th generation Centrino platform, Calpella, will become available in July or August next year.

Intel thinking use of jet engine's cooling technique

Intel is looking to cool your laptop with the exact same technology that a jet engine does. The issue of burning legs (that’s right, burning legs) has been an issue on the mind of Intel for some time now, and they’re looking to soothe that with their latest breakthrough.

Intel jet engine's cooling technique

Intel has been focusing on the increasing issue of hot thighs with something called Laminar Flow. Laminar Flow occurs when a fluid or gas/air flows in parallel layers, allowing a non-turbulent way to misdirect hot air away from the surface of a jet engine (or laptop). As demonstrated, this technology allows efficient cooling of temperatures upwards of 1,000 °C.A demo of this technology was given at this week’s Intel developer forum in Taiwan by Mooly Eden, Intel’s head of Mobile Platforms Group. “We are licensing it to our customers so they can keep making thinner and thinner laptops,” said Eden.

Technorati Tags:laptop, notebook, processor, intel jet cooling, jet engine cooling

Intel's Clarksfield to begin production in second half of 2009

Penryn-based Monteniva just hit the streets a few months ago, but hardware geeks are fickle fellows, so Intel said today that Clarksfield, its Nehalem-based mobile CPU, will go into production in the second half of next year; sadly, there was no word on when the chips will actually be available to consumers in the Calpella platform, which (rumor has it) will have built-in WiMAX. Clarksfield will include an onboard memory controller and a bunch of other new techs like its desktop brethren, so the promise of performance gains comparable to what we've seen in the desktop benchmarks is there -- we'll just have to wait at least a year to enjoy them, so get comfy.

Memory Kits For Intel i7 CPUs from OCZ

Memory Kits Optimized for Intel i7 CPUs from OCZ

OCZ Technology unveiled the industry's first triple channel memory kit designed specifically for the impending Intel Core i7 processor / Intel X58 Express Chipset, available in 3GB and 6GB kits.

The new memory modules are Optimized for the Core i7's triple channel mode, combining low voltage requirements, speed, and latency.

Memory Kits For Intel i7 CPUs from OCZ

Developed for enthusiasts and early-adopters, the low voltage OCZ Triple Channel solutions are the choice counterparts for leading-edge performance that won't inhibit the functionality of Core i7 CPUs."OCZ engineered these 3GB and 6GB kits specifically for Intel's imminent platform," commented Dr. Michael Schuette, VP of Technology Development at OCZ Technology. "By using special-screened, low voltage-capable ICs, these modules are the perfect complement for the Intel X58 Chipset to deliver the maximum performance by enabling low latency settings and higher system frequency."The advanced memory modules offer clock speeds up to 1600MHz at just 1.65V memory voltage. Elpida memory has also anounced the development of DDR3 memory that match these specifications.

OCZ will release the triple channel kits under the Platinum and Gold series designed specifically for enthusiasts and gamers:

OCZ PC3-10666 Gold 3GB (3X1GB) 1333MHz CL 9-9-9-20 1.65V
OCZ PC3-10666 Gold 6GB (3X2GB) 1333MHz CL 9-9-9-20 1.65V
OCZ PC3-12800 Gold 3GB (3x1GB) 1600MHz DDR3 CL8-8-8-24 1.65V
OCZ PC3-12800 Gold 6GB (3x2GB) 1600MHz DDR3 CL8-8-8-24 1.65V
OCZ PC3-10666 Platinum 3GB (3x1GB) 1333MHz CL7-7-7-20 1.65V
OCZ PC3-10666 Platinum 6GB (3x2GB) 1333MHz CL7-7-7-20 1.65V

OCZ?s Triple Channel kits feature propriety XTC (Xtreme Thermal Convection) heatspreaders. The company has not provided pricing details.

