Thursday, May 3, 2012

2HApr. mainstream NAND flash average contract price falls 2-9 percent due to slow season effect and seesaw-battle price negotiations

USA: According to DRAMeXchange, a research division of TrendForce, 2HApr. inventory replenishment for China’s Labor Day holiday was not as expected. Furthermore, aside from restocking for certain new tablet PC and smartphone models hitting the market in 2Q12, most system product and memory card application markets remain sluggish, affected by the 1H slow season.

Moreover, NAND Flash price has fallen dramatically since the beginning of the year, which has already impacted NAND Flash suppliers’ profit margins. Thus, since the beginning of April buyers and vendors have taken opposing stances on price negotiations - some firms have still not reached a consensus on contract price for some chips by the end of April. Therefore, the average mainstream NAND Flash contract price decreased about 2-9 percent in 2HApr roughly.Source: DRAMeXchange, Taiwan.

TrendForce provides the following analysis of supply and demand factors affecting the NAND Flash market:

1. Looking at demand, based on the release schedule of new processors, operating systems, and low-cost 20nm-class-process SSD products, many manufacturers are not likely to significantly increase shipments of new ultrabook, smartphone, and tablet PC models until 4Q12. Therefore, restocking of related NAND Flash application products will not increase markedly until mid-3Q12.

2. As for supply, due to market demand that has been weaker than expected in 1H12 and the potential effect of delayed release of some system products in 2H12, many NAND Flash suppliers have indicated they plan to carefully manage 2012 capital expenditures and bit growth, as well as temporarily suspend or slow wafer expansion plans or reallocate a portion of product lines to non-memory IC products.

Additionally, some NAND Flash suppliers will gradually decrease the sales portion of memory cards and UFDs in channel market, which have relatively lower demand, favoring instead embedded applications such as eMMC, mSATA and SSD – this will increase the sales portion of system product application revenue, reducing the impact of price decline on profitability.

Hopefully, excess inventory from 1H12 will be digested before the peak sales season in 4Q12 and the market oversupply situation will improve.

In summary, although the 2Q12 NAND Flash market remains in a state of oversupply, as NAND Flash suppliers’ profits were impacted by 1Q12 price decline, and new 20nm-class-process products’ yield rate, performance, and reliability all require some time to improve, makers will be less willing to adopt aggressive price-cut policy in 2Q12.

Therefore, NAND Flash contract price will continue to show a slight downtrend in the 2Q12 slow season, but price is expected to stabilize in early 3Q12 due to recovery of inventory restocking demand caused by some system product clients’ new model releases.Source: DRAMeXchange, Taiwan.

Dongbu HiTek to showcase specialized BCDMOS and analog CMOS chip technologies at PCIM Europe

SEOUL, KOREA: Dongbu HiTek announced that it will present new insights into its best-in-class Analog CMOS and BCDMOS processing/design technologies at the PCIM (Power Conversion Intelligent Motion) exhibit and conference, which commences its three-day run on May 8th in Nuremberg, Germany.

In addition to showcasing its best-in-class analog and mixed-signal semiconductor process technologies in the main exhibit hall (Booth 764), the Korean company will be participating in a “DC/DC Conversion” panel discussion on May 9th (Booth 670, 12:20-13:20) and making UHV 700V BCDMOS presentations two other days as follows:

May 8th (Foyer Ground Floor, 15:30-17:00) Poster Dialogue Session
“Design & Optimization of Robust 700V DR-LDMOS Using Thin Epitaxial Technology”
Presenter: Bon-Keun Jun, Dongbu HiTek Senior Manager, Process Development

May 10th (Booth 670, 14:40-1500) Forum Program
“UHV 700V Technology for High Efficiency Power IC”
Presenter: Lyes Djama, Dongbu HiTek European Marketing Director)

During the May 9th panel discussion, Dongbu HiTek will introduce a specialized Analog CMOS process (AN180X) for achieving highly efficient DC/DC conversion. Panel participants from Texas Instruments, Intersil, International Rectifier and Infineon are also expected to discuss their respective DC/DC conversion technologies.

Dongbu HiTek’s specialized BCDMOS technologies at the 0.35-micron node continue to emerge best-in-class across diverse applications. During PCIM, the company will be exhibiting two options for its superb UHV 700V process: an N-Epi version for industrial, automotive, military and medical electronics; and a Non-Epi version for lighting and power supply applications. The N-Epi process is ideal to implement gate drivers, motor drivers, solar energy inverters and IGBT modules; the Non-Epi version is well suited to implement ADCs, DC/DC converters and PFC controllers.

