Ati Radeon HD 4850
What's better than press photos of the upcoming Radeon HD 4850 but not as good as actual performance results you ask? Since it's Saturday and we're feeling pretty lazy let's just say non-press photos. Not coincidentally that's exactly what you can see below, some real, up close and personal pictures of the cheapest RV770-powered card currently in plan, the Radeon HD 4850.
The new card is set to offer 480 Stream Processors, a 256-bit memory interface, 512 MB of GDDR3 memory and something that apparently Nvidia can only dream about, DirectX 10.1 support. From the looks of it, the HD 4850 is not very different from its predecessor, with the upcoming card's length, cooling solution and connectivity being strongly resembling the HD 3850. The Radeon HD 4850 is expected to be released around June 18.
Pictures courtesy of bit-tech.
Radeon HD 4850
Intel and Micron manufacture first sub-40nm NAND chip
AMD party trasher Intel and memory heavyweight Micron have now started bragging with the manufacturing of the first sub-40nm NAND chip. The record-braking chip is made using the 34nm process and is a 32 Gb multi-level cell part. The advancement in manufacturing technology has been made by Intel and Micron's NAND joint venture IM Flash Technologies and it will see the new 32 Gb chip being sampled as soon as next month with mass production being planned for later this year.
"This new 32 Gb device provides the best bit storage density available in the industry," said Brian Shirley, vice president of Micron's Memory Group. "Together with our partners at Intel, we're proud to have now taken the lead in production process technology."
The 34nm-build 32 Gb NAND chip is expected to help bring higher-capacity solid state drives while also lowering the costs of such solutions. It's a win-win situation.
Intel® Core™2 Extreme Processor QX9650
Processor Specifications:
CPU Speed:
PCG:
Bus Speed:
Bus/Core Ratio:
L2 Cache Size:
L2 Cache Speed:
3 GHz
05B
1333 MHz
9
12 MB
3 GHz
Manufacturing :
Core Stepping:
CPUID String:
Thermal Design Power:
Thermal Specification:
VID Voltage Range:
45 nm
C0
10676h
130W
64.5°C
0.85V – 1.3625V
Intel Core 2 Extreme Mobile Processor
Experience with the world’s highest-performing mobile processor.‡1 The Intel® Core™2 Extreme mobile processor delivers when you need the ultimate in gaming and multimedia performance. Take your system with you to LAN parties or competitive gaming. Process your HD videos from wherever you want to be without slowing down.
Designed from the ground up for an unbeatable gaming experience, the Intel® Core™2 Extreme mobile processor enables gaming realism like never before. And its energy efficiency kicks you more battery and thermal resources for even more gaming power, faster memory, greater storage and higher performance graphics. With the unrivaled power of Intel® dual-core technology, you get extreme mobile performance and advanced power management. Translation? Play harder in more places with better battery life. And with integrated wireless LAN, the only thing you’ll have to plug in is your instincts.
Extreme performance unwired. Get yours.
6MB of Shared L2 Cache
Takes on complex game physics, artificial intelligence, and multi-threaded games: welcome to a vengeance for unrivaled realism.
800 MHz Front Side Bus (FSB)
Provides incredible gaming performance and play so realistic that you’ll think you’re in the game.
44W Thermal Design Power (TDP)
Helps save energy with the option of fine-tuning for optimum gaming performance.
Increase your domination with superior ammunition.
- Intel® Wide Dynamic Execution enables better gaming and multimedia execution time.
- Intel® Intelligent Power Capability provides extreme energy-efficient mobile performance in game.
- Intel® Advanced Smart Cache with mega performance smart cache is optimized for multi-threading games.
- Intel® Advanced Digital Media Boost enhances ultra-realistic game physics and AI.
- Flexible AP Roaming provides enhanced wireless capabilities so you can remain unwired longer.
Via Launches Crysis-Capable Nano Processors
VIA Launches VIA Nano Processor Family[VIA]Power efficient processors based on 'Isaiah' architecture designed for optimized performance for mainstream PC markets and new device types
Beijing, China, 29th May 2008 - VIA Technologies, Inc, a leading innovator of power efficient x86 platforms, today announced the new VIA Nano processor family based on the VIA Isaiah Architecture.
Building on the market-leading energy efficiency of the VIA C7 processor family, the VIA Nano processor family offers as much as four times the performance within the same power range to extend VIA's performance per watt leadership, while pin compatibility with VIA C7 processors will ensure a smooth transition for OEMs and motherboard vendors, and provides them with an easy upgrade path for current system or board designs.
The first 64-bit, superscalar, speculative out-of-order processors in VIA's x86 platform portfolio, VIA Nano processors have been specifically designed to revitalize traditional desktop and notebook PC markets, delivering truly optimized performance for the most demanding computing, entertainment and connectivity applications, including Blu-ray Disc™ HD video playback and the latest PC games, such as Crysis™.
The VIA Nano processor family leverages Fujitsu's advanced 65 nanometer process technology for enhanced power efficiency, and augments that with aggressive power and thermal management features within the compact 21mm x 21mm nanoBGA2 package for an idle power as low as 100mW (0.1W), extending the reach of power efficient green and silent PCs, thin and light notebooks and mini-notes around the world.
"VIA Nano processors represent the next generation of x86 technology, providing the fundamental building blocks for a new genre of optimized computing solutions," said Wenchi Chen, President and CEO, VIA Technologies, Inc. "'Small is Beautiful' is more than a design strategy; it's our vision of where the PC market is heading and our new processors will help the market realize that dream."
VIA's 'nano' association also extends to VIA's signature silicon and platform design characteristics of power efficiency and form factor size reduction, as demonstrated by VIA's ultra compact Nano-ITX boards and the processor packaging used for the current VIA C7 processor family and the first generation of VIA Nano processors.
About the VIA Nano Processor Family
Initially to be launched in two skus, the VIA Nano L-series processors for mainstream desktop and mobile PC systems and the ultra low voltage U-series for small form factor desktop and ultra mobile devices such as mini-notes.
The VIA Nano processor family boasts the highly efficient VIA V4 bus interface and brings a host of technology firsts to VIA's processor platform line-up, including:
— 64-bit Superscalar Speculative Out-Of-Order MicroArchitecture: Supports a full 64-bit instruction set and provides for macro-fusion and micro-fusion functionality, and sophisticated branch prediction for greater processor efficiency and performance.
— High-Performance Computation and Media Processing: The high-speed, low power VIA V4 Front Side Bus starting at 800MHz, plus a high floating point unit, support for new SSE instructions, and two 64KB L1 caches and 1MB exclusive L2 cache with 16-way associativity gives a big boost to multimedia performance.
— Advanced Power and Thermal Management: Aggressive management of active power includes support for the new "C6" power state, Adaptive PowerSaver™ Technology, new circuit techniques and mechanisms for managing the die temperature, reducing power draw and improving thermal management.
— Scalable Upgrade to VIA C7™ Processor: Pin-to-pin compatibility with current VIA C7 processors enables a smooth transition for OEMs and mainboard vendors, enabling them to offer a wider range of products for different markets with a single board or system design.
— Greener Technology: In addition to full compliance with RoHS and WEEE regulations, product manufacturing will be halogen-free and lead-free at launch, helping to promote a cleaner environment and more sustainable computing.
— Enhanced VIA PadLock™ Security Engine: Industry-leading on-die hardware cryptographic acceleration and security features, including dual quantum random number generators, an AES Encryption Engine, NX-bit, and SHA-1 and SHA-256 hashing.
Intel's Centrino 2 Platform Delayed
Aside from delaying new Centrino 2 notebooks generally (duh), it also significantly quashes the chances of new MacBooks at WWDC—particularly since one of the issues is with the integrated graphics chipset, which is causing serious failures in OEM notebooks.