Dual-core Phenom pops up again

Much (digital) ink has run on the subject of dual-core Phenom derivatives. Rumor sites initially expected such chips to launch quad-core Phenoms last year, but we later heard that AMD had shelved the concept until the 45nm switch. Then, 65nm dual-core Phenom listings somehow made their way onto a price search engine in Europe, and unofficial benchmarks of a 2.2GHz model surfaced. We still haven't heard a peep from AMD on the subject, though.

Expreview now claims to have the latest scoop. The Chinese site says AMD may have given up on the 2.3GHz chip we heard about last month, but the chipmaker's October roadmap includes two new dual-core Phenom derivatives. Branded Athlon X2 7550 and Athlon X2 7750, the two CPUs have respective clock speeds of 2.5GHz and 2.7GHz, which they each pair with 3MB of total cache—likely 512KB of the L2 variety per core and 2MB of shared L3 cache per chip.The roadmap reportedly pegs the arrival of these CPUs in the first quarter of next year. However, Expreview says AMD will manufacture them using 65nm process technology, and both models will have 95W thermal envelopes. This claim sounds a tad implausible, although AMD has made some strange moves in the past.

Intel Core i7 920, 940 and 965 models

While some companies are already out there promising Core i7-based gear, we haven't yet heard many official specifics about the actual processors that'll be at the heart of them. That's now changed in a pretty big way, however, with some supposedly authentic leaked materials revealing the complete spec list and some pricing details. Apparently, you can expect Core i7 920, 940 and 965 models to roll out in November, with 'em clocking in at 2.66GHz, 2.93GHz, and 3.2GHz, respectively. Each, as expected, are quad-core, 45nm processors, and each boast the same 8MB of shared L3 cache, 256kb of L2 cache per core, and TDP rating of 130W. No word on individual unit pricing just yet, but it looks like the Core i7 920 will run $284 in quantities of 1,000, with Core i7 940 upping things considerably to $562, and the "extreme" Core i7 965 demanding a hefty $999.

Model number	Intel Core i7 920 (mainstream)	Intel Core i7 940 (performance)	Intel Core i7 965 (extreme)
Clock	2.66GHz	2.93GHz	3.20GHz
Cores	4	4	4
Threads	8	8	8
Process	45nm	45nm	45nm
Socket	LGA1366	LGA1366	LGA1366
TDP	130W	130W	130W
L2 cache	256KB per core	256KB per core	256KB per core
L3 cache	8MB shared	8MB shared	8MB shared
Price (in thousand unit quantities)	$284	$562	$999

Intel Officially launches six-core Xeon 7400

Thanks to Google's Chrome and Microsoft's IE8, discrete processes are the new hotness, and more apps running simultaneously on your desktop is a great excuse to buy a CPU with more cores! On cue, Intel has officially launched its new Xeon 7400 processor, hitting 2.6 GHz on six cores and boasting an advertised 43 percent jump in performance over the lowly quad-core 7300, which had only half the 7400's 16MB of L3 cache. Impressive stuff, especially considering a bonus 10 percent drop in power consumption, but at $2729 for the top of the line model it's not exactly consumer-oriented.


Intel has no current plans to release a six-core CPU to the mainstream market, and few applications would be able to scale well enough to take full advantage of the additional two cores. Intel seems to be pushing Nehalem for the consumer market, which will launch as a quad-core. Dunnington customers – large Web 2.0 companies like Myspace – will be the ones who benefit most from the extra performance and power efficiency, which may enable them to develop compute-intensive features like high-definition video sharing.
source

New Chip Design to Challenge AMD from INTEL

Intel Corp. cracked the lid Tuesday on a new chip design that is at once a big challenge to smaller rival Advanced Micro Devices Inc. and an admission that AMD nailed a key design feature before it slipped into a severe financial slump.

Intel, the world's largest computer chip maker, showed off the new blueprint, known as a microarchitecture, for its chips at a developers conference in San Francisco.Though some of the details were already known, the design's formal unveiling represented another demonstration of Intel's advantage over AMD in cranking out new chip designs once every two years, a factor that helped send AMD's stock price down 5 percent in an overall down day for technology shares.