The company will also showcase four superb single-chip BCDMOS solutions: a high-voltage motor driver for industrial applications; and three others that address advanced automotive applications including Adaptive Front Lighting System (AFLS) to control headlights, Automatic Power Folding System (APFS) to control side-view mirrors and a Heated Steering Wheel Controller (HSWC).

Dongbu HiTek technologists will be on hand to discuss each of these specialized process technologies in greater detail, and how they can be fully leveraged to speed time to market and reduce overall manufacturing costs.

Arasan Chip Systems announces MIPI compliant low latency interface (LLI) IP solution

SAN JOSE, USA: Arasan Chip Systems Inc., a leading provider of Total IP Solutions, announced the availability of their MIPI LLI controller IP along with a matching Type 1 M-PHY, the latest additions to its prominent MIPI portfolio.

Both the Applications Processor and the Baseband Processor for mobile platforms are complex SOC's. Although the two chips are often integrated into one SoC by a number of chip vendors, a number of high end mobile chipsets are still split into two separate processors.

They each have their own system level memory to allow efficient cache refills. LLI is a chip-to-chip link layer interconnect protocol that allows efficient , low-latency cache refills from the DRAM associated with a companion chip, thereby removing the need for two separate sets of DRAM's and substantially reducing the cost of mobile platforms. LLI requires M-PHY Type 1 as the physical layer.

Arasan has developed a combined LLI controller and M-PHY Type 1 solution, which can be configured for a variety of host buses (like AHB, AXI and OCP), and bandwidth/latency requirements across multiple traffic classes. Using up to six lanes of M-PHY's this solution offers up to 17 Gbps bandwidth in each direction, with only one clock domain crossing in the LLI controller. Customers are given a choice of either source synchronous or independent clocking in the M-PHY's for clock and data recovery mechanisms in the analog receivers.

Benchmark announces tablet reference design based on TI's OMAP 4 processors

ANGLETON, USA: Benchmark Electronics Inc., a leading integrated contract product design and manufacturing service provider, announced that the company's design engineering team has created a reference design for a tablet platform based on the OMAP 4 applications processors from Texas Instruments Inc. (TI).

This reference design, called the Grizzly Tablet, can now be leveraged by customers in the key end markets that Benchmark serves, such as medical and industrial control equipment, including aerospace and defense. The Grizzly Tablet will be showcased at the TI Technology Tech Day at the River Centre in St. Paul, Minnesota, on May 2, 2012.

By selecting Benchmark's reference design as a building block, customers have access to TI's OMAP 4 processors for differentiated solutions targeting applications such as telemedicine, rugged military and industrial applications, security applications and monitoring systems.

"Benchmark's status as a Platinum Member of the TI Design Network provides early access to this cutting-edge family of OMAP 4 processors for our customers. The Grizzly Tablet reference design offers the combination of design flexibility and state-of-the-art technology with great value and performance," said Mark Troutman, VP of US Design Engineering Services for Benchmark Electronics.

"As a Platinum member of TI's Design Network, Benchmark plays an important role in providing access to a broader base of customers interested in working with our leading OMAP processors. Benchmark's experience and track record of designing, integrating and manufacturing electronics products accelerates the design process, and helps customers bring proven, solid products to market," said Mike Schoonover, OMAP Industrial Business Development manager, TI.

Wednesday, May 2, 2012

Active-Semi announces three high performance quasi-resonant (QR) mode PWM ICs for “Green Power” AC/DC adapter apps

SAN JOSE, USA: Active-Semi announced the release of the new ActiveQR family of high performance QR mode secondary side regulator (SSR) ICs. These devices integrate benchmark protection features, offer fast load dynamic response, achieve accurate overload protection (OLP) and improve efficiency and EMI performance.

The ACT510, ACT511 and ACT512 belong to the patented ActiveQR Family of Universal-input AC/DC off-line controllers for general purpose adapter applications. The ACT510 and ACT511 are designed for Flyback topology working in discontinuous conduction mode (DCM) with full load switching frequency of 120kHz and 65kHz respectively. The ACT512 may be designed to work in either continuous conduction mode (CCM) or DCM at full load. This improves the efficiency in higher power applications, up to 60Watts, at a lower cost point.

"We have been very pleased with the ACT512's performance, especially in efficiency margin and protection coverage,” said A. Cao, manager of Technical Dept. of DareGlobal. “I was impressed that such extensive protection and QR mode control could be integrated into a tiny SOT23-6 package. The circuit design is easy and neat. Normally, such a performance controller requires SOP-8 package with more complex and expensive circuit design."