Apparently, the Centrino 2 launch will be rolling, with some chipsets hitting in July and others following later. One of the other problems is with FCC certification of the sets with 802.11n, so that wireless N support "may see a slower ramp as Montevina can only be shipped with support 802.11 a/b/g for now."
If you recall the Santa Rosa launch last year, new MacBook Pros shipped about a month after the chipset debuted, pointing toward an August launch for new Centrino 2-based notebooks. [TGDaily]
Dell XPS M1730 17in gaming laptop
Review Dell’s acquisition of gaming manufacturer Alienware, followed by the introduction of 13.3in and 15.4in laptops to its XPS range, signalled a shift from hardcore gaming machines, with the brand targeting the regular consumer instead.
Despite rumours circulating that the company may even stop making XPS–branded machines altogether, the XPS M1730 represents a return to core values - it's Dell’s most focused gaming laptop to date. You’ll find the use of high-end components, two cutting-edge graphics chips and every feature you’re ever likely to need, all wrapped in a 17in chassis that screams for attention.
Dell has christened this machine "The Beast" – a moniker it more than lives up to, not least in terms of size. It’s absolutely massive, taking the opposite route to Apple’s waif-like Macbook Air, and cramming in as many features as possible. It measures 56mm at its deepest, and weighs in at a spine-crippling 5kg.
You’ll need the arms of an Olympian to carry it to the next room, then, and possibly a golf cart if you want to lug it further. Criticising the XPS for is weight is perhaps a little unfair, as it’s designed purely as a desktop replacement machine, and is similar in size to most of its rivals.
The sheer bulk also manifests itself in several plus points, one of which being excellent ventilation. The use of two GPUs and twin hard drives typically result in a machine hot enough to warm a small group of homeless folk, but the M1730 remained cool to the touch at all times.
Lights everywhere
There’s also loads of room for a good-sized set of controls. The spacious keyboard is ideal for all-night gaming sessions, and is one of the most comfortable we’ve used on a laptop. The large, individually mounted keys respond to the lightest of touches, yet have a long enough travel to prevent typos. That said, the quality of the keyboard does lag behind that of the XPS M1330, with the board flexing slightly under pressure.
read full article:http://www.reghardware.co.uk/
Intel Six-core Processor Into 65 Watt Power Envelope
If there were still any doubts that over the scalability of the Core architecture, Dunnington should put any of those to rest. Like the quad-core Harpertown design, the multi-socket Dunnington will be targeting all three major server segments, covering the 130 watt, 95 watt and 65 watt power envelopes.
Considering that some Harpertown CPUs were rated at a 150 watt power envelope recently (updated to a 120 watt rating last week) and the fact that the 45 nm Dunnington will add two more cores, 9 MB L2 cache and a massive 16 MB L3 cache, the power achievement is impressive by any standard.
The Dunnington flagship, the X7460 model, will run at 2.66 GHz and will carry a thermal design power of 130 watt. The 65 watt version, which could become a major hit in blade-server designs will be clocked at 1.83 GHz. All Dunnington processors will be using FSB1066.
Intel to use Centrino 2 to promote solid-state drives
Intel will offer to bundle its promised solid-state drives with its upcoming Centrino 2 platform, it has been claimed. The plan is being portrayed as a bid to push SSDs into the mainstream.
Today, SSDs generally command a significant premium over the price of equivalent notebooks fitted with a hard drive. Quite how far Intel will use the bundle approach to drive down the cost of implementing SSDs remains to be seen.
Centrino 2 - codenamed 'Montevina' - is due to debut late June, according to industry moles. Back in March, Intel executive Troy Winslow said the chip giant would ship 80GB and 160GB 2.5in- and 1.8in-format SSDs in Q2, so the timing of both product types could easily coincide.
According to insiders cited by Digitimes, the Flash drives will be branded Intel High Performance SSD, with the model codes X25-M and X18-M for 2.5in and 1.8in sizes, respectively. The first SSDs will be 80GB. They'll use a SATA interface.
That makes the 1.8in model a logical update for Apple's MacBook Air - reviewed here - allowing the company to deliver a second-gen machine that connects to storage over a SATA bus rather than the slower, parallel ATA bus the current version uses. The report claims the 160GB version won't arrive until late Q4, with 250GB SSDs following next year.
AMD to launch 'Puma' laptop platform next month
Intel may be gearing up to release Centrino 2, but let's not forget its arch-rival, AMD, is to release a notebook platform of its own, codenamed 'Puma'. The cat will be let out of the bag at Computex, in Taipei on 3 June.
So what's AMD going to announce? Puma comprises a new (ish) processor, 'Griffin', and AMD's 780M chipset. Griffin is a dual-core part based around a couple of old Turion 'K8'A-class cores. Unlike existing Turion chips, Griffin was design specifically for laptops and so features extra, mobile-friendly circuitry for better power management.
Indeed, power rather than performance is Griffin's watchword, though we expect it will still push past AMD's current mobile CPUs when the speeds and feeds are announced in June.
Griffin is expected to contain at least 2MB of L2 cache, with each core having 1MB all to itself. The two cores will be able to run at different clock speeds, allowing either or both to slow down - as far as 300MHz - when their workload lightens.
Griffin's northbridge circuitry combines the usual HyperTransport 3 controller and a DDR 2 memory manager capable of handling 667MHz and 800MHz memory, AMD has said in the past. These elements and the CPU cores all operate on separate voltage planes, to allow unneeded components to power down.
So if the 780M's DirectX 10 integrated graphics is churning through HD content, the CPUs can slow right down or go to sleep without affecting the memory controller's ability to keep the GPU fed with video data.
Likewise the bandwidth made available by the HyperTransport controller can be squeezed according to need, reducing the power required to support in-bound data, out-bound data or both.
The 780M's graphics core will work with a discrete graphics chip, if one's present, to render all the basic stuff and leave the main GPU to power right down until its needed for 3D rendering. Again, that should boost the battery life of Puma-based systems.
The 780M - or, possibly, the M780 - will also incorporate AMD's SB700 chipset, which can link in up to 14 USB ports, six SATA ports and parallel ATA devices, and provide HD audio.
A once touted Puma component was HyperFlash, AMD's answer to Intel's TurboMemory: a Flash-fitted module that provides a small chunk of solid-state storage for Windows Vista's ReadyBoost and ReadyDrive technologies to make use of.
Turbo Memory is available in a fair few top-end notebooks, but it's largely failed to grab attention as a performance booster that Intel may have hoped it would. Partly that's because of corporate indifference to Vista - though this may change now Service Pack 1 is here, and AMD is likely to want to tap into that with HyperFlash.
Support for HyperFlash is integrated into the SB700.© The Register
Intel plans Q3 Core 2 Quad, Duo line-up shuffle
The latest leaks coming out of Asia point to Q3 introductions of faster Core 2 Quads and Duos. The quarter will apparently see the debut of the 3GHz Q9650, the 2.66GHz Q9400, the 3.33GHz E8600 and the 2.66GHz E7300.
They'll be priced at $530, $266, $266 and $133, respectively.
All three introductions will see existing processors come down in price to make room for the new parts. Some CPUs will get pushed off the list as a result: the 65nm Q6700, and the 45nm Q9450 and Q9300 will be dropped from the four-core range, while the E8300 will be excised from the two-core list too.© The Register.
Intel nudges Centrino 2 debut back
Intel's next major Centrino revamp - dubbed 'Centrino 2' - isn't coming until the end of June, and not this month, as the company's executives have stated in the past.
According to whispers coming out of Taiwan's notebook manufacturers, the chip giant will release the X9100, T9600, T9400, P9500, P8600 and P8400 'Montevina'-oriented processors in that post-Computex timeframe.
They'll be priced at $851, $530, $348, $316, $241 and $209, respectively, we understand. They're all dual-core, and they'll all be fabbed at 45nm.
The 'P' chips have a maximum power draw of 25W, the 'T's 35W. The 2.4GHz P8600 has 3MB of L2, as does the 2.26GHz P8400. The P9500, like the T9600 and T9400, has 6MB of L2. Those three are clocked at 2.53GHz, 2.80GHz and 2.53GHz, respectively. All five CPUs sit on a 1066MHz frontside bus (FSB).
The X9100's FSB is 1066MHz too, but it'll draw 45W, be clocked to 3.06GHz and pack in 6MB of L2.
Centrino 2's late-June arrival gets it just into the H1 2008 launch window announced by Intel in the past. In September 2007, CEO Paul Otellini let slip that Centrino 2 would arrive in May 2008.
Once Centrino 2's out of the way, Intel can focus on its key Q3 launch: the first mobile quad-core chip. The $1038 QX9300 contains 12MB of L2 - 6MB per core pair - and sit on a 1066MHz FSB. Its clock speed has been said in the past to be 2.53GHz.
Q4 will see the debut of the four-core Q9100, but its speeds and feeds remain unknown© The Register.
DIY Intel Atom Computer
AMD announces the GAME initiative
The AMD Game logos will appear on AMD-based desktops and notebooks that meet the system requirements to run nearly all of the latest games.
AMD has begun pitching its processors, graphics engines and chipsets as the ideal foundation for gaming PCs.
The AMD Game! initiative takes in two specs: the regular Game! PC and the Game! Ultra PC. The latter requires a four-core Phenom X4 9500 processor and ATI Radeon HD graphics in CrossFire configuration, all sitting on a motherboard based on AMD's 770 chipset.
The more basic Game! set-up also centres on the 770, but this time with an Athlon X2 5600+ CPU and a single Radeon HD 3650 GPU. In between, are various versions with Phenom X3 CPUs, Phenom X4s but no CrossFire, and so on.
All configurations come with suitable case stickers on the off-chance the hapless punter forgets what's inside his or her computer.
AMD's Game! stickers: for! your! box!
The Game! approach might sound like a take on Intel's notebook-centric Centrino brand, this time for gaming desktops, but it's actually closer to Nvidia's SLI Marketplace initiative - a way of highlighting systems based on each vendor's various components and suitably certified that all the bits work together.
The initiative takes in not only CPUs, GPUs and chipsets, but also rates mice, controllers, keyboards, cases, cooling systems and memory for compatibility with each trio of AMD silicon products. All this will make for "console-like simplicity". Apparently.
Like SLI Marketplace, AMD's Game! initiative will steer the build-your-own PC brigade along the path of righteousness - as AMD sees it - and also help novices see red with by nudging them toward system builders who've put together pre-assembled AMD-based rigs.©THE REGISTER
Firefox 3 RC1 ready for download
Go kick the tires and light the fires of Firefox 3 Release Candidate 1, which is now available for download at the Mozilla Firefox site. The update delivers a few aesthetic changes, as well as what Mozilla is calling major security enhancements. Users should also expect increased performance and stability for applications like Google Mail. There's an exhaustive list of known bugs to squash too, but that's par for the course with any software. [Firefox 3 RC1]
Intel Cuts Power Consumption Rating Of "Skulltrail Xeon"
Santa Clara (CA) - Intel is transitioning all of its 45 nm Xeon DP processors from the C-O to the E-0 stepping. There are no changes in the feature set of the Harpertown quad-core CPUs and Wolfdale dual-core processors, with the exception of the flagship model, which is the sister model of the company’s Skulltrail processor Core 2 Extreme QX9775. The thermal design power (TDP) rating of the Xeon X5482 drops by 20%.
According to a product change notification sent to Intel customers yesterday, all 18 current 45 nm Xeon processors are transitioned to the new stepping with first samples expected to become available on July 3. The updated processors are scheduled to begin shipping on October 6 and Intel asks its customers to be prepared to receive these CPUs by October 20 of this year.
Since there are no new features and (almost) no specification changes, Intel says that only "minimal re-qualification" will be required on the system builder side. However, the new chips will require a BIOS update.
The only CPU that does see specification change is the Xeon DP X5482: According to Intel, the 3.2 GHz chips carries a 120 watt TDP rating in the E-O stepping, compared to a 150 watt rating in the C-O version. The current Skulltrail processor Core 2 Extreme QX9775 is also clocked at 3.2 GHz and rated at a 150 watt TDP.
We don’t know whether the power consumption of the processor in fact drops, or if the new TDP rating is simply a bit closer to reality. According to a review posted by Tom’s Hardware, a Core 2 Extreme QX9775 consumes a maximum of 114 watts at 3.2 GHz. The power consumption climbs to about 138 watts when overclocked to 3.6 GHz and 176 watts at 4 GHz.
But enthusiasts who are interested in building a Skulltrail PC should have a closer look at Harpertwon processors - and not just because of the upcoming stepping update: Even the pricey X5482, which currently lists for a tray-price of $1279 per piece and sells for about $1400 in retail, is substantially cheaper than the QX9775, which has a tray-price of $1499 and sells for closer to $1700 per CPU in retail. Even if you are sinking more than $10,000 into a Skulltrail PC, a few hundred dollars of savings aren’t bad.
Intel’s first Nehalem mobo is X58
As we known, Intel’s next gen processor will be Nehalem, and desktop Nehalem will be Bloomfield. It is a 45nm process technology 4 core CPU, integrated with memory controller, SMT (Simultaneous Multi-threading) up to 8 processing threads, and w/ 8M cache. It has been revealed in IDF 08, but at that time no the supporting mobo is still being kept in secret.
Click to read more.
Today we find out that the first Bloomfield supporting mobo will be Intel X58(with ICH10/R) coming out Q4 this year. Because Bloomfield have already got memory controller integrated, so north bridge will be simplified. The biggest point of the board will be the odd lay out, dual channel, Triple channel DD3 supporting and Dual PCIe 2.0 X16 or Quad PCIe 2.0 X8.
The north bridge have couples of new features, but Intel keeps ICH10/R as its south bridge, so there is nothing more to say about it.
Also here is a pic we believe it is a X58,which leaked to the web earlier.
Fight card: AMD Radeon 4800 vs. Nvidia GTX 200
The high-end graphics card market isn’t the straightforward, mano-a-mano battle it used to be. Graphics cards with dual GPUs, systems with multiple cards in CrossFire or SLI implementations, and most recently hybrid graphics have all muddied the picture.
Meanwhile, Intel has improved its integrated graphics and has long-term plans in Larrabee to challenge AMD and Nvidia in discrete graphics.
But next month we should get a good, old-fashioned GPU showdown when AMD and Nvidia release entirely new, high-end GPUs. Details of these GPUs, AMD’s RV770 and Nvidia’s G200, and the graphics cards based on them, are trickling out.
The RV770, or ATI Radeon 4800 series, will launch on June 16, according to several sites. Graphics cards based on the first GPU in the series, the Radeon 4850 with 512MB GDDR3 memory, will be available immediately for $230. This GPU is the follow-up to the current Radeon 3850. Higher-end cards based on the Radeon 4870 with 512MB GDDR5 and the dual-GPU Radeon 4870 X2 with 1,024MB GDDR5–a first–will be announced on June 23. The Radeon 4870 cards won’t ship immediately because of the lack of availability of GDDR5 memory, but when they do show up prices should be around $350. The Radeon 4870 X2 cards won’t show up until early 4Q and will cost about $550.
One of the more intriguing specs of the Radeon 4800, according to the site TGDaily which has many details on the 4800 series, is “physics processing capabilities.” Physics is used to define how objects in a game move and interact. It can deliver a more realistic, compelling experience, but few games are designed to take advantage of physics.