AMD has racked up nearly $5 billion in losses during the past 18 months and last month replaced Hector Ruiz, who had been running AMD for six years, with a new chief executive, Dirk Meyer.

The details of Intel's microprocessor architecture are always highly technical. But they're also closely watched because of the ubiquity of Intel's chips in personal computers and corporate servers.

One of the most significant changes was already known. Intel now plans to build a part called an integrated memory controller - which moves information between the microprocessor and the computer's memory - directly into the processor itself.

That's a key change because processors are asked to do more and more, and any lag in communication can seriously hurt performance. AMD has already been incorporating integrated memory controllers into its processors.

Because of that and other tweaks, Intel said its new design, which is code-named Nehalem, will triple the speed at which data can be written to memory or read back, compared to previous generations. Intel says Nehalem also will have nearly double the 3-D animation capabilities as past chips, and better utilize the multiple "cores," or processing engines, on each chip.

Chip makers are adding multiple cores to their chips, essentially jamming many separate processors onto the same slice of silicon, to make sure they're able to continue ramping up performance without running into overheating problems.

Intel said four-core Nehalem chips, which are due to be in production by the end of 2008 and will first target servers and desktop computers and later laptops, have the ability to turn individual cores on and off and can be programmed to boost the speed of active cores when the workload ramps up.

AMD: Promises, promises

Here we go again. AMD’s latest server roadmap promises six and then 12 core delights. Sorry, but given AMD’s current track record of over-promising and under-delivering, color me doubtful.

This is a perfect example of what I was saying yesterday about AMD. While the marketing machine is hyperventilating over technologies that are more than an arm’s reach away, when it comes to actually delivering these goods in a timely manner (and the timeliness window for processors is shrinking daily - a 3 to 6 month delay can be disastrous) AMD is notorious for its failure to deliver.

Also, this jump from six to 12 cores that AMD is promising, maybe it’s me but that sounds gimmicky. AMD is good at packaging arguments for triple-core processors and jumps from six to 12 cores as being “what out customers want.” I’m certain that AMD is trying to carve out a market with such statements.

Intel, SGI To Build 10 PFlops Supercomputer For NASA

Santa Clara (CA) - Intel and supercomputers is a combination of words we have been hearing quite a bit lately. Today we learned that the company has secured another contract with NASA, expanding on existing relationship that began with the deployment of the Columbia supercomputer in 2004. The new "Pleiades" system is planned to deliver 1 PFlops of computing power by 2009 and 10 PFlops by 2012.

Under the terms of a Space Act Agreement, NASA said that will work closely with Intel and SGI to increase computational capabilities for modeling and simulation at the NASA Advanced Supercomputing (NAS) facility at NASA’s Silicon Valley-based Ames Research Center at Moffett Field. As NASA’s Columbia supercomputer, currently listed at #20 in the list of the world’s 500 fastest supercomputers, is past its prime time, the new Pleiades is aimed to deliver the computing capability of new technologies and take NASA back to the top of the list.

Few details about Pleiades have been released, but NASA said that it targets a peak performance of 1 PFlops or 16x the performance of Columbia, which stands at about 61 TFlops. A 2012 is expected to result in a ten-fold increase in performance to about 10 PFlops.

This dramatic increase in performance raises questions how Intel and SGI will be able to deliver this speed, especially if we heard just yesterday that a 10 PFlops system would currently cost about $1 billion to construct. It seems as this system could be another project for Intel that takes advantage of Larrabee accelerator nodes.

Columbia went into operation in 2004 and is based on SGI Altix architecture with 10,240 Itanium 2 (Madison-9M core with 9 MB of L3 cache) processors, with the majority running at 1.5 GHz and some running at 1.6 GHz. Each Itanium 2 processor is rated at a maximum performance of 6.4 GFlops - which is only about 10% of what Xeon quad-core processors are estimated to be capable of today.
Last week, Intel said that it will be developing supercomputers in collaboration with Cray.
source:http://www.tomshardware.com