Tag: chip

Intel to spend $550 million in Israel

Intel-logoIntel has pledged to write a cheque for at least $550 million in the Promised Land over the next five years.

This is part of Intel’s promise to spend a total of $6 billion to upgrade its Kiryat Gat plant for the manufacture of new advanced chips for its next generation devices.

The $550 is part of Intel’s offset purchase arrangement with the state, which is providing the company with grants of up to $600 million over the next five years as well as a major tax break through 2023. Intel will get two $300 million grants, distribution of which will be spread over five budget years.

Although these figures look great for Chipzilla, executives will be happier with the news that it will only have to pay a corporate tax of only five percent until 2023.  Others in Israel have to give the tax man 26.5 percent. In return, Intel committed to hiring at least 1,000 new employees, at least half of whom will be residents of communities in southern Israel. In addition, the company promised to spend at least $550 million.

Intel is committing to spend what it is getting from the government in direct grants, but the Economy Ministry claims the arrangement was great for the Israeli economy.

“This arrangement will have a very positive effect on hundreds of small businesses and suppliers,” said Ziva Eiger, director of investments at the Industrial Cooperation Authority.

“Offset agreements such as this are platforms for leveraging public expenditures for the benefit of the Israeli economy, both for training and encouraging further expansion of small suppliers for the local and world market, and to enhance Israel’s brand as an attractive place for foreign investment,” Eiger added.

“As a result of this agreement, Israelis can look forward to thousands of more jobs being available. It is a model for offset agreements that can provide benefits to all sides.”

 

Calxeda tech gets second life

calxeda_energy_core_handTechnology from Calxeda, which shut down last year, is being  repackaged and built by Silver Lining Systems.

Last year founder Barry Evans shuttered Calxeda which pioneered low-power ARM-based chips for use in servers for scale-out data centre environments. At the time, the industry was not interested. In December 2013, Calxeda ran out of money when financing fell through.

A year later Calxeda’s technology is re-emerging with a company called Silver Lining Systems (SLS), which is developing compute and storage servers that will use the IP Calxeda developed. It will introduce those systems early next year.

Evans told EWeek that it was exciting to see the technology he helped create, which many in the industry thought was gone, re-enter the highly competitive market. He is working with Silver Lining as adviser.

SLS is the cloud subsidiary of AtGames Cloud Holdings, a 13-year-old cloud gaming vendor that last year was developing systems based on Calxeda technology. When the company shut down, AtGames wanted to continue to develop its systems, so it bought Calxeda’s assets and found a home for them in Silver Lining, which was created by AtGames as part of a larger R&D effort to build scale-out cloud compute capabilities.

Silver Lining is working with ARM and Foxconn it has run successful proof-of-concept projects, and the systems will be introduced in early 2015, according to an email sent to media outlets.

Samsung and Apple back together

Samsung HQ Silicon Valley - MM picThe dark satanic rumour mill has manufactured a hell on earth yarn which claims that Apple is now back in love with Samsung and the pair have produced a new monstrous off-spring called the A9 chip.

According to the Korean IT News  Samsung Electronics has begun production of ‘A9,’ the application processor for Apple’s next-generation smartphone. It applies the 14nm FinFET microprocess for system semiconductors, for the first time.

Samsung began production of Apple’s A9 in the Austin plant in the US using the 14nm FinFET technology. Samsung has production lines capable of FinFET process production in Austin, US and Giheung, Korea, but began to produce A9 only in Austin as it is in the initial stage.

The outfit said that it would start production of the 14nm FinFET chip at the end of this year, but did not disclose whether the company received an order from Apple for the production of A9 chips or whether the production line is actually running.

Samsung is happy with the yield of the 14nm FinFET process, and supplied samples as good as finished products early enough.

The Austin plant began official production first at Apple’s request, and industry insiders said it is a move to produce the chip in the US, not Korea. They guessed that the Austin plant was chosen because of the next-generation chip’s problems with performance security and supply.

The initiation of the A9 chip production enabled Samsung to recover the foundry quantities from Apple, which have been discontinued for some time, and get the upper hand in the 14nm FinFET technology competition with TSMC, killing two birds with one stone.

However, this is clearly a burying of the hatchet between the two companies. Apple and Samsung stopped AP production as they were embroiled in patent litigation back in 2012. It appears that Apple has been lured back to Samsung with its winsome 14nm FinFET ways.

Relationship counsellors are quick to warn that it is early days yet.  Apple has been seeing other people during the break. Taiwan’s TSMC began the risk production of the 16nm FinFET plus (16FF+) process, and began to produce chips in July earlier than originally anticipated Q3.

Apple is effectively two timing the rivals. Shuttling between Samsung and TSMC, if TSMC’s production line is stabilised in the future, there is no knowing how SEC will respond.

Samsung’s foundry business was hit hard when Apple orders stopped. Although the entire semiconductor business is booming, securities companies predict that the system LSI business, including the foundry business, will suffer a loss to the tune of KRW800 billion this year. SEC is expected to recover sales loss to a certain extent with the production of Apple A9.

 

Intel’s China plans show mobile agenda

1220aIntel’s recently announced plans to invest shedloads into its Chengdu plant might be revealing much about its cunning plan for the future.