There are really only two players here: Havok, which makes software for game developers and movie editors, and Ageia, a competitor best known for pioneering the PhysX Accelerator boards for gamers. Intel purchased Havok in September 2007, and Nvidia completed the acquisition of Ageia in February and plans to integrate the PhysX Accelerator into its GPUs. ATI previously worked with Havok on physics (and AMD reportedly considered buying Ageia late last year), but it is unclear where AMD is getting the physics technology now. (There have been some reports that AMD borrowed a programming language developed at Stanford University for the GPUs that power the Folding@home research project and extended it to support game physics.)
Nvidia’s G200 will reportedly launch on June 18 (some sites put it in early July). This GPU family was previously known at the GeForce 9900 series, but several sites are now reporting that Nvidia will switch to a new naming scheme: the GTX 200 series. Initially this should consist of two products, the GTX 280 and GTX 260, but otherwise there are few details aside from the fact that G200s supposedly contain more than 1 billion transistors (not including memory), use GDDR3 memory with a 512-bit interface, have an updated Unified Shader architecture, and will be in cards that cost $450 or more.
With the massive Computex tradeshow in Taipei coming the first week of June, I suspect we will know a lot more about these new GPUs before the rumored announce dates.
HP begs AMD PC owners to put XP SP3 on ice
Hewlett-Packard has told customers not to install Windows XP service pack three (SP3) on AMD-based desktops until Microsoft and HP cough fixes to the endless reboot snafu that has wreaked havoc on PCs.
Microsoft confirmed yesterday that it was scurrying to patch the problem after hundreds of angry XP customers first grumbled about the sizeable cock-up last week.
"HP is working diligently with Microsoft on a software update and will be proactively distributing a patch this week through HP Update that will prevent this error from occurring," said the computer giant in a statement. "HP recommends consumers with AMD-based desktops wait until after HP's or Microsoft's updates have been deployed on their systems to install Service Pack 3."
Microsoft’s message board was, within hours of the firm pumping the operating system’s final service pack out to the masses, inundated with complaints about machines being crippled by the endless reboot cock-up.
On Wednesday HP, which is the world’s biggest computer vendor, posted advice on its website about its Pavilion and Compaq Presario desktop PCs:
“After installing the initial release of Service Pack 3 for Windows XP an error condition can occur. The Service Pack 3 update copies an Intel power management driver to the computer that was not on the computer before the update.
“During Windows startup, computers with AMD processors may experience a blue screen error,” it said.
That acknowledgement, which somewhat swerves any admission of guilt, follows Microsoft’s statement earlier this week in which it blamed OEMs for “improperly placing a Windows XP image created for an Intel-based computer onto machines with non-Intel chipsets”.
Notably, Microsoft avoided specifics on which computer manufacturers had been guilty of mistakenly loading the wrong Sysprep image on their machines.
Microsoft also confirmed – much to the chagrin of XP customers who have patiently awaited SP3 after its release was spurned by a Vista-centric Redmond several times – that the endless reboot problem first surfaced way back in 2004 when service pack two landed.
The error, it said, was brought on by an orphaned power management Intel-only driver (intelppm.sys) that remains in the Windows Registry when the wrong image has been loaded onto AMD-based machines.
However, customers have also complained that the glitch has occurred following the install of XP SP3 on PCs from Dell, Gateway and Lenovo, while others have suggested Asus chipsets have also been crippled by the reboot problems.
Unsurprisingly, AMD, HP and Microsoft have all claimed that the issues have been fairly limited to some desktop computers. But, as is often the case with this type of FAIL, no tech vendor has been brave enough to step forward to provide a definitive number of how many PCs have actually been affected by the issue.
HP said it hopes to push out a patch named SP37394 to customers within the next week. No word yet on when y’all can expect to see a fix from Microsoft, however.©THE REGISTER
Intel: future iPhone to be Atom powered
An Intel executive has apparently claimed a future iPhone will be based upon the chip giant's Atom processor.
Speaking in Germany, one Hannes Schwaderer, Intel's MD for Central Europe, made the claim, according to a German-language ZDNet report.
Sachwderer also promised a raft of Atom-based devices in the next 12 months, all of them "a bit bigger than the iPhone".
Of course, what one Intel executive claims and what Apple will actually do are two separate things. More to the point, the current generation of Atom isn't sufficiently power efficient for device as small as the iPhone.
The 'Silverthorne' Atoms, launched last month, are aimed at what Intel calls Mobile Internet Devices. Yes, that's a term that could be applied to the iPhone, but what the chip maker has in mind are larger, more tablet-like devices not gadgets like phones.
For that, we have to wait until 2009-2010 when 'Moorestown', the next generation of Atom, complete with on-board graphics and a power consumption characteristic suitable for phones, is due to ship.
Even then, will Apple shift from the ARM chip the iPhone currently uses? It could. The current iPhone Software Development Kit already includes an iPhone emulator. Since that code runs on Intel's x86 architecture and is, therefore, fully compatible with Atom, running iPhone emulation on a future, Atom-based handset isn't beyond the bounds of possibility.
Emulation would be essential to allow Apple to carry forward all the third-party development work done on iPhone software up to that point.
We shall see...
Alun Taylor contributed to this reportAmd Phenom x3 vs Amd Athlon x2, 3 core slower than 2 core
Performance: Phenom X3 8450 15% Slower
The small Phenom X3 model, the 8750, clocks at a rate of 2.10 GHz. When compared to the Athlon 64 6400+ with 3.20 GHz and 6000+ with 3.00 GHz, it simply can’t keep up for many applications.
The 6400+ is approximately 15% faster than the Phenom X3 8450 in the benchmark course. Compared to the 6000+ it is almost 9% slower. The difference in price between the small Phenom X3 and the Athlon 64 X2 6400+ is 26 euros and it costs approximately the same as the Athlon 64 X2 6000+.
source:http://www.tomshardware.com/
AMD refreshes low-power Quad-Core Opterons lineup
Advanced Micro Devices is shipping B3 versions of its low-power Quad-Core Opteron processors.
AMD first detailed these processors in September 2007, when it unveiled the Quad-Core Opteron processor. However, earlier versions of the chips were affected by a bug discovered in December that reportedly forced AMD to suspend some processor shipments. The B3 version of the chips announced Monday fixed that bug.
The five chips run at clock speeds ranging from 1.7GHz to 1.9GHz. Three of the chips -- the 2344 HE, 2346 HE, and 2347HE -- are designed for servers with two processors, while the other two -- the 8346 HE and 8347 HE -- can be used in servers with four or eight processors. They are priced from US$255 to $873 in 1,000-unit quantities, a standard way of quoting chip prices.
The low-power Quad-Core Opteron chips have an average power consumption of 55 watts, AMD said.
Intel P45 boards said to have CrossFire problems
With the drivers now available for both the motherboard and AMD graphics cards the CrossFire setups have noticeable display issues. There's no indication as to where the problem is coming from but new driver version or BIOS revisions may solve the problem before the P45 motherboards make officially introduced, early next month.
AMD Rolls Out Low-power Quad-core Opteron CPUs
Sunnyvale (CA) - AMD's B3 stepping Barcelona processors appear to be finally on track as the company continues to announce new design wins and new models joining the existing line-up. The company is now offering ’HE’ of its dual- as well as 4P and 8P socket versions of its processors, offering an alternative to six competing Intel quad-core CPUs.
The new processors include three 2P versions (2344HE, 1.7 GHz; 2346He, 1.8 GHz and 2347 HE, 1.9 GHz) as well as two 4P and 8P models (8346HE, 1.8 GHz and 8347HE, 1.9 GHz) and expand AMD’s portfolio into a market in which the company needs to regain market share from Intel in order to achieve its profitability goals.
The fact that the TLB bug delayed the launch of these processors by about 8 months has turned out to be a major problem for AMD, as Intel has been the only option for companies that transitioned their servers to faster or energy-efficient quad-core processors. Even if AMD claims that these new HE-series CPUs are the first power-efficient quad-core
AMD rates its new HE Opterons at a power consumption of 55 watts. However, this value is described as "ACP" (Average CPU Power), which is AMD’s way to describe the power consumption of its processors and which is different from the "TDP" (Thermal Design Power) that Intel uses to come up with a value for its processors. Server vendors told TG Daily that there is virtually no way to reliably predict which processor will be more power efficient in different application environments. The only way to reliably measure the power consumption of server systems is to actually put test systems in place and evaluate them in a real-world scenario.
Romanian and Turkish scientists turn circuit boards into oil
Whether through a force of expanding environmental activism or just compliance with government edicts, the IT sector is in a pinch over how to safely recycle defunct computers and equipment.
Unfortunately, IT kit is packed with enough environmental hazards to put the Toxic Avenger's codpiece to shame. And with each generation of electrical merchandise released, dealing with the jettisoned remains of yesterday's gear turns into a bigger problem.
But a team of scientists from Romania and Turkey say they've found a simple and effective method to turn printed circuit boards from discarded IT kit into material suitable as fuel or for industrial use.
The researchers note that the plastic portion of waste electrical and electronic equipment (WEEE) is particularly tricky to recycle because it contains additives, heavy metals, and extremely toxic flame retardants. (You don't want too much polybrominated diphenyl ethers in your diet if you cherish your liver and brain.)
In their paper "Feedstock Recycling from the Printed Circuit Boards of Used Computers," the scientists describe using a process of heat and chemical decomposition to destroy or remove almost all of the hazardous toxic compounds. A copy of the paper can be found here. (PDF warning.)
The process isn't exactly light reading — but when it's done, what's left of the printed circuit board is pyrolysis oil (or bio-oil), which can be refined in a similar fashion as crude petroleum for fuel or can be used by industries to make other useful chemicals.
Indeed now more than ever, is there anything adding more RAM can't do?©THE REGISTER
AMD rejiggers management; Forms central engineering group
AMD on Monday named a new head of its computing solutions group and created a central engineering unit to improve execution at the chipmaker.
Among the moving parts:
- Randy Allen was promoted to lead the computing solutions group at AMD, which focuses on the company’s consumer and commercial processors. Allen had been responsible for AMD’s server and workstation business and had been in charge of microprocessor engineering. He replaces Mario Rivas, who was responsible for the Barcelona launch (see Tom Krazit’s take).
- Chekib Akrout joins AMD from Freescale to help lead a new central engineering unit. Akrout was in charge of design technology at Freescale and had developed chips for IBM. Jeff VerHeul, corporate vice president of design engineering, will be co-leader of the group along with Akrout.
- Allen Sockwell becomes chief talent officer replacing Michel Cadieux.
According to AMD chief operating officer Dirk Meyer, the moves are designed to “enhance our execution” and focus the company.
Intel said to feed Google solid state disks
Sources have told Digitimes that Google plans to test out SSD storage in an effort to lower power consumption at its vast data centers. The ad broker-cum-search engine will turn to Intel for the SSD gear. And, if true, that's great news for Intel, which hopes to make a run at turning SSDs into a big business.
As the report tells it, Intel will supply flash memory, while Marvell will provide the controllers.
For what it's worth, our sources tell us that Digitimes report is flat out wrong.
Still, it's interesting to speculate about Google paying for the more expensive SSDs on the data center scale. Like many service providers, Google is finding that lowering energy costs is its top priority.
Of course, there are performance implications as well.
Given the secretive nature of Google, it's rather hard for outsiders to tell what the company is up to. It's also damn hard to tell if the company's supposed data center magic really lives up to its billing or if the company just blows tons of cash and time designing its own systems.
Anyway, the report goes on to cite memory makers grumbling about a potential shortage of NAND flash chips due to increased server-side interest in SSDs.
Google and Intel enjoy a very tight relationship with Intel making custom motherboards for its chum. Intel's server chip customers have taken notice of this arrangement
©THE REGISTER
AMD to hire TSMC to fab 'Fusion' CPUs?
AMD is said to be planning to add chip foundry TSMC's name to its (short) list of processor production partners later this year. TSMC's alleged role: to fab AMD's upcoming 'Fusion' CPU.
So say unnamed industry sources cited by DigiTimes, and the notion isn't without merit or precedent.
AMD already uses Singapore's Chartered Semiconductor to produce processors. Chartered was signed up in November 2004, but it was almost two years - July 2006 - before AMD said the deal was yielding it revenues. The deal centred on 90nm chips, and in October 2006 it was claimed Charted had already been certified to punch out 65nm Athlons and Opterons.
The moles suggest TSMC will come on stream in the second half of the year. It's not clear whether that's the point at which the partnership will be announced, or that the foundry will begin production. We'd suggest the former, given the sources' claim that TSMC has begun testing an silicon-on-insulator (SOI) process - a key element of AMD CPU construction, incidentally.
Testing the process, fine-tuning it and winning AMD approval is likely to take some time, so if TSMC is gearing up to fab AMD64 parts, it won't be doing so in volume until 2009 at the earliest.
That's when Fusion is scheduled to debut. Fusion is AMD's modular architecture for multi-core processors, where CPU cores can be swapped out at the design stage for specialist processing units like GPUs and TCP/IP packet handlers. This, it believes, will allow it to build mix'n'match CPUs, allowing it to target different applications with the same core technology, more efficiently and thus more cheaply.
Of late, AMD has refused to discuss its manufacturing strategy, CEO Hector Ruiz saying only that the company's plans are "bold".©THE REGISTER
AMD announces low-power quad-core chips
The new HE (high-efficiency) processors have a thermal envelope of 55 watts, compared to higher thermal envelopes of 105 watts or 75 watts for other AMD quad-core server processors.
The low-power Opterons are available in both the 2300 and 8300 series. Prices are as follows: $873 for the 1.9GHz 8347 HE and $377 for the 1.9 GHz 2347 HE. (Keeping in mind that 2300 series processors are designed for servers that use two processors, while 8300 series processors are for systems that use four or eight processors.)
“Our new Quad-Core AMD Opteron HE processors were designed to help data center managers who see power consumption and virtualization as the keys to solving their overall performance equation,” Randy Allen, general manager at AMD’s Server and Workstation Division, said in a statement in the press release.
If AMD doesn’t suit you, don’t forget that Intel announced energy-efficient quad-core Xeon processors with a thermal envelope of 50 watts at core frequencies as high as 2.50GHz in March. (Thanks to Brooke at the Crave blog)
65 Nm GPUs For Xbox 360 In Production, Xbox '540' Coming In 2009
It has been almost a year since Chartered has begun taking the Xbox 360 Xenon CPU from 90 nm to 65 nm and it really was just a matter of time until other hardware would follow. According to the Taiwan Economic News, TSMC has initiated first wafer starts of the 65 nm Xenos GPU and Northbridge. Microsoft apparently has ordered 10,000 300 mm wafers from TSMC at this time.
As it is the case with any die-shrink, Microsoft should see substantial economic advantages from this move, supporting the company’s ongoing strategy to reduce the production cost of the console (the reduction of the Xbox 360 production cost has been one of the key reasons why Microsoft’s entertainment division has been able to notably increase its profits over the past seven quarters). If the 65 nm Xenos "v2" scales down linearly from 90 nm, the new die size should be around 125 mm2, while the eDRAM chip will remain at 70 mm2. The new production process should yield about 35% more GPUs per wafer than before.
TSMC will continue to be in charge of the wafers, while Nanya will be delivering the flip-chip packaging substrates. ASE combines the silicon and substrate and is responsible for QA.
Quite honestly, we were a bit surprised to hear that Microsoft did not decide to die-shrink both the CPU and GPU at the same time, especially because two different foundries are manufacturing the chips. However, our sources at TSMC explained that Microsoft has the same production philosophy as Nvidia: Wait for a manufacturing process to mature and then run the initial wafer order. Apparently, the transition was simulated in detail by ATI and the tapeout happened without problems, at least according to our sources close to ATI. Keep in mind that ATI is only a contracted partner for Microsoft: Both the CPU and GPU are officially Microsoft parts, and the Ballmer-Gates company is the only one in the console segment following through with such a strategy.
The Xbox 360 is scheduled to ship in an "all 65 nm" package (Jasper platform) this August. Consumers won’t notice the refresh, unless Microsoft decides to put a Blu-ray drive into the Xbox 360. We were not able to receive a confirmation either way, and we keep digging to find out if the Asustek subsidiary Pegatron will manufacture regular Xbox 360’s or units with an integrated Blu-ray drive. In any case, Celestica, Pegatron and Wistron will have a busy summer cranking out millions of refreshed Xbox 360 consoles.
A more dramatic and perhaps visible change will happen next year: TSMC plans to begin producing the Valhalla chip, which will be the foundation of the mid-cycle refresh of the Xbox 360, thus called ’Xbox 2.5’ or simply ’Xbox 540’ (360+180), in fall of 2009. We learned that this new chip is apparently much more than a die-shrink and end up as a system-on-a-chip design. This change is likely to enable to redesign the Xbox 360 casing and go towards a slim-design, much like what Sony did with the Gen1 and Gen2 PS2. We believe that TSMC will use a 45 nm process for this Multi-Chip-Module package (CPU+GPU+eDRAM).
There are also some interesting pieces of information that Microsoft is shopping for a more efficient cooling solution - efficient in more ways than just one: Several people close to the cooling industry told us that Microsoft approached them and asked for better and cheaper cooling than what is used in the Xbox 360 right now. Some may claim that the current Xbox 360 cooler design is already as cheap as it gets, but we have no doubts that Microsoft will find a way to drop the cost once again.
AMD six-core CPU prepared for second half of 2009
Build on 45nm and bearing the 'Istanbul' codename, AMD's six-core part will be introduced in the second half of 2009, will feature HyperTransport 3.0 and will be compatible with Socket (F) 1207 boards.
Trying to catch up with Intel, AMD is planning to release yet another 6-core CPU in the first half of 2010, this time for a new Socket - G34. Dubbed Sao Paolo, the CPU will have DDR3 support and battle Intel's 32nm hexa-core processor which should be released around the same time.
Intel X58 to be the first Nehalem desktop chipset
Whether AMD likes it or not, Intel sill be introducing the first 45nm-built Nehalem processors in Q4 and with them, a whole new range of silicon chips will arrive. One of them, the Tylersburg chipset, has now been exposed to bearing the X58 marketing name. Set to be paired up with the ICH10 southbridges, the X58 will be used on Socket 1366 motherboards and act as the assistant of the quad-core Bloomfield processors.The platform based on the X58 and Bloomfield couple will provide triple-channel DDR3-1333 support, two or four PCI-Express 2.0 x16 slots and will act as Intel's top desktop offer. Just about the same time the X58 rolls out AMD should start delivering its first quad-core 45nm Phenom CPUs so expect this year to end with quite a clash of titans. Well, it'll be more David and Goliath.
AMD to stockholders: Everything will be all right
In looking back, 2007 was a difficult year of transition for AMD. And I speak for every AMD employee when I say that I could not be more disappointed with our financial results. The Barcelona processor delay really hurt. But that is behind us. As a matter of fact, we've learned some lessons very important to us. In our materially cyclical industry, we have learned to anticipate and prepare for the inevitable downturn. And that's exactly what we did. We have a healthy cash balance sheet approaching $2 billion, and we have some financial cushioning in place if the macro-economic environment continues to disappoint, particularly in North America.
Ruiz reiterated that AMD is on track to become operationally profitable again in the second half of the year, and he called 2008 "a year of opportunity." He also explained that the company is taking steps to avoid a repeat of the past few quarters:
While our short-term goal is to achieve operational profitability in the second half of this year, our long-term goal is to achieve consistent profitability in good times and bad times. To that end, we are re-architecting the business so that our financial success is not invariably dependent on continuous component performance and leadership.
In order to keep AMD's head above the water, Ruiz said, "[we will be] reducing our break-even point by several hundred million dollars so that we will be profitable on our current revenue stream." In other words, if the previously announced job cuts and restructuring work out, AMD may not need to double its market share to be profitable again.
Speaking of restructuring, Ruiz also had a little more to say about AMD's long-awaited "asset smart" strategy. As part of the strategy, Ruiz said, AMD will "deploy [its] manufacturing assets to most cost-effectively stay at the leading edge and deliver customer value." While vague, the statement further hints that AMD might spin off its manufacturing business as a separate entity.
Did Ruiz's promises successfully convince stockholders? That's hard to say: the only question after his monologue came from an elderly Austin resident, who talked for several minutes in a slow, baritone drawl about golf tournaments.
©THE REGISTER
12-Core Chips from AMD in 2010
Intel To Launch 4-series Chipsets At Computex 2008
More here at Digitimes
Intel: Skulltrail Supports Crossfire & SLI
"For the team creating world-class games here at id, time is one of our most valuable assets," said Robert A. Duffy, Programming Director, id Software. "Having eight powerful Intel cores in a single machine helps our team create and test our latest titles at record speed. We have seen one of our most time-consuming asset generation processes cut from over 4 hours to under 20 minutes by utilizing all eight cores and threading the generation code. Long term this translates to better games on the market faster than previously possible."
The Intel Core 2 Extreme processor QX9775 is available now at an MSRP of $1,499 each. Estimated street pricing for the Intel Desktop Board D5400XS is $649. Platform components are sold separately. Several enthusiast PC manufacturers plan to offer systems based on this new platform starting today and over the next 30 days, including Armari, Boxx Tech, Digital Storm, Falcon Northwest, Maingear, Puget Systems, Scan, Velocity Micro, Vigor Gaming, Voodoo, @Xi Computers and others.
AMD: Promises, promises
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 DX10 Drivers; NVIDIA Laughs
However, NVIDIA has been reportedly been sending tech journalist tidbits on the performance of Intel’s DX10 update – mainly that there is no performance gain.
NVIDIA ran the usual benchmark of games under DX10 settings and to no surprise found Intel’s integrated graphics to be “downright incapable” of being used with DX10 applications.
NVIDIA goes on to say, “Saying GMA 3500 is DirectX 10 capable is kind of like saying Styrofoam is "nutrition capable". I guess Intel’s definition of capable is a lot different than our definition... a lot.” NVIDIA may be patting itself on the back for drawing that conclusion, but PC enthusiast have widely linked Intel’s IGP to poor performance for quite some time now. While NVIDIA should not be worried too much about Intel upping the performance on its IGP, they should be worried about Intel’s upcoming Larrabee project.
Larrabee, due in 2010, is a multicore discrete graphics initiative from Intel based off of the x86 architecture and supports OpenGL and DirectX instructions. Larrabee, has drawn much interest from the technology community and is one of the most anticipated product launches in the industry.
Jen-sun Huang, NVIDIA CEO, and the rest of NVIDIA have been on the initiative of opening “a can of whoop ass”, before Larrabee takes off. This latest act from NVIDIA appears to be just another example of the company’s aggressive PR plan — pushing the consumer to link poor graphics performance with Intel. Conversely, Intel previously stated in a press call that it thinks computer graphics is hitting a severe bottleneck and that is vector based solutions coming down the line will be the way of the future for 3D.
source:http://www.tomshardware.com/
AMD Announces 6 And 12-core Opterons
When Intel rolls out its Nehalem processors later this year, AMD appears to be better prepared than in late 2006 when Core began its triumph and until today pretty much wiped out all of the gains AMD Opteron has made in previous years. On the server side, AMD will counter Intel with the Shanghai quad-core CPU, which will be based on the current Barcelona technology, but will be manufactured in 45 nm and bring performance and power consumption benefits.
According to vice president Randy Allen, Shanghai will deliver about 20% more speed than Barcelona. While the chip will remain compatible with the current Socket F (1207) platform, the L3 cache will triple to 6 MB, 3-lane HT3 support will be integrated and DDR memory support will be extended to 800 MHz devices. AMD claims that the idle power consumption of the new CPUs will be 20% below their 65 nm counterparts.
Production of the 45 nm chip, which has been sampled to server vendors, is scheduled to begin late this year, while first products should be surfacing in Q1 2009. In the second half of that year, AMD plans to release a single-die 6-core Socket F processor code-named Istanbul. Offered for two- and four-socket systems, the chip will be clocked at lower clock speeds than the quad-cores.
A completely new platform (read: a departure from Socket F) is planned to be introduced in 2010. Called the "Maranello" platform, AMD will then transition to DDR3 memory and four HT3 lanes. The multi-socket processors will include the monolithic 6-core "Sao Paulo" with 6 MB of L3 cache, new hardware throttling and power management features, as well as the dual-die 12-core "Magny Cours" processor with 12 MB L3 cache. Both processors will be built in 45 nm. At least at this time AMD does not anticipate any Nvidia chipsets to support this platform and says that it will couple the CPU with its own RD890S and RD870S northbridges as well as the SB700S southbridge.
Eagle-eyed readers may wonder about those production dates. A late 2008 45 nm volume production of these processors matches the information we received from industry sources and would mean that AMD will put 45 nm into production just about one year behind Intel. Intel plans on introducing 32 nm processors in late 2009 and AMD now indicates that its Sao Paulo and Magny Cours processors will be manufactured in 45 nm in the first half of 2010, leading us to believe that 32 nm won’t arrive until the second half of the year in a best case scenario. This could mean that AMD keeps up with Intel, but won’t gain any ground on Intel in this discipline anytime soon.
Randy Allen responded to our inquiry by saying that this assessment would not be accurate as server processors do not reflect the entire product range of AMD. However, at least for 45 nm, we have credible sources indicating that 45 nm AMD CPUs won’t arrive before Q4 of this year. The 2010 processors are too far out to speculate whether Intel will deliver them at the end of 2009 and whether AMD will have any 32 nm chips available in Q1 2010. But clearly, at least for now, the 2006 promises AMD made to analysts aren’t likely to be kept.
Until Shanghai arrives, AMD will have to survive with its Barcelona quad-core and it is not surprising that the company touts the power-efficiency drum to compare itself favorably to competing Intel products. Allen claimed that AMDs 75 watt part Opteron 2356 saves about 20% of power when idle, 14% in high-load scenarios and 10% under full-load when compared to Intel’s (45 nm) 80 watt Xeon 5440. The executive also noted that the Opteron delivers on average 13% more performance than its Intel counterpart.
Those claims are based on five SPEC benchmarks as well as two fluid dynamics tests. It often depends on a very specific need that exploits a very specific capability of a processor as the results in each of those benchmarks can change dramatically. While AMD claims a 19% lead in SPECompMBase2001, a 13 and 19% in SPECfprate_base2006 and SPECfprate_2006, we were not able to verify the latter two results. In fact, official SPEC results indicate that at least one Intel Xeon 5440 systems is 29% and 26% ahead of the best performing Opteron 2356 system listed.
source:http://www.tomshardware.com
Intel, SGI To Build 10 PFlops Supercomputer For NASA
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
Intel delays next-gen integrated chipsets to fix video bug
When does an Intel chipset launch? When it's first mentioned by the company in public? The next time? The time after that? The 'Eaglelake' chipset family has been discussed by Intel on several occasions, but it'll apparently be truly launched next month.
The Eaglelake line - aka the 4 series - comprises the P45, P43, G45, G43, Q45 and Q43, and will be formally launched at the Computex show, held in Taipei in early June.
No great surprise there: Intel has already confirmed a Q2 launch date, and Computex is the customary location for launches of this kind.
More interestingly, the chips will go out to motherboard makers in the third week of June, industry moles have told local newssite DigiTimes, which probably means mobo availability in July.
That's for discrete, 'P' parts - integrated, 'G' chipsets could arrive up to three weeks further down the track thanks to a glitch in GPU hardware that decodes the VC1 hi-def video codec, the site claims. Ditto the enterprise-oriented, 'Q' parts, which are being delayed pending a fix for a Virtualisation Technology bug.
The Eaglake series is designed to be PCI Express 2.0 compliant and support 1333MHz DDR 3 memory. Some integrated versions will be able to host DisplayPort monitor connectors.
© The Register.
AMD Phenom™ X4 9000 Series Quad-Core Processors
INCREDIBLE PERFORMANCE
The ultimate megatasking experience. Featuring true multi-core design and award-winning AMD64 technology with Direct Connect Architecture, AMD Phenom™ processors deliver the ultimate megatasking experience by providing direct and rapid information flow between processor cores, main memory, and graphics and video accelerators. AMD Phenom™ processors have the technology to break through the most challenging processing loads. AMD Phenom™ processors feature low latency access to main memory for amazingly rapid response and phenomenal system performance. AMD Phenom™ processors were designed for megatasking—running multiple, multi-threaded applications. Surge through the most demanding processing loads, including advanced multitasking, critical business productivity, advanced visual design and modeling, serious gaming, and visually stunning digital media and entertainment.
Phenomenal performance with advanced processor design. The AMD Phenom™ processors are the most advanced processors for true multitasking with true quad-core design. Don’t get bogged down by non-native quad-core processors and obsolete front side bus architectures. With an integrated memory controller and shared L3 cache, AMD Phenom™ processors have low-latency access to main memory for amazingly rapid system response and phenomenal system performance.
Blast through performance bottlenecks. All AMD Phenom™ processors feature AMD64 with Direct Connect Architecture to blast through performance bottlenecks. Award winning HyperTransport™ 3.0 technology just got faster, providing support for full 1080p high-definition video and extreme total system bandwidth.
Shatter the memory barrier. Superior AMD64 architecture offers direct access to DDR2 memory. Enjoy virtually unlimited memory options with AMD64 technology and 64-bit Windows Vista.® Shatter the memory barrier with AMD Phenom™ processors and 64-bit Windows Vista.®
INTENSELY VISUAL
Experience Windows Vista.® Harness the power of Windows Vista® with the AMD Phenom™ processors. AMD Phenom™ processors divide and conquer the most complex tasks with true multi-core design. Enjoy the ultimate megatasking experience on Windows Vista.® Enjoy virtually unlimited memory options with AMD64 technology and 64-bit Windows Vista.® Shatter the memory barrier with AMD Phenom™ processors and Windows Vista.®
STRIKINGLY EFFICIENT
Strikingly efficient Cool‘n’Quiet™ 2.0 technology.With the next generation of award-winning Cool‘n’Quiet™ technology, Cool‘n’Quiet™ 2.0 technology reduces heat and noise so you can experience amazing performance without distraction. Combined with core enhancements that can improve overall power savings, AMD Phenom™ processors deliver seamless multitasking and optimum energy efficiency. Work, play, talk, and share a PC that’s seen, not heard.
Purchase with Confidence
Founded in 1969, AMD has shipped more than 240 million PC processors worldwide. Customers can depend on AMD64 processors and AMD for compatibility and reliability. AMD processors undergo extensive testing to help ensure compatibility with Microsoft Windows XP, Vista, Windows NT®, Windows 2000, as well as Linux and other PC operating systems. AMD works collaboratively with Microsoft and other ecosystem partners to achieve compatibility of AMD processors and to expand the capability of software and hardware products leveraging AMD64 technology. AMD conducts rigorous research, development, and validation to help ensure the continued integrity and performance of its products.
source:amd.com
AMD Phenom X3 processor family performance
AMD known for its fighting underdog spirit recently relaunched the Phenom with the B3-stepping, which we looked at just recently. The B3-stepping has not exactly breathed new life into the Phenom processors, but it has given AMD a chance to regroup and retarget its products, forcedly aiming at cheaper and more affordable Intel processors.
The latest processor series from AMD is a little unusual as it makes use of not one, two, or even four cores, but rather three! That rsquo;s right, the new Phenom X3 carries an unusual core configuration, and I guess the question most of you are probably asking yourselves (as we did) is why? The most reasonable explanation is that this still allows AMD to sell Phenom X4 processors with a defective core, minimizing their loss. Second, it gives some leverage for AMD to compete with Intel rsquo;s dual-core processors, being able to pull the "additional core" card.
The Phenom X3 family consists of three new processors: the 8450 (2.10GHz), 8650 (2.30GHz) and 8750 (2.40GHz), all of which we are testing here today.
View: AMD Phenom X3 processor family performance @ TechSpot
Read full story...
AMD Must Double Processor Market Share to Survive
Advanced Micro Devices needs to more than double its share of the microprocessor market to survive, according to a brief filed by the company's lawyers in its antitrust lawsuit against Intel.
At the end of 2007, AMD had 13 percent of the processor market, "less than half of what it requires to operate long-term as a sustainable business," the brief said, explaining that Intel's alleged efforts to shut the company out of the processor business had largely succeeded.
"Measured on a revenue share basis, AMD made little progress growing its slice of the pie," it said.
View Full Article: PC World
AMD now has 'more than allegations' against Intel
As Intel and AMD near the end of the discovery process in their US antitrust battle, the two companies have begun fighting over whose testimony will make it to the big dance. In a legal filing, AMD has pointed to the employees at some of the technology world's biggest names - HP, Dell, IBM and others - who it thinks will help make its case. Intel has responded in kind, and it's now up to a judge to decide on the strength of the vendors' arguments.
Following a dispute over the number of depositions allowed in the case, Special Master Vincent Poppiti appointed to hear evidence from both companies ordered up a pair of "preliminary pre-trial briefs." Well, it's those very briefs which have now been turned over to the press.
AMD's beefy filing claims to offer new evidence to support the original accusations filed back in 2005, at the U.S. District Court in Delaware.
"The current brief reflects that the allegations are more than allegations," said AMD spokesman, Michael Silverman. "The evidence exists."
Unfortunately, most of this purported evidence has been redacted from the public version of the brief. Both companies had previously signed a confidentiality order to guard trade secrets, making for long, tantalizing passages buried behind this kind of nonsense and more of it in AMD's latest 108-page summary.
What can be gleaned from the filings are the companies that AMD claims have been influenced by Intel's anti-competitive behavior.
AMD fingers a laundry-list of major OEMs: Dell, Hewlett-Packard, IBM, Lenovo, Gateway, Acer, Sony, NEC, Toshiba, Fujitsu, and Hitachi as holding anti-competitive, exclusive arrangements with Intel. It also calls for testimony from smaller builders such as Supermicro and Rackable. All the sections covering these vendors are heavily censored, however, making it hard to grasp anything beyond that AMD claims to have acquired specific evidence of misconduct.
For instance, in the section about Dell:
How to destroy 60 hard drives an hour
Disk drives are extraordinarily resilient. I've seen one that had been run over, dropped in a toilet and then thrown out of the window of a seven-storey office block. It looked unusable. Kroll Ontrack managed to recover virtually all the data on it by drying it and taking it apart.
Hard disk crusher
Clam shell casings are very strong and deleting data doesn't work effectively, and passing magnets over the disk isn't sure enough. Even repeated writing of zeroes isn't necessarily effective and certainly takes a long time.
So, if you really, seriously, want to render your hard drive unreadable and haven't used major league encryption of all its contents then what can you do?
Drill the little sucker right through its spindle heart is one answer.
EDR Solutions sells a machine to do just this. It's inaccurately called the Hard Disk Crusher but it doesn't crush the drive. Instead it drills through the spindles and this creates ripples in the platters rendering them effectively unreadable.
It can destroy the readability of disk drives at the rate of 60 an hour, taking 10 seconds once the disk is clamped in place. One EDR customer rendered 800 unreadable in just one day.
The machines costs $11,500, which includes a one-year warranty (return to factory).
More information hereNext-Gen PC Concepts Include Laser Screens, Tactile Interfaces, Glossiness
The Microsoft-sponsored Next-Gen PC Design Competition asks its contestants to design the next generation of Windows-based PCs and judges the resulting entries on innovation, user experience, aesthetics and whether it can ever be manufactured. This year's contest focused on PC designs that "help people do what they feel passionate about." Public voting is over and the winners will be announced sometime this month, but even if we can't participate in the voting process, a lot of these concepts are worth gawking at. Check out some weird and interesting prototypes after the jump.ngpcentry1
Palette is a hardware system design concept that designated for digital artist, painter, design visualizer and people with passion in visual arts. The system were designed based on the background of basic painting behavior and its physiology activities. Palette is designed to have connectivity with other devices such main PC system, Main monitor or other wireless peripherals. connectivity between them are wireless and the user interface screen surface of the palette is using photo-optic visual sensor with color recognition as its main advantage.
ngpcentry2
A biometric scan of a finger will recognize users and display all saved personal files and settings. As an optional security feature, users will be capable of enabling a voice recognition program with a library distinct voice commands. Once access has been granted, cutting-edge laser technology will display mid-air 2D and 3D images with pristine quality on a dual-head display, ideal for designers.
Bloom’s two “side branches” house the PC’s sound system and assist in the display of both 2D and 3D images. The branches allow the displays to range from 15 to 24 inches in size with a maximum viewing angle of 80 degrees. Users can effortlessly choose from a variety of display sizes and viewing angles by re-positioning the branches.
ngpcentry8
Momenta, the neck worn PC also captures the best and most exciting moments of your life. Ever thought, "Man I wish I had that on tape!" When everybody is laughing themselves to tears, Momenta has already captured the previous 5 minutes from its rolling buffer and continues to record until you tell it to stop. Triggered by increased heart rate, it captures those hilarious or exciting moments that are usually lost forever. Whether it's an exciting sports experience, a funny social scene, the scene of an accident, etc. you can capture it and share it.
ngpcentry7
ZEN is a system of software and hardware specifically designed for the visually impaired. It is comprised of a hardware called the ‘Sandbox PC’ and ZEN edition operating system which work together to create a computing environment that can be used with eyes closed. The innovation of ZEN is embedded in its name. ZEN is for ‘Z-axis Enabled,’ meaning it is a truly interactive 3D computer. The Sandbox PC features Active Surface, an electronic surface that can take different physical shape and can be manipulated by the user. You can read Braille text on it, make out windows, icons, and different textures. You can press into the surface, or scratch it with your fingers. All input output is made with the sense of touch. ZEN software uses a Tactile User Interface (TUI). The visually impaired are a negated market. It is fitting, then, to bring them to light using the negated sense of touch, by introducing a negated dimension of depth.
[Next-Gen Design Competition