“At Active-Semi, we focus on developing innovative charger solutions as well as general purpose adaptor solutions,” said Peter Huang, director, AC/DC Marketing and Applications at Active-Semi. “With our unique architecture, the feedback system can monitor an extensive range of abnormal fault conditions and provide protection upon these conditions. We provide value to the end customer in both cost and performance. In 12V/2A applications, we can improve efficiency by 2 percent with standby power below 90mW and fast load dynamic response. This is one of the lowest standby power levels achieved in its class.”

All three products are sampling today.

Sample pricing for the ACT510-US-T is $0.18 each; for the ACT511US-T, it is $0.19 each and for the ACT512. $0.16 each.

INSIDE Secure named cool vendor by Gartner

AIX-EN-PROVENCE, FRANCE: INSIDE Secure, a leader in semiconductor solutions for secure transactions and digital identity, announced it has been named a "Cool Vendor" in the latest "Cool Vendors in Semiconductors, 2012" report published on April 25, 2012 by leading analyst firm, Gartner Inc.

In the new report, written by Gartner analysts Jim Tully, Ganesh Ramamoorthy and Mark Hung, Gartner recognizes INSIDE Secure as an innovative, impactful and intriguing vendor. INSIDE Secure is one of the pioneers in near field communication (NFC) technology, and has developed a unique ability to provide a complete end-to-end NFC solution with its lineup of secure element, NFC controller and software stack offerings.

"As a leader and innovator in NFC, we are pleased by Gartner's recognition. We consider it a testament of our long-standing commitment to the NFC market and the technological advances we have brought to it," said Remy de Tonnac, CEO for INSIDE Secure. "With our MicroRead and SecuRead NFC hardware, the industry's only commercial-grade, royalty-free, open-source NFC protocol stack, an extensive suite of OEM and carrier NFC apps and our ecosystem partnerships, INSIDE Secure is uniquely able to provide complete NFC solutions."

Dual-core processors power nearly 20 percent of smartphones in 2011

BOSTON, USA: Dual-core processors, which power ultra high-end smartphones, gained strong traction in 2011, accounting for nearly 20 percent of total smartphone applications processors shipped, according to, “Smartphone Multi-Core Apps Processor Market Share: Samsung Leads with 60 Percent Volume Share in 2011,” from the Strategy Analytics Handset Component Technologies (HCT) service.

This Strategy Analytics research shows that Samsung led the dual-core smartphone applications processor market in 2011, with 60 percent volume share, followed by Qualcomm, Texas Instruments and NVIDIA. Samsung’s success in dual-core processors can be attributed to its strong sales at Apple and in Samsung’s own Mobile Handset Division.

Stand-alone applications processors accounted for 90 percent of total dual-cores shipped in 2011. Single-core applications processors are much more likely to be integrated—72 percent of total single-core processor chips shipped were integrated with baseband applications processors in 2011.

Sravan Kundojjala, senior analyst, commented: “Strategy Analytics would advise NVIDIA to focus on high volume tier-one design-wins in 2012 in order to maintain its first-mover advantage with quad-core processors. NVIDIA, whose Tegra 2 smartphone shipments declined 8 percent at the end of 2011--compared to shipments in the first half of the year -- lost momentum, despite being early to market with dual-core ARM Cortex-A9 processors.”

Stuart Robinson, director of Strategy Analytics Handset Component Technologies service, said: “Qualcomm captured 16 percent volume share in the dual-core smartphone applications processor market in 2011 on the strength of its dual-core Snapdragon applications processor products MSM8x60 and APQ8060. Strategy Analytics believes that Qualcomm is well-positioned to make significant share gains in 2012 with the help of its LTE-integrated dual-core Snapdragon processor MSM8960.”

Kundojjala continued, “Strategy Analytics anticipates dual-core applications processor penetration into mainstream smartphones will accelerate through 2012 and into 2013, and new ARM architectures such as Cortex-A5 will help dual-core penetrate entry-level smartphones at the start of 2013.”

CEA-Leti promotes LETI-3S (Silicon Specialty Solutions) concept

GRENOBLE, FRANCE: CEA-Leti announced a focus on a new collaboration mode for a wide range of industry oriented customers from start-ups to manufacturers.

The research oriented unique CEA-Leti’s world-class process and metrology tool set should also offer:
* Advanced single step processes to high end process modules to realize some or the entire product or simply make a proof-of-concept demonstration;
* Flexible customer owned processes to meet small manufacturing requirements, through prototyping or small pre-series before transfer to commercial foundries;
* Equipment access through the expertise of CEA-Leti engineers.

LETI-3S for Silicon Specialty Solutions is targeting start-ups, component integrators, fabless or fablite companies, equipment or consumable suppliers. Other entities should be concerned such as foundries or research centers with a limited processes offer, micro and nanotechnologies companies who don’t want low-volume activities, and high-value silicon wafers dealers.

3S addresses following main domains:
* Deposition.
* Front Side/Back Side cleanings, wet etch and stripping.
* Lithography with dual side alignment capability. LETI-3S can provide generic masks but usually works on customer designs adapted to CEA-Leti lithography tools.
* Etching.
* Implantation, epitaxy, diffusion.
* CMP, bonding, grinding, dicing.
* Advanced in-line metrology.

LETI-3S is operated on the Minatec Campus in Grenoble on 24/7 200mm and 300mm technological platforms. Lower wafer sizes should be available thanks to an innovative holder technology. LETI-3S can provide substrates or work on customers wafers, pre-processed or not, through stringent contamination protocols.

The ISO9001 certification reinforces CEA-Leti’s leadership position as an industry oriented research center. LETI-3S ensures a full traceability from process flow conception to products delivery. SPC, FMEA, audits or conformance certificates to specifications are available according to the type of inquiry.

“LETI-3S is totally in accordance with the CEA-Leti missions to create and to transfer innovations to the industry“, said Laurent Malier, CEO of CEA-Leti. “This is a great opportunity for us to give a simple access to our resources and thus enlarge significantly the panel of our industrial partners on different markets in our application domains.”

Cavium announces shipment of OCTEON Fusion - industry's most powerful “base station-on-a-chip” family

SAN JOSE, USA: Cavium Inc. announced initial shipments of OCTEON Fusion small cell base station-on-a-chip family of SoC’s to Telecom Equipment Manufacturers (TEMs). The OCTEON Fusion family delivers breakthrough capacity and throughput for LTE small cell base stations, including micro, pico and enterprise femto base stations, through a high performance multi-core architecture.

OCTEON Fusion CNF71XX integrates Cavium’s award-winning OCTEON multi-core architecture along with purpose-built Baseband DSP cores, extensive LTE hardware accelerators and digital front end (DFE) features into a single chip.

The first release of this innovative new family of silicon includes four high performance MIPS64 cores and six DSPs, along with baseband hardware acceleration units interconnected with a low-latency crossbar and shared memory subsystem, allowing service of more than 128 connected users while delivering line-rate throughput for Uplink and Downlink. Compared to alternate solutions available today, OCTEON Fusion provides up to 4x greater performance within the same power and cost envelope.

Augmenting the macro cell network with a very large number of small cell base stations is essential to solving the mobile broadband capacity crunch and delivering 4G data throughput in a cost effective manner. For carriers to effectively deploy such a large and distributed radio access infrastructure, base stations need to be compact, inexpensive, easily manageable and energy efficient. Most of the TEMs and carriers are working to implement small cell based network topology and according to industry analysts, the small cell base station semiconductor market is expected to reach over $1 billion by 2016.

“With the shipment of OCTEON Fusion CNF71XX, service providers now have available unmatched carrier grade feature set and performance on all layers of LTE protocol in their small cell base stations,” said YJ Kim, GM of Cavium’s Infrastructure Products Group. “Increasing 4G momentum and consumers’ insatiable demand for mobile data are causing service providers to accelerate small cell deployments. OCTEON Fusion along with Cavium’s fully integrated and proven FusionStack™ LTE protocol software deliver standout performance while accelerating time to market for our TEM customers and their service provider customers.”

OCTEON Fusion’s innovative power efficient architecture allows the same scalable architecture to be deployed on both indoor and outdoor radio access solutions. Additionally, OCTEON Fusion maintains the same OCTEON architecture used in OCTEON II SoC’s that are widely deployed by several Tier 1 wireless OEM’s in their macro base stations and radio network equipment, enabling seamless reuse of existing software modules.

Cavium is currently sampling OCTEON Fusion CNF71XX to select customers. The OCTEON Fusion reference platform ships with comprehensive RF solutions that work in multiple bands and a complete set of fully optimized deployment ready software.

Fairchild's smart high-side switches provide designers reliable alternatives to discrete solutions for driving automotive apps

SAN JOSE, USA: In today’s automotive applications, designers need to reliably and safely switch high currents into grounded resistive or inductive loads. This includes applications such as incandescent lamps, motors, and heaters. To do this currently, they either have to rely on discrete or electro-mechanical solutions, or are hampered by limited solutions in the marketplace.

Fairchild Semiconductor, a leading global supplier of high performance power and mobile products, has developed a smart high-side switch family that specifically addresses advanced electrical load control in automotive body electronics. The family provides an alternative to discrete MOSFET solutions, integrating protection and diagnostic features in order to reduce component count and PCB complexity, while offering better system reliability.

The FDDS100H06_F085 smart high-side switch is an N-channel power MOSFET with charge pump, ground-referenced CMOS-compatible input and diagnostic output with integrated protection functions. The diagnostic feedback capability of the FDDS100H06_F085 provides system control options to minimize impact under various fault conditions.

The FDBS09H04A_F085A and FDDS10H04A_F085A are smart high-side switches incorporating an N-channel power MOSFET device featuring a charge pump, current controlled input and diagnostic feedback with load current sense and integrated Smart Trench chip-on-chip technology.

eASIC enables 3X increase in energy efficiency for astrophysical simulation supercomputer

SANTA CLARA, USA: eASIC Corp., a provider of NEW ASIC devices, announced that Tokyo Institute of Technology, in collaboration with Hitotsubashi University, has successfully demonstrated a 3X increase in energy efficiency for a green supercomputer used for performing astronomical simulations. By leveraging eASIC’s low power Nextreme-2 NEW ASIC devices, Tokyo Institute of Technology was able to achieve an energy efficiency ratio of 6.5 GFLOPS/Watt for its GRAPE-8 supercomputer.

The key component in achieving the highest energy efficiency ratio for the GRAPE-8 system is the use of eASIC’s Nextreme-2 NEW ASIC devices, which are commonly used for replacing FPGAs for high volume and power sensitive applications. Manufactured using a low-power 45nm LP process, and employing single via programming which eliminates the need for power hungry SRAMs that FPGAs require for programming look-up tables and routing, eASIC devices typically enable FPGA users to achieve up to 80 percent lower power consumption. eASIC’s unique GreenPowerVia technology also enables users to completely turn off any logic and memories that are unused in a design.

“eASIC’s devices have enabled us to take a giant step forward in resolving the power issue with next generation supercomputers,” commented Junichiro Makino, Professor at Tokyo Institute of Technology Graduate School of Science and Engineering. “Standard cell ASIC was prohibitively expensive and FPGA power consumption simply could not allow us to beat the previous record of 2.1GFLOPS/Watt which was held by IBM’s BG/Q green supercomputer. eASIC’s NEW ASIC allowed us to achieve a new record of 6.5 GFLOPS/Watt,” added Makino.

“This great accomplishment by Junichiro Makino and his team is testament to the innovative power saving technology eASIC has developed,” said Jasbinder Bhoot, VP of Marketing, eASIC. “The energy efficiency numbers they are demonstrating are very impressive. In many industries ranging from wireless infrastructure to enterprise storage, we are helping customers overcome the power problems caused by FPGAs. When low-power consumption is paramount, like Tokyo Institute of Technology’s GRAPE-8 processor, eASIC’s Nextreme-2 devices are the ideal choice.”

Broadcom announces industry's lowest power, highest performance 100GbE multi-rate Gearbox PHYs

IRVINE, USA: Broadcom Corp. has announced the industry's lowest power and highest performance multi-rate 100 Gigabit Ethernet (GbE) physical layer transceivers (PHYs) for next-generation data center, enterprise and core networks.

The new Broadcom BCM84790 and BCM84793 gearbox PHYs extend Broadcom's industry-leading portfolio of 10GbE, 40GbE and now 100GbE physical layer devices while offering the first multi-rate interface for 4-lane (4x25 Gbps) and 10-lane (10x10 Gbps) bi-directional transmissions.

The low power 40 nm CMOS architecture dissipates only 2.5 watts per port, delivering power savings of more than 35 percent as compared to previously available solutions. The BCM84790 features a space-saving 17x17mm BGA package specifically engineered to support CFP network applications. The BCM84793 features a 19x19mm BGA package with the ability to support 10-lane bi-directional transmissions at 10 Gbps each. This added flexibility ensures support for future-generation networking formats such as CFP2 and CXP.

Optimized for IT professionals who design and manage high-bandwidth data center and enterprise networks, the BCM84790 and BCM84793 gearbox PHYs have the ability to multiplex and demultiplex data across four 25 Gbps channels to (or from) ten 10Gbps channels. Both devices can also be configured to support four bi-directional lanes at 10Gbps for 40GbE repeater applications, support Ethernet and optical transport networking, and are compliant with the IEEE 802.3ba standard for 100GbE and ITU OTL 4.4 signaling.