The move was a little unusual, as it did not come with the huge tax breaks and other sweeteners that Intel has required from Israeli governments and the US. This indicates that Intel needed a plant in China.

The move followed another similar-sized investment aimed at consolidating China’s wireless chip sector. The smart money suggests the two moves are linked Intel wants China to become a major centre for the company’s belated push into wireless chips.

Intel could be positioning both of these plants to manufacture smartphone chips. Intel failed to recognise the importance of mobile devices earlier, with the result that most of the market is now dominated by companies like sector leader Qualcomm, and mid-sized players like MediaTek, which mostly use chips based on an architecture supplied by European chip giant ARM.

Beijing wants to create a homegrown player to take down Qualcomm but its domestic chip sector consists of mostly small design houses that lack the resources to become major global players.

Unigroup, which is based out at the Tsinghua University, merged two of the biggest domestic players, Spreadtrum and RDA Microelectronics, into a single company. Intel joined that initiative in September, when it purchased 20 per cent of the new company for $1.5 billion. This makes it bigger than China’s largest chipmaker SMIC.

What Intel appears to be doing is getting itself onside with the Chinese and helping their domestic chip making plans with the idea of getting a foot in the door behind the bamboo curtain. Antitrust watchdogs are less likely to snap at its heels, or treat it as a problem, like they do Qualcomm and in the long term, it will have its fingers in the pies of a growing Chinese chip industry.

War between Nvidia and Samsung gets ugly

Newspaper Seller, 1939The war between Samsung and Nvidia has escalated and handbags are being distributed to the troops.

Samsung says that Nvidia has infringed several of its semiconductor-related patents and for making false claims about its products. This is a counter-suit following Nvidia’s lawsuit against the Korean company in September.

Samsung, which filed its lawsuit on Monday, is seeking damages for deliberate infringement of several technical patents, including a few that govern the way semiconductors buffer and use data.

Samsung said that Nvidia is guilty of false advertising when it says its “Shield” tablet sports the world’s fastest mobile processor, the Tegra. Samsung cites benchmarking studies performed by researchers at Primate Labs as proving that claim false.

Nvidia said it would review and respond to these new claims against it, and looked forward to presenting its case on how Nvidia GPU patents are being used without a licence.

Nvidia also pointed to a benchmarking study that supported its claim that the Tegra was the fastest mobile processor on the market.

Everyone knows that industry claims that “our chip is faster because we use x benchmark” always goes nowhere. It will be interesting to see what a court will make of them.

Intel expects Chinese small chipmaker rebellion

1900-intl-forces-including-us-marines-enter-beijing-to-put-down-boxer-rebellion-which-was-aimed-at-ridding-china-of-foreigners-Intel boss Brian Krzanich has been consulting his i Ching and expects ARM to be a spent force in China within a few years.

He claims that new semiconductor partners in China will migrate to Intel and give up on ARM technology more widely used in smartphones and tablets.

Intel this year signed deals with Rockchip and Spreadtrum Communications to use Intel’s technology to make chips for low-cost smartphones and tablets aimed at China’s fast-growing consumer market.

Spreadtrum and Rockchip specialise in smartphone and tablet platforms that are easy for manufacturers to use. At the moment they use ARM technology.

Intel has been writing agreements with the Chinese chipmakers which still allow them to make ARM-based chips, but Krzanich thinks that in a couple of years they will not see the point.

With Qualcomm offering high-end chips based on ARM and Taiwan’s MediaTek attacking the Chinese market with inexpensive chips also designed using ARM, adopting Intel’s architecture is the only way that anyone can offer a way to differentiate with better performance and features.

The only way for small manufacturers to compete is by going to Intel, he reasons.

Intel and Rockchip are working on an Intel-branded tablet SoC, with Rockchip contributing expertise on connectivity, graphics and its experience in China’s domestic market. Spreadtrum, is working with Intel on SoCs expected out next year.

Since both outfits are small, they lack the resources to make separate chips based on Intel and ARM technology over the long term.

Krzanich said Intel might collaborate with more companies there.

Cambridge boffin dismisses the existence of a general processor

martin-luther-nails-thesis-1David Chisnall, of Cambridge University become the general processor’s first atheist, by refusing to believe in the existence of a general-purpose chip.

Writing in Queue  he said that there was a general trend to categorise processors and accelerators as “general purpose.” Of the papers published at this year’s International Symposium on Computer Architecture (ISCA 2014), nine out of 45 referred to general-purpose processors; one additionally referred to general-purpose FPGAs (field-programmable gate arrays), and another referred to general-purpose MIMD (multiple instruction, multiple data) supercomputers, stretching the definition to the breaking point.

However Chisnall nailed a statement on the door of the Intel research division that there was no such thing as a truly general purpose processor and that the belief in such a device was harmful.

Risking the wrath of the great Intel God he said that many of the papers presented at ISCA 2014 that did not explicitly refer to general-purpose processors or cores did instead refer to general-purpose programs, typically in the context of a GPGPU (general-purpose graphics processing unit), a term with an inherent contradiction.

He said that a modern GPU has I/O facilities, can run programs of arbitrary sizes (or, if not, can store temporary results and start a new program phase), supports a wide range of arithmetic, has complex flow control, and so on. Implementing Conway’s Game of Life on a GPU is a common exercise for students, so it’s clear that the underlying substrate is Turing complete.

Chisnall argues that it is not enough for a processor to be Turing complete in order to be classified as general purpose; it must be able to run all programs efficiently. The existence of accelerators (including GPUs) indicates that all attempts thus far at building a general-purpose processor have failed. If they had succeeded, then they would be efficient at running the algorithms delegated to accelerators, and there would be no market for accelerators.

Nvidia turns in good results

nvidia-gangnam-style-330pxGraphics chip maker Nvidia posted higher fiscal third quarter revenue than many of the cocaine nose jobs of Wall Street expected.

Revenue in the fiscal third-quarter ended Oct. 26 was $1.225 billion, up 16 percent from the year-ago quarter, compared with Wall Street’s random guess of $1.202 billion.

For the current fourth quarter, Nvidia said it expects revenue of $1.20 billion, plus or minus two percent. Analysts on average expected fourth quarter revenue of $1.198 billion.

Third-quarter net profit was $173 million compared to $119 million a year ago.

Nvidia did better by focusing on using its Tegra chips to in entertainment and advanced navigation systems in cars made by companies including Volkswagen’s Audi, BMW and Tesla.

In the third quarter, revenue from Tegra chips for automobiles and mobile devices jumped 51 percent to $168 million. Nvidia’s PC graphics chip business expanded 13 percent to $991 million.

Intel’s Skylake is an escargot

Cooked_snailsWord on the street is that Skylake processors may be delayed.

Intel has been promising that Skylight will be in developers paws early 2015 ahead of shipping “by the second half of the year.”

Digitime’s Joanne Chen claims that Skylake will be held back so as not to dent sales of Broadwell-based notebooks.

The report claims that Intel has set the RTM schedule for the Skylake to the 37-47th weeks of 2015 which is the end of August to the end of October.

The later schedule, Chen claims, would damage the release of Windows 10 notebook sales, because Skylake-based models will not be able to reach retail until late September.

“Windows 10-based notebooks are already facing many negative factors that could impact their shipments in 2015: most enterprises are expected to finish their PC replacement by the end of 2014 after Microsoft terminated support for Windows XP; Microsoft will offer free upgrade to Windows 10 for existing Windows 8/8.1 notebooks; and Windows 10 lacks attractive features,” Chen wrote.

While it is possible that Intel has allowed itself room in September for a later ship date without exposing itself to accusations of breaking its promises, the matter of Windows10 is important. The Skylake delay is expected to worsen Windows 10-based notebooks prospects, and may in turn weaken the notebook market’s performance in the second half of 2015.

Chip price war unlikely

Normans_BayeuxSamsung does not expect a price war to break out in the semiconductor industry next year even though it is ramping up capacity.

CEO Kwon Oh-hyun said that he will have to “wait and see how things will go next year, but there definitely will not be any game of chicken”,

Memory chip makers have reported strong profits this year thanks to better-than-expected demand for PCs and servers.  Most analysts believe industry conditions will remain favourable in 2015.

But Samsung’s plan to invest $14.67 billion in a new South Korea chip plant stoked concerns about the industry’s profit outlook. Some investors worry that the firm could ramp up supply and undercut prices to squeeze Hynix and Micron’s bottom line.

Cooler heads have pointed out that Samsung’s plant will not begin production  until 2017. They say margins in the memory business are important to Samsung, given the mobile division’s falling smartphone profits.

 

 

Kaveri makes this Acer Aspire E5 sing

Acer E5-551_1AMD has been making a big noise about its Kaveri chip range of Mobile APUs and we got a look at what it could do to Acer E5-551 15.6 laptop.

AMD A-Series APU family, codenamed Kaveri, is starting to pop up in systems from Acer, Asus, Dell, HP, Lenovo, Samsung, and Toshiba. According to AMD, the new mobile APUs mark the debut of Heterogeneous System Architecture (HSA) features and Graphics Core Next (GCN) Architecture for mobile. Acer’s Aspire E5 is first to arrive in the UK and will be sold through Debenhams so this makes it interesting from a user or a channel perspective.

The AMD A8 Kaveri chip is now powering the new Acer E5-551 laptop is designed for multi-tasking from gaming and light photo-editing to running multiples instances of software and streaming music. The E5-551 features a DVD reader and writer for long plane flights where you need a better movie. It is also well placed for the school market as it can practically anything a school kid could throw at it.

The machine is priced for about £409 which makes it a good budget model for smaller companies who are providing portables for those who have to do work, rather than just look at tablets. Normally there is not much between computers in this range. Most have WXGA screens based on TN technology and conventional hard drives and plastic bodies.

 Acer E5-551_9The Spec

Processor: AMD Quad-Core A8-7100

Mainboard: AMD A76M

Memory: 8192 MB, DDR3L, 1600 MHz, dual-channel, 2 memory banks, both filled

Graphics adapter :AMD Radeon R6 (Kaveