All the latest quantum computer articles

See the latest stories on quantum computing from eeNews Europe

Thursday, November 26, 2009

Printable programmable logic

Nano ePrint in Manchester has secured a grant for Research and Development from the UK's Northwest Regional Development Agency (NWDA) for a £390k project to demonstrate the world's first printed programmable logic device.
This will be based on developments led by Ian Barton, previously Managing Director of Silistix, who joined Nano ePrint as Vice President of Engineering in May. Ian brings significant experience in the management of integrated circuit design and is leading Nano ePrint's engineering team in the design and production of printed programmable logic devices. Silistix, another spinout of the University of Manchester, develops network on chip technology.
Nano ePrint was formed in 2006 following pioneering research at the University of Manchester in the UK that allows planar nanoelectronic circuits to be built in a single layer of semiconductor via single-step patterning. Nano ePrint’s devices achieve very high speeds while dramatically reducing circuit size and simplifying manufacture. Nano ePrint has also developed a configurable circuit architecture to deliver the world’s first printed programmable logic circuits, allowing one design to simultaneously address multiple applications, opening up prospects in low-cost, high-performance printed electronics across a wide range of markets.

First 1.1in QVGA LED display

OKI Digital Imaging has developed a 1.1 inch QVGA LED display while reducing power consumption to one-tenth that of conventional LCDs.
This is the direct result of development in its LED print head technology that enhances luminance efficiency using an extension of its proprietary Epi Film Bonding technology into a two-dimensional array.
This array has the world’s first 65-micrometer pitch between LED chips to give a high definition image and, as it is self-illuminating, the new display has a faster response, smaller size and lower power than LCD displays that require a separate light source. It also features high contrast (over 5000:1) for high visibility even in bright daylight.
OKI Digital Imaging will manufacture the LED display at the LED manufacturing facility in Gunma prefecture, Japan, that it acquired from Reneasas back in July. Shipments of samples are expected to begin by the end of fiscal year 2010.
“The new technology has made it possible for us to develop a groundbreaking LED display that provides high definition images while maintaining compact dimensions,” said Hiroshi Kikuchi, President of OKI Digital Imaging. “The development of the world’s first thin-film-bonding technology in 2006 marked a significant milestone for the OKI Group. As shown by today’s announcement, that technology laid the foundations for continuing developments in LEDs. The OKI Group plans to leverage the expertise and technical knowledge accumulated over 20 years of experience with LED printing to develop new technologies in other areas and to expand its LED-related businesses.”

Altium targets Poland on software piracy

Design tool vendor Altium is targetting illegal use of its software specifically in Poland, offering an amnesty for just one month for existing users.

Sets December deadline for illegal users

“Piracy of Altium software, both of Altium Designer and older Protel software, is a known problem throughout Poland. With this programme, Altium helps illegal Protel users continue with their favourite EDA tool for another half a year without having to fear legal consequences,” said Jean-Paul Seuren, Licence Compliance Manager EMEA, Altium Europe.
“At the same time, and of much more value to these users, we will help them to move their old Protel projects and legacy data to Altium Designer. Designers will see the benefit from using this next-generation electronics design software and increase their competitiveness in the Polish and worldwide electronics design sector. This software amnesty programme helps illegal users take the first step to battle software piracy.”
Under the new scheme, companies doing electronics design can legalize their illegal copies of Altium’s Protel and Altium Designer software by buying the latest version of Altium Designer at a price of €3,200 that includes half a year amnesty for each legalized seat.
Earlier this year Altium lowered the pricing for Altium Designer to €3,200, a reduction of nearly 70% designed to make electronics design software affordable to every electronics designer. This highlights the key reason why designers were using the tools illegally. The amnesty programme extends that new pricing by adding the support Polish users need to become up-to-date in both compliance and their software.
During this amnesty period, designers can retain their illegal Protel seats to migrate their legacy data to Altium Designer. Altium will also provide free software assurance and support, webinars as well as special migration training. Users will receive an official certificate that enables them to use one unlicensed Protel seat legally for six months for each Altium Designer licence purchased.
Altium’s Value Added Reseller Evatronix SA, based in Bielsko-Biala, will be the local point of contact in Poland to facilitate the software amnesty programme.
Altium Designer is the successor of the well known and widely used Protel PCB design tool but takes a very new and different approach to electronics design. In a single solution, using a single data model, it lets designers create products in all three design domains: PCB layout, programmable hardware (using FPGAs) and software.

Lego robots at Robocon 2009

I can't resist robots, it's just the kid in me! The Annual Robocon Competition took place in Yokohama, Tokyo on November 18. 40 teams from all over the world were competing in various disciplines, including the LEGO MINDSTORMS, in which the Royal Danish Embassy in Tokyo sponsored one of the 'Innovation Prize' awards.

Wednesday, November 25, 2009

USB3 solid state disk from OCZ

New names in the industry are emerging with USB3.0 coming through (despite the problems at Intel) with OCZ Technology of San Jose teaming up with Symwave to show an external high speed solid state disk (SSD) drive.

Symwave is a fabless chip supplier of SuperSpeed USB3.0 devices with headquarters in Orange County, Calif., and design centers in Shenzhen, China, and San Diego, Calif. “Thanks to Symwave’s industry leading USB 3.0 storage controller, our external SSD device delivers 10x the transfer rate of USB 2.0 at 5Gb/s, as well as several ‘green’ improvements including superior power management and lower CPU utilization,” said Eugene Chang, Vice President of Product Management at the OCZ Technology Group. “OCZ’s products are synonymous with high performance. Combining our USB 3.0 controller with their SSD technology, both architected for performance without compromise, is a perfect match,” said John O’Neill, Vice President of Marketing at Symwave. “CES will be a showcase for new and innovative products that provide consumers with a dramatically improved user experience. Symwave is committed to delivering silicon and software solutions that will transform the way consumers view and use USB storage devices.”
Interestingly, Symwave has seen an investment by networking leader SMSC alongside venture capital firms including Kodiak Venture Partners and CMEA Ventures.

OCZ has just launched the Colossus 3.5” Solid State Drive (SSD) Series, designed to meet the performance and storage demands of high-end desktop users and enterprise clients. Available in up to one terabyte with RAID0 configurations, the Colossus Series marks a milestone in solid state drive technology, making it possible to have both high performance and high capacity in one solution.
“The new Colossus Series is designed to boost desktop and workstation performance and is for high power users that put a premium on speed, reliability and maximum storage capacity,” said Chang. “The Colossus core-architecture is also available to enterprise clients with locked BOMs (build of materials) and customized firmware to match their unique applications.”
The Colossus SSD is available in 128GB, 256GB, 512GB, and 1TB with 1.5 million hour mean time between failures (MTBF).

Atom moves into automotive

Kontron's Microspace MPCX28: in-vehicle PC with Intel Atom Z530 processor

Kontron has launched its first in-vehicle PC with extended support, based on the Intel Atom Z530 processor and Intel System Controller Hub US15W, with a flexible PC/104 socket for ultra flat and space-saving PC/104 expansion cards. Further options for customization include 2 PCIe Mini Card Slots for custom specific options such as GPS, GSM/UMTS, and WLAN.
Its fanless design, easy expandability and high flexibility make the precision Swiss-engineered Kontron Microspace MPCX28 Box PC a perfect fit for various applications in busses, cars, trucks, and (automated) heavy-duty vehicles used in intralogistics. In cabs and busses it can be used for passenger infotainment, video surveillance, or ticketing systems. Typical truck applications include CAN-based trailer and goods management, GPS navigation combined with communication via GSM, and UMTS/GPRS for fleet management. For vehicles used in intralogistics, the applications range from the control of automated vehicles, to logistics for forklift trucks with WLAN and barcode/RFID interfaces.
Designed to work in the extended temperature range -25 to 70°C, the passively cooled in-vehicle Kontron Box PC is based on the Intel Atom Z530 1.6 GHz processor with 533 MHz FSB and supports up to 1 GB DDR2 RAM. The embedded graphics feature 256MB of memory DirectX 9 3-D and HDTV support with MPEG2 and H.264 decoders to process compressed HD videos to offload CPU usage. To accommodate storage devices, it provides front side slots for a 2.5" SATA HDD or SDD and Compact Flash media. Thanks to the externally-accessible slots, the change of storage media is quick and simple without the use of tools. This makes it particularly easy to swap out the hard drive within seconds to provide new content for, e.g., infotainment applications. The internal PCI/104 and 2x PCIe MiniCard expansion slots allow for various application-specific extensions, such as Dual-CAN, GPS, GSM/UMTS, and WLAN.

Application specific expansions via ultra flat PC/104 extension cards

The Microspace MPCX28 also features a comprehensive set of interfaces to fulfill the demands of nearly all in-vehicle applications. On the front, two USB 2.0 interfaces and 3.5 mm jacks for stereo output and input are accompanied by a SIM slot for mobile communications. On the back, the new in-vehicle PC offers 2 x Gigabit Ethernet, 2 additional USB 2.0 ports, as well as an SATA port to connect additional storage devices. The DVI-I interface transmits analog or digital video signals to external monitors with a resolution of up to 2,048 x 1536. Two DSUB connectors for COM ports (RS232) and a parallel interface for digital I/O, which provide CAN-bus connectivity via an optional PCI/104 expansion card, round off the feature set.
With a wide-range power supply (8 to 58 VDC), the PC can be directly connected to the vehicle's power supply without additional converters. A plug-in power supply for 110/220 VAC is also available. The system supports remote power on/off and can accommodate customer-specific functions such as wake on ring, event wake up or a wake on LAN.
It is scheduled to be available (with CE and E1 certification) in Q1, 2010 and supports Windows XPe/XP/Vista/7, and Linux.

SME survey of UK companies

The UK's Electronics Leadership Council, distributor Premier Farnell and The UK Electronics Alliance are co-ordinating a survey to gain knowledge from technology-based SMEs on how to better advise government to direct the necessary support for greater economic growth.

The survey is here

All UK based small and medium enterprise (SME) technology companies are encouraged to take part in a survey through to 9th December 2009 developed to garner appropriate support from the UK Government and trade associations to ensure the necessary future economic growth for these SMEs.
The intent of the survey is to gain support to influence regulation and legislation that will encourage success for SMEs, ensuring their views are well understood by government as decisions are made. The findings will also be used to attract funding for business in the UK and develop targeted programmes that make a positive contribution. The ELC and UKEA have excellent lines of communication already established with government that will ensure the issues raised by the survey get directly to those most able to influence change.
A second stage of focus group consultation is planned in early 2010 involving interested survey participants. It is hoped that as many SME’s as possible will complete the initial survey to ensure it will be truly representative of opinion and make a powerful contribution towards shaping the decision making processes of government.
All participants in the survey will have the opportunity to receive a copy of the results once the survey is concluded. All trade associations that participate in the survey will also receive a final copy of the survey findings to distribute to their members.
“In recent months we have met with a number of key government leaders and industry bodies, all of whom want to stimulate growth, innovation and enterprise in the UK. Our industry is critical to leading the market out of the downturn and affecting change on a global scale as the pervasion of technology continues in every aspect of our lives,” said Harriet Green, CEO of Premier Farnell.

MEMS microphones for a dollar from ST and Omron deal

ST Microelectronics has teamed up with Omron to drive the cost of micromachined, highly integrated silicon MEMS microphones.
The deal will see the devices falling to under $1, says ST, using Omron's sensor technology. ST is already the leading supplier of MEMS for consumer and portable applications and the third largest MEMS manufacturer overall. , but will use Omron's MEMS sensor technology.
The key is that ST is the only manufacturer that can mount and package multiple MEMS microphone devices at a time, facilitating high-volume production, using a proven packaging technology very similar to that used for motion sensors.
MEMS microphones can be made smaller than the most compact microphones and are less susceptible to mechanical vibration, temperature variations and electromagnetic interference, which is important in cell phones and other devices with an audio input, such as notebook computers, video recorders, digital cameras, as well as hearing aids or electronic stethoscopes.
“In the past, microphones were the domain of expert acoustics companies, but now it’s time for semiconductor MEMS players to drive the growth of this market. We’re aiming to increase the size of the MEMS microphone market by an order of magnitude,” said Benedetto Vigna, Group Vice President and General Manager of MEMS and Healthcare Division, STMicroelectronics. “This market can explode only with big and long-term committed suppliers, operating their own leading-edge MEMS fabs. Working together with our Japanese friends, we'll drive the microphone market growth as we have done in motion sensors.”

Samples of digital MEMS microphones that integrate ST’s electronic control circuit and OMRON’s micro-machined sensor in a single package will be available before the end of this year, at less than one dollar for large-volume orders.
According to iSuppli’s research dated September 2009, the market for micro-machined acoustic devices for consumer electronics and mobile handsets is forecast to grow at a revenue CAGR of 18% between 2008 and 2013, when it will exceed one billion parts per year.
MEMS microphone startup Akustica was acquired earlier this year by Robert Bosch, while another MEMS startup, Oligon, was bought by Wolfson Microelectronics to develop a range of highly integrated microphones.

Tuesday, November 24, 2009

Instant products from rapid prototyping

Design tool vendor Altium has added an instant deployment option to its FPGA development board. Designers can now take their FPGA-based designs from concept right through to deployment without the need to create a custom PCB by clipping the NanoBoard 3000 into a new range of enclosures designed in-house by Altium.
The modular enclosures let designers deploy FPGA–based designs created and hosted on the NanoBoard 3000 in a number of different ways: on desks, on walls, in either commercial or industrial locations, in extended options for multiple board designs, with or without the TFT display that comes with the enclosures.
Instantly deploying the NanoBoard 3000 in the new modular enclosures gives designers the option of creating small production runs in commercially attractive cases, again without being compelled to manufacture custom enclosures or custom PCBs.
Altium’s NanoBoard 3000 is a programmable design environment supplied complete with hardware, software, ready-to-use, royalty-free IP and a dedicated Altium Designer Soft Design license. This is a complete design environment that lets electronics designers construct sophisticated ‘soft’ processor-based systems inside FPGAs without any prior FPGA expertise, specialist VHDL or Verilog skills. Designers simply use their existing board layout and systems design skills to construct, test and implement FPGA-based embedded systems. The supplied Altium Designer license lets them select, drag and drop the large range of IP blocks to add processors, memory controllers, peripherals and software stacks. Unlike alternative programmable design environments, designers no longer need to search the web for drivers, peripherals or other software, and then have the hard work of integrating all these elements to make them work together.
The new NanoBoard 3000 enclosures mean that designers can now start with a purely 'soft' prototype on the NanoBoard and then deploy it into the field. But Altium Designer’s unified architecture also gives them the option of upgrading to a board-level Altium Designer license and moving into custom PCB design. Their 'soft' design work completed on the NanoBoard 3000 is simply ready to be used on their custom PCB.
Weeks are saved on getting high-quality proof-of-concept prototypes designed without any custom PCB design work required at all. Newcomers to FPGA design have a low-risk, low-cost design environment that has everything they need to get started. Experienced FPGA designers can use their expertise in new ways to focus on creating the intelligence of their products.
Altium’s first NanoBoard 3000 features a Xilinx Spartan 3AN FPGA. Two more NanoBoards, featuring Altera and Lattice FPGAs, are in manufacture and design. The new deployment enclosures at Euro99 will accommodate all these NanoBoard options.
The NanoBoard 3000 is available for a recommended retail price of $295 and includes a 12-month subscription to the Altium Designer Soft Design License.

Friday, November 20, 2009

Namecalling Begins Around Intel's USB 3.0 Slowdown, Slapfight Imminent

Namecalling Begins Around Intel's USB 3.0 Slowdown, Slapfight Imminent

Posted using ShareThis

DDR3 memory transistion next year

Shipments of the next generation high speed, low power DDDR3 memory will rise to account for more than half of the global DRAM market by the second quarter of 2010, surpassing DDR2 for the first time as the leading technology for PC main memory says market researcher iSuppli. DDR3’s share of the DRAM market in terms of gigabit-equivalent shipments will rise to 50.9 percent in the second quarter of 2010, up from just 1 percent in the second quarter of 2008 and 14.2 percent in the second quarter of 2009. By the end of 2010, DDR3 will grow to account for 71 percent of shipments.
“DDR3 is 50 percent faster than today’s dominant DRAM technology, DDR2, while using about 30 percent less power,” said Mike Howard, senior DRAM analyst for iSuppli. “For PC users across the board, this means faster performance. For notebook users, it can result in longer battery life.”
Two major factors are driving the industry’s transition to DDR3: new Intel Corp. microprocessors and the increasing manufacturability of the part.
Intel’s latest-generation microprocessor microarchitecture, dubbed Nehalem, employs a memory controller that supports only DDR3, unlike the previous Penryn line, which works with both DDR2 and DDR3. With Intel transitioning its microprocessor line to Nehalem-based chips, PC makers will have no choice but to migrate to DDR3 SDRAM.
On the supply side, production of DDR3 has advanced to the point were memory makers can produce it at a competitive cost using leading-edge semiconductor process technology. Because of this, all major DRAM suppliers now are producing DDR3, making the memory more attractive to PC makers who are loathe to be beholden to a single source.
“With DDR3 commanding higher pricing than DDR2, memory makers realize where the big money will be in 2010,” Howard said. “Because of this, they are more than willing to transition production to the new memory technology.”
As availability has increased and prices for DDR3 have fallen, the cost of DDR2 to PC makers actually has risen in recent months because of supply constraints. And despite the rapid rise of DDR3, DDR2 is expected to command significant volumes throughout 2010, accounting for 15.4 percent of gigabit-equivalent shipments the fourth quarter.
For more information on this topic, see iSuppli’s new report: DDR2 to DDR3 – The Anatomy of a Technology Transition.

Thursday, November 19, 2009

Atom module adds dual GbE, SATA and PCIe

Anders Electronics in London has developed a module that uses Intel's Atom processor with two Gigabit Ethernet connections for faster connectivity. Measuring just 65x75mm, the new CM-iAM SoM (system-on-module) packs more features than any other Atom Z530 SoM - notably up to 2GB RAM, 4GB on-board Flash Disk, SATA interface, dual GbE, and up to 3 PCIe lanes.
Used as a building block for any embedded application, the CM-iAM packs up-to-date technologies into the most compact, lightweight PC-on-module available in the market.
With on-board resources that smoothly run Linux, Windows XP or Windows 7, the CM-iAM’s standard X86 architecture runs most software applications written for PCs. And, the CM-iAM’s credit-card sized board boasts a low power consumption of 4-6W, making it ideal for battery and “always on” applications.
The basic feature set of the CM-iAM includes a 32-bit X86-compatible Atom Z530 CPU (up to 1.6GHz), up to 2GB RAM, 4GB on-board Flash Disk, SATA interface, and a Micro-SD slot. The CM-iAM also offers multiple display interfaces (RGB, LVDS and SDVO), with resolutions of up to 1920x1200 – enabling it to run high the definition graphics needed for such applications as digital signage.
For embedded applications, the CM-iAM provides a variety of display interfaces, up to 3 PCIe lanes, dual GbE ports, an RS232 serial port, up to 15 GPIO lines and many other essential functions. The user interface is supported by an enhanced graphics controller, 6 USB 2.0 ports, and 5.1 channels audio codec (analog+S/PDIF or HDA output).
To enable quick evaluation of the CM-iAM in the full range of usage scenarios, the SoM ships with an advanced evaluation base-board, the SB-iAM. This evaluation kit features a wide range of extension and I/O interfaces – most notably PCI and PCIe slots, RGB, LVDS, DVI and CRT display, and digital, analog and optical audio.
“Anders has always attempted to keep our clients on the cutting edge of technology,” said Rob Anders, CEO of Anders Electronics. “The CM-iAM is a true first – the smallest and most cost-effective SoM on the market, that doesn’t compromise on processing power or featureset."

National Instruments expands HIL

National Instruments is extending its hardware-in-the-loop (HIL) simulation platform, which includes a range of products that optimise embedded system validation. During the past six months, NI has released nearly 40 new products targeted at delivering flexible HIL solutions to embedded control system developers within a variety of industries. The portfolio of NI HIL simulation tools helps engineers maintain reliability and time-to-market requirements while reducing costs, even as their products become more complex.
“We continually hear that engineers are struggling with traditional test systems to meet increasing product complexity and performance requirements within tight budgets and timelines,” said Mike Santori, Business and Technology Fellow at National Instruments. “These engineers need an HIL simulation platform that is highly productive out of the box but also open and flexible to adapt to fast-changing testing demands. The NI HIL simulation platform provides unprecedented openness and performance for HIL applications. The platform’s highly flexible architecture helps engineers address a wide range of applications, from those in automotive and aerospace to new fields such as alternative energy and medical device development.”

Recent product releases include NI VeriStand software for real-time testing and simulation; the NI TestStand 4.2 automated test management environment including support for Python scripts; a new family of fault insertion units; NI-XNET high-performance CAN and FlexRay bus interfaces optimised for HIL applications; ARINC 429, MIL-STD-1553 and AFDX (ARINC 663) military and aerospace avionics bus interfaces; low-cost and high-performance real-time processor cards; and several other I/O interfaces. To ensure that applications can easily scale and meet evolving requirements, the NI HIL simulation platform supports third-party hardware interfaces and integrates with C, C++, .NET and Python programming languages. In addition to integrating seamlessly with the NI LabVIEW graphical system design environment, the platform works with a variety of modeling environments such as Simulink software; ITI SimulationX; Maplesoft MapleSim; and Gamma Technologies GT-POWER.
Engineers can increase their system performance and flexibility while reducing overall costs by taking advantage of the open PXI hardware standard, advanced multicore technology and graphically programmed FPGA interfaces. Additionally, the platform’s software-defined instrumentation approach makes it possible for HIL applications created with NI products to scale from low-cost desktop validation systems to multiprocessor distributed simulators, a benefit that provides engineers a flexible and cost-effective toolset for all HIL testing applications.
The platform delivers commercial off-the-shelf (COTS) solutions that offer alternatives to complex proprietary configurations and bulky, inefficient traditional simulation systems. In today’s challenging economic climate, NI HIL simulation products are ideal for making projects more efficient and cost-effective for design engineers in multiple industries, from aerospace, alternative energy, automotive and consumer electronics to government, industrial transportation, mechatronics, medical technology and semiconductor manufacturing.

Wednesday, November 18, 2009

PowerVR moves to HD

Imagination Technologies has launched a video encoder family of IP that provides H.264 High Profile (HP) at HD resolutions.
The newest member of the POWERVR VXE encoder family delivers the highest quality encode ever seen in a synthesisable IP core, and offers the most complete implementation of the High and Main Profile toolsets of any IP core available in the market today, says the company. This translates to improved image quality at all bit rates, and enables ultra-low bitrates to be achieved for encoded streams with minimum degradation from the original source image data.
"With the relentless growth in demand for high quality user-generated content in all forms of mobile devices, we've seen tremendous interest in our VXE HD video encoder products. Our partners have evaluated our advanced encoding algorithms used in VXE380 against many other solutions, and the feedback we're getting indicates we have a winning solution that is the ideal complement to the growing numbers of HD resolution image sensors now being deployed in mobile devices," said Tony King-Smith, VP marketing at Imagination.

POWERVR VXE encodes video sequences and still images to a wide range of compression standards including H.264 High, Main and Baseline Profiles, H.263, MPEG-4, M-JPEG and JPEG. These cores offload the entire video/still encode processing to hardware, minimising host CPU processing load, enabling lowest power solutions without compromising performance. The core is fully supported by an accompanying software stack, which includes an OpenMAX IL compliant API under Linux.
The core is scalable, allowing Full HD video to be encoded, as well as offering the ability to record at extremely high frame rates (e.g. >1200 fps at QVGA) for detailed slow motion capture - one of the most sought after features for encoders today. Support for 4K resolutions is also possible, due to Imagination's unique scalable architecture.
The third generation, intelligent motion search algorithm dramatically reduces memory bandwidth requirements, whilst offering an even larger effective search range to give the best possible video quality and compression ratios. As memory access is one of the key sources of high power consumption for a SoC (System-on-Chip), this reduction in bandwidth also helps lower the overall SoC power requirements.
The IP core contains a small embedded 32-bit META processor, which ensures that the rate control algorithms are both flexible and highly capable, without burden or reliance on the host processor. This flexibility gives designers the ability to meet specific rate control requirements for the most demanding applications.
The VXE380 IP core, like all of Imagination's IP cores, is designed with ease of system integration in mind, requiring minimal external components, supporting memory interfaces optimised for all popular DRAMs including the latest DDR3 devices.
The POWERVR VXE family of encoders is ideal for a wide range of consumer and embedded applications including mobile phones, digital cameras, Blu-Ray disc authoring, portable media players, video conferencing, surveillance, personal navigation devices, personal video recorders and hardware encode accelerators.
The POWERVR VXE380 IP core is available for licensing now.

DDR3 module price increase catches PC and chip makers

Pricing continues to increase at a rapid pace for both DDR2 and DDR3 memory modules according to a market update from Converge. Earlier in the year market expectations were for another down year for the memory market, and yet at the close of business for October, the DDR2 and DDR3 markets had experienced eight straight months of rising prices. Clearly, computer builders did not see this coming or they would have forecasted for more supply in the second half of 2009. Most builders are getting caught short, and savvy buyers are finding that the spot market has been instrumental in assisting with supply shortages. The sweet spots are still 1GB and 2GB PC800 for desktops and notebooks, and 2GB DDR3 PC1066 and PC1333 for desktops.
While it appeared that DDR3 had stabilized for a few weeks in October, activity has once again picked up as we close out the first week of November. DRAM manufacturers have their hands full as they decide which technology to produce more of so they can take advantage of these high resale prices. Converge believes November will continue to bring shortages for modules as well as for chips. While all the talk is about module shortages, the chip market is also experiencing severe shortages. Micron parts in particular have been on a lot of contract manufacturers' radar screens for shortages, with lead times running 10+ weeks.

Tuesday, November 17, 2009

Supercomputing moves into harsh environments

GE Fanuc Intelligent Platforms, part of GE Enterprise Solutions, has developed a rugged processing board with the performance of a supercomputer.
The OpenVPX-compatible GRA111 high performance graphics board is one of the first commercially-available products to feature Nvidia’s newly-announced GT 240 GPU (graphics processing unit) and is also the first rugged implementation of a GPU using Nvidia's CUDA (Compute Unified, Device Architecture). The board is the first result of a relationship with Nvidia announced on October 8th.
The GRA111 is the first member of GE Fanuc’s planned family of CUDA-enabled products that are expected to revolutionize military and aerospace embedded computing by bringing a supercomputing capability to applications and environments that require leading edge processing such as radar, signals intelligence and video surveillance and interpretation. For example, radar systems can run up to 15 times faster.
“The enormous potential of the GPGPU – the general purpose graphics processing unit – with its massively parallel computing architecture is just beginning to become apparent to developers in the military and aerospace field,” said Rob McKeel, Vice President, Embedded Systems at GE Fanuc Intelligent Platforms. “What we are announcing today is an exciting first: a new compact, lightweight platform that can deliver exceptional levels of computing performance in demanding applications and challenging environments.”

NVIDIA’s CUDA is a general purpose parallel computing architecture that takes advantage of the parallel compute engine in NVIDIA GPUs to solve many computational problems in a fraction of the time required on a general purpose CPU. It includes the CUDA Instruction Set Architecture (ISA) and the parallel compute engine of the GPU.
CUDA has been shown to deliver a 100x increase in speed in a broad range of applications that lend themselves to parallel computing. Early tests at GE Fanuc have shown that military applications can benefit from similar increases in speed. A major prime contractor in the military/aerospace industry has evaluated the CUDA architecture in a radar system, and found that performance improvement of 15x is achievable with minimal reprogramming effort.
The 3U format GRA111 derives much of its performance from the 96 cores, 128-bit memory interface, 1 GByte of DDR3 video memory, 16-lane PCI Express Gen 2 interface, 540MHz graphics clock and 1.3GHz processor clock of the GPU.
Optionally available as an LRM (Line Replaceable Module) in accordance with the VPX-REDI (VITA 48) standard and with a rich set of I/O, the GRA111 can also be applied to many of the most common graphics and video applications. Dual, independent channels mean that it is capable of driving RGB analog component video, digital DVI 1.0, and HDMI standards. In addition, the GRA111’s video input capability allows integration of sensor data using RS170, NTSC or PAL video formats.

8bit microcontrollers fight back with dual core for payTV

While 32bit microcontrollers are moving down into the 8bit market at 65cents, 8bit devices are moving to dual core.
Atmel has developed the industry's first dual-core secure microcontrollers targeting demanding smart card applications such as the PayTV conditional access market.
Atmel's TwinAVR microcontrollers (MCUs) offer two independent cores, memory and peripherals on a single chip, unlike current secure MCUs on the market today that offer a single core.
The Master Core is built around Atmel's 8-/16-bit RISC secureAVR, with 128K ROM, 36K EEPROM and 6K RAM. The Secure Core is also built around Atmel's 8-/16-bit RISC secureAVR, with 64K ROM, 18K EEPROM, 6K RAM, but with AdvXTM Crypto Processor, TDES, AES & AIS31 TRNG. This dual core architecture simplifies the implementation of the RED/BLACK concept of segregation between highly sensitive plaintext data (RED) and encrypted data (BLACK), leading to an easier and faster certification.
"As the PayTV conditional access market continues to grow and become more complex, smart card developers demand higher performance and storage capacity while maintaining the highest level of security," said Eric Le Cocquen, Atmel's marketing manager for PayTV products. "Atmel's TwinAVR addresses all these issues. Segregation of the RED and BLACK domains traditionally relies on hardware firewalls and a lot of software. The TwinAVR goes one step beyond by offering, on the same chip, a master core to implement the BLACK domain and a secure core to implement the RED domain, thus providing an indisputable segregation between the two domains. The two cores also increase the overall system performance considerably, equaling a 32-bit MCU, while the data flash interface enables large storage capacities. This new family offers a complete solution for smart card developers in the PayTV market."

The TwinAVR also features an ISO 7816-3 smart card interface and a built-in DataFlash interface. Large amounts of data can be stored with flash die stacking to the microcontroller and by fitting both chips in a standard smart card module. This product meets Common Criteria EAL5+ security certification.
Pricing starts at $1.85 USD for quantities of 100Ku in bare die form, with samples of the AT90SDC104 4Mbit TwinAVR samples will be available Q1 2010 at $4.00 USD for quantities of 100Ku in smart card modules.

Atmel launches low cost 34cent RFID chip for next generation access control

Access Control System Diagram

RFID block diagram

Atmel has launched a low cost, One Time Programmable (OTP) transponder chip for next-generation access control systems. Application areas include, for example, hotel rooms, engineering departments, offices, time recording systems and parking lots, as well as customer loyalty and membership cards. The IDIC ATA5575M1 provides OTP functionality, which simplifies the production process and provides increased flexibility compared to read-only devices. Customers can program the required code before shipping, reducing the lead time and time-to-market by many weeks.
A new architecture enables longer read distances with different types of coils and readers. The write distance has been reduced since a short write distance is sufficient and helps to minimize both cost and chip size. Large write distances are also in many cases unfavorable due to the risk of multiple tag programming.
Further system cost reduction results from the integration of the optional trimmed 250- or 330-pF on-chip capacitors. These capacitors eliminate the need for external components, since a coil is all that is needed for a complete system, enabling extra-small access control systems. Since the ATA5575M1 measures only about 0.9 square millimeters, the chip can be used in almost any transponder package, including glass transponders for very small tags such as plastic key housings.
A unique ID is vital for guaranteeing access to secured areas and enabling traceability of the end product. The device's user memory contains a unique, manufacturer-programmed ID. In standard applications, the system designer can simply lock the tag and use this pre-programmed unique ID as is. In applications requiring an individual, unique ID, the designer can overwrite the existing ID by using the desired customer-specific code.
As the ATA5575M1 is an LF device, it can be used worldwide. It is insensitive to rugged environments and can also be used under conditions that normally complicate the application of RFID devices, e.g. under water, on metal, in dirt, out-of-sight, or worn on the body.
The ATA5575M1 transponder IC supports ASK modulation and Manchester coding with a fixed bit rate of RF/64 and is designed for passive identification systems with a 100 to 150 kHz magnetic field. It can replace almost all available LF RFID read-only devices with the 'unique format'.
Samples are available now as wafer or die-in-waffle pack. Pricing for the ATA5575M1 as wafer shipment start at US$0.34 in quantities of 15k pieces and is available as 64- or 128-bit memory version. To support the engineer and to simplify the design of complete RFID systems, Atmel provides the evaluation kit ATA2270-EK1 including ATA5575M1 sample tags. The kit is based on Atmel's well-established AVR microcontrollers and is accompanied with a Windows PC software, C-source code for the AVR and PCB Gerber data for the reader board.

Monday, November 16, 2009

32bit microcontrollers fall to 65cents

The latest family of 32bit ARM Cortex-M0 microcontrollers from NXP is driving real time oerating systems straight into the 8bit and 16bit space. The LPC1100 microcontroller family will be available from distributors in December from 65cents in 10K volumes. With 15 members initially, the family offers a seamless entry point for any 8/16-bit customers looking to start with the scalable ARM architecture throughout their entire range of product development.
This provides an excellent opportunity for RTOS vendors such as
with the FreeRTOS and SafeRTOS. This is also a direct challeng to Texas Instruments with the new Coretex M0 family from recently acquired Luminary Micro.
“Existing 8-bit architectures have their origins in the early era of the semiconductor industry, resulting in limitations of address range, register restrictions, limited functionality, unsuitability for high-level languages, and little attention to power and scaling issues,” said Geoff Lees, vice president and general manager, microcontroller product line, NXP Semiconductors. “The Cortex-M0 processor core and system architecture take full advantage of today’s optimized low-power design tools, techniques, and the latest low-power, high-density silicon Flash process.”

With over 45 DMIPS of performance compared to the sub-DMIP performance typical of 8bit MCUs and 3 - 5 DMIPS for 16-bit MCUs, the LPC1100 can not only execute basic control tasks but also sophisticated algorithms, making even the most complex tasks within reach. Shorter time to do more tasks translates directly into lower energy consumption. This level of performance is delivered at 50 MHz, with extensive power optimization, at less than 10 mA. The industry standard Coremark benchmarks also show that the LPC1100 requires 40-50 percent smaller code for most common microcontroller tasks.
“It may be a big surprise to embedded users how much the LPC1100, a 32-bit microcontroller, outperforms in efficiency compared to 8- and 16-bit microcontrollers. If performance and energy consumption are important criteria for selecting a microcontroller, based on the results generated from EEMBC's CoreMark benchmark, embedded designers should check out the LPC1100 (especially with its $0.65 price tag) before committing to any 8- or 16-bit options with comparable features and pricing. The scores for the LPC1100 are already posted on,” said Markus Levy, EEMBC President.

Features of the NXP LPC1100 family of microcontrollers include:
· 50 MHz Cortex-M0 processor with SWD/debug (4 break-points)
· 32 Vectored Interrupts; 4 priority levels; Dedicated Interrupts on up to 13 GPIOs
· UART, 1 or 2 SPI, I2C (FM+); 2 16-bit and 2 32-bit timers with PWM/Match/Capture
· 12MHz Internal RC Oscillator with 1% accuracy over temperature and voltage
· Power-On-Reset (POR); Multi-level Brown-Out-Detect (BOD); 10-50 MHz Phase-Locked Loop (PLL)
· 8-channel high precision 10-bit ADC with ±1LSB DNL
· Up to 28 or 42 fast 5V tolerant GPIO pins for HVQFN33 and LQFP48 respectively, high drive (20 mA) on select pins
· Single 1.8 – 3.6V power supply; over 5kV ESD for rugged applications

Future product features will include: Ultra Low Power options, CAN, 12-bit ADC and DAC, temperature sensor, high resolution timer features, and advanced sensor interface.

The LPC1100 family is supported by development tools from IAR, Keil, Hitex, Code Red, and many others. NXP will also offer an easy to use, comprehensive development tool platform for under US$30.
Recommended distribution unit pricing in 10,000 piece quantities for the 33-pin package is: LPC1111FHN33/101 at US$0.65, LPC1112FHN33/101 at $0.75, LPC1113FHN33/201 at $0.85, and LPC1114FHN33/201 at $0.95, with Flash sizes of 8K, 16K, 24K and 32K respectively. In addition, 48-pin LPQFP and PLCC44 packages will be available for socketed applications.

Bada emerges as OS for handhelds

Samsung is set to launch its Bada operatings system as a challenge to Android. Details and prospects at the Portable Multimedia blog.

Wednesday, November 11, 2009

European car telematics market to boom

Despite the collapse of the new car market, iSuppli is predicting a boom in telematics equipment for vehicles in Western Europe over the next ten years, and the details of the usage of telematics for each brand make very interesting reading (below), particularly the relatively low levels of telematics used today.
Sales of automotive telematics systems are set to rise by a factor of five during the period from 2008 to 2016 as carmakers offer more telematics-equipped models in the region, reaching 24.8 million systems by 2016, expanding at a Compound Annual Growth Rate (CAGR) of 22.5 percent from 4.9 million systems last year.
“Automotive telematics is defined as the integrated use of telecommunications and informatics, allowing the sending and receiving of information,” said Anna Buettner, analyst for automotive electronics at iSuppli. “Telematics can provide a range of benefits to motorists, from notifying an emergency operator when a car’s airbags have been deployed, to reporting vehicle conditions to a remote monitoring center.”
Two major types of telematics solutions are being offered on the market: embedded systems, whose functionality is integrated into the headunit of cars; and mobile-device oriented systems, which use a wireless product like a cell phone to communicate information.
“iSuppli believes that embedded and mobile-device-based telematics systems will continue to coexist in the car market in Western Europe and elsewhere,” Buettner said. “Embedded solutions are suitable for high-end luxury models, while mobile-device-oriented systems can bring telematics services to high-volume, entry-level cars. Regardless of which approach is used, pipes used for communications and delivery of content will become increasingly transparent and will be modified to cater to a vast array of data types.”
GM’s telematics sales in Western Europe are projected to grow to more than 1.4 million units in 2016, rising at a CAGR of more than 41 percent from 210,000 million in 2008. GM currently offers only mobile-device-based telematics systems.
Ford is expected to introduce the Sync system in Europe in 2011. This should increase its mobile-device telematics system shipments in the region to more than 1.7 million systems in 2016, rising at a Compound Annual Growth Rate (CAGR) of 31 percent from 190,000 in 2008.
The Japanese brands are not as strong in Europe as they are in North America, so their telematics sales will be much lower in the region. Toyota, which maintains the largest auto sales among the Japanese brands in Western Europe, will have the highest telematics sales there—growing to 990,000 units in 2016, up at a CAGR of 31 percent from 110,000 in 2008. The Japanese leader most likely will bring an embedded solution to the market in the near future.

Domestic telematics

The Top-6 European auto manufacturers collectively control about 64 percent of the Western European auto market.
PSA now has the highest telematics sales in Europe due to its use of both embedded and mobile-device systems. Company sales are forecasted to grow to 3.1 million systems in 2016, up from nearly 620,000 in 2008. Embedded systems accounted for 42 percent of PSA’s 2008 telematics sales.
VW’s telematics sales are projected to grow to nearly 3.2 million units in 2016, increasing at a CAGR of more than 26 percent from 490,000 systems in 2008. While the company’s four brands currently use only mobile device telematics systems, the German OEM is expected to introduce an embedded telematics system in the future.
Fiat’s three brands will sell 2.3 million units in 2016, up at a CAGR of 27 percent from 340,000 in 2008. Embedded systems accounted for about 20 percent of Fiat’s telematics sales in 2008.
Renault has been slow to adopt telematics and has just recently added mobile device systems. Renault’s telematics sales are projected to grow to 1.6 million units in 2016, rising at a CAGR of 30 percent from 190,000 in 2008.
The two main luxury brands in Europe, BMW and Mercedes-Benz, already sell a high percentage of their autos with telematics systems.
BMW uses both embedded and mobile device telematics systems. Company telematics sales will rise to more than 1.3 million units in 2016, up at a CAGR of 16 percent from 390,000 in 2008.
Mercedes-Benz currently sells only mobile-device telematics systems. Sales of Mercedes-Benz telematics systems are forecasted to grow to 1.4 million systems in 2016, expanding at a CAGR of almost 20 percent from 320,000 in 2008. Similar to Toyota/Lexus and Volkswagen, Mercedes-Benz also is expected to introduce an embedded telematics solution for its future offerings.

Cognitive radio for old TV spectrum

Cambridge Consultants has developed a novel, low-cost ‘spectral sensing’ cognitive radio technology platform that will allow any radio product to transmit without interference over so the called ‘whitespace’ frequencies at 700MHz recently vacated by the US digital TV switchover. With InCognito, it is aiming to do for whitespace what it did for Bluetooth and Zigbee, spinning off companies such as Cambridge Silicon Radio.
Whitespace is highly desirable TV band frequencies that easily penetrate walls, potentially extending the range of Home Area Networks (HANs) and enabling a range of novel new applications, such as reliable high-definition video streaming from a single access point to every room in a house. Innovative use of whitespace radio will also make it possible to increase the accessibility of low-cost high speed wireless internet services, including those rural communities which are currently poorly served.
The new frequencies, recently authorised in the USA by the Federal Communications Commission (FCC), greatly increase the overall wireless bandwidth available to computers, set top boxes, laptops, WiFi hot spots and other radio devices that currently use the unlicensed band around 2.4GHz. They also have the potential to greatly extend the range of these devices.
“We’ve seen so much game-changing innovation in the unlicensed 2.4GHz band, but I believe the FCC’s decision to open up the ‘whitespace’ radio frequencies for innovation promises even more. We will quickly see a wave of innovation in wireless products and services around 700MHz, bringing benefits both to consumers and to the innovative businesses that move quickly into the whitespace market,” said Luke D’Arcy of Cambridge Consultants.
But the decision is not without controversy. The primary incumbent users of these TV band frequencies are concerned about the potential for interference caused by the influx of new users. “Based on highly complex cognitive ‘spectral sensing’ radio technology which, until now, has only been used in defence and security applications, the InCognito platform enables ‘whitespace’ radios to quickly and accurately detect and avoid other broadcasts," saidD'Arcy. "We also understand that the most successful technologies in use in these markets today are based on open industry standards, which is why we have taken our platform to the CogNeA Alliance. This, together with our track record of fast product development, means that any client can take advantage of the new spectrum quickly and easily - adding this potentially low cost technology to a client’s device could take as little as a few months. I believe we will start to see the first cognitive radio enabled products in mid 2010 and that the market will develop quickly.”
The CogNeA Alliance aims to drive the definition and adoption of industry-wide standards for wireless devices to operate over the new TV white spaces. Kiran Challapali, CogNeA’s Chairman, said “Cambridge Consultants has helped get many successful new standards such as Bluetooth, DECT and Zigbee off the ground. We welcome them to our alliance, which includes companies like Philips, Samsung Electro-Mechanics, ETRI and Georgia Tech, and we look forward to working together to make CogNeA a successful standard."

Intel moves into equipment with the Reader

Intel's Digital Health group has launched a handheld device in the US that reads out text. This is one of the few times that Intel has sold consumer equipment under its own brand and is a key step towards the Intel phone. Intel already makes notebooks and netbooks, but these are all labelled by, and sold by, other companies.
The Intel Reader is designed to increase independence for people who have trouble reading standard print. About the size of a paperback book, it converts printed text to digital text, and then reads it aloud to the user. It combines a high-resolution camera with an Intel Atom processor, allowing users to point, shoot and listen to printed text.
The original concept for the Reader came from Ben Foss, a researcher at Intel who was identified in elementary school as one of the estimated 20 percent of people nationwide who have symptoms of dyslexia. Throughout high school, college and graduate school, he had to depend on others to read to him or work through the slow process of getting words off of a page himself. As an adult, much of the content he wanted, from professional journals to pleasure reading, just wasn't available in audio form.

"As someone who is part of this dyslexic community, I am thrilled to be able to help level the playing field for people who, like me, do not have easy access to the printed word," said Foss. "Feelings of loneliness are often the experience of not being able to read easily. We hope to open the doors for people in these communities. The Intel Reader is a tool that can help give people with dyslexia, low-vision, blindness or other reading-based disabilities access to the resources they need to participate and be successful in school, work and life."

Intel says it can assist the estimated 55 million people in the U.S. who have dyslexia or other specific learning disabilities, or have vision problems such as low-vision or blindness, which makes reading printed words difficult or impossible.
The Intel Reader will be available in the United States through select resellers, including CTL, Don Johnston Incorporated, GTSI, Howard Technology Solutions and HumanWare.
When the Intel Reader is used together with the Intel Portable Capture Station, large amounts of text, such as a chapter or an entire book, can be easily captured for reading later. It is available through selected outlets in the US.
"The Intel Digital Health Group's expertise is in finding innovative technology solutions to improve quality of life," said Louis Burns, vice president and general manager of Intel's Digital Health Group. "We are proud to offer the Intel Reader as a tool for people who have trouble reading standard print so they can more easily access the information many of us take for granted every day, such as reading a job offer letter or even the menu at a restaurant."

Monday, November 02, 2009

DDR3 memory - what happend?

I've just had to buy a new laptop, and wanted something that would last for a good few years, so naturally wanted high speed, low power DDR3 memory. Could I find much? No, really just Dell and Acer.
So this analysis of what happened from Malcolm Penn at Future Horizons is well timed, and an excellent description of how technology does, or doesn't get adopted, and the implications for the embedded market! It looks like my timing is good though as now I have a lovely DDR3 system at a reasonable, not bleeding edge, cost.

Semiconductor Spotlight – DDR3

The DRAM manufacturers as a whole made a combined loss during 2008 of US$20 billion. At the moment memory pricing is firming but semiconductor memories are still plagued by a regular cyclicality of over-capacity and declining ASPs countered by tightening capacity and rising ASPs. However, although improvements are evident in the short term, the longer-term trends are still up for debate. One of the major trends currently under way is the trend to DDR3 memory from the current mainstream DDR2.
DDR, DDR2 and DDR3 are types of SDRAM (Synchronous Dynamic Random Access Memory) and use a clock signal for synchronization. DDR stands for Double Data Rate, meaning that memories from this category transfer two data chunks per clock cycle.
One of the main differences between DDR, DDR2 and DDR3 is the highest transfer rate each generation can achieve. For example, higher speed DDR2-1066 memory operates at a clock rate of 533 MHz. Whilst there is an equivalent DDR3 memory that also operates at this speed, the DDR3-1600 operates with a clock rate of 800MHz. In addition, the internal data path of DDR2 memory is 4 bits whilst DDR3 is 8 bits. DDR3 memories also operate at lower voltages compared to DDR2 memories.
This lower voltage means that DDR3 consumes less power than DDR2 for a given clock rate and lower power is becoming increasingly important for laptops and the general trend of electronic equipment to become ‘greener’.
On the minus side, DDR3 memories have higher latencies than DDR2 memories. This means that DDR3 memories take more clock cycles initially before delivering data compared to DDR2 memories.
The primary benefit of DDR3 is the ability to transfer at twice the data rate of DDR2. DDR3 can operate at higher clock rates than DDR2 and DDR3 memory uses less power for the same clock rate. An emphasis on low power and higher data rates make DDR3 memory a more attractive option provided there is a low premium on price and the motherboards are ‘DDR3 ready’.
The first DDR2's replaced DDR in PCs by Dell during the summer of 2004 and were DDR2-533 specification. Dell has been one of the leaders in new system development and it was expected that that the transition to DDR2 would start getting underway very soon after. However, history showed that the uptake of DDR2 was not a smooth transition with memory manufacturers reverting to DDR and then going back to DDR2. The changeover eventually occurred during early 2006 when DDR2-667 had both a performance and price advantage.
It looks as though history will repeat itself with the transition to DDR3 from DDR2 but perhaps over a shorter time-span. DDR3 has been in gestation since 2005 and has appeared on some motherboards during 2007. Optimistic predictions were made for adoption during early 2008 with predictions of 30 percent or more penetration of DDR3 memory with less than ten percent price difference in price over DDR2 by the end of 2008.
By mid-year 2008 this prediction was looking increasingly unlikely and eventually did not happen. However, the pricing has been altering in favour of DDR3 in the last four months and the percentage of DDR3 has been creeping up to over 20 percent of total PC memory in the first half of this year.
PC OEMs are inexorably price driven and will tend to use the cheapest available memory given comparable performance in mainstream machines. Another factor is, of course, the type of memory the motherboard can take. DDR3 has become more popular since July and, as a consequence, the price for this memory rose as demand increased.
As the DDR3 prices increased, PC manufacturers started to go back to using DDR2 as these were lower in price. However, increased DDR2 demand, in turn, drove up prices. With prices now higher for DDR2, PC OEMs will switch back to DDR3 parts. This cycle will likely continue for the next six months until DDR3 memory becomes the mainstream memory for PCs probably during mid-2010.
The adoption of any new technology is always beset by problems in its introduction and the transition from DDR2 to DDR3 is promising to be no exception. It has been clear for some time that DDR3 memory has had the potential to become the mainstream PC memory but market adoption depends on a careful mix of price and performance with a lot of emphasis on price.
As such we believe DDR3 will really come into its own when the new generation 50nm designs hit the market. This advance will also allow a large increase in die-per-wafer and either an improvement in profitability or a tool to increase market share depending on the individual memory manufacture’s strategy.

China supercomputer breaks performance barrier

China's National University of Defense Technology (NUDT) has developed a supercomputer that breaks the petaflop barrier, putting China in the same league as the US and Europe for high performance computing.
The supercomputer, called "Tianhe", meaning Milky Way, is based in Changsha, capital of central China's Hunan Province.
The 155-ton system, with 103 refrigerator-like cabinets covering an area of about 1,000 square meters cost $88m and is expected to process seismic data for oil exploration, conduct bio-medical computing and help design aerospace vehicles. It uses 6,144 Intel CPUs and 5,120 AMD GPUs, and reaches a peak performance of 1.206 petaflops and 563.1 teraflops on the Linpack benchmark, making it currently the fourth most powerful machine on the planet.
China's Dawning Information Industry Company is attempting to build its own supercomputer that overcomes the petaflop barrier by 2010, and the NUDT is planning to add "hundreds or thousands of China-made CPUs" to the machine to improve its Linpack performance to over 800 teraflops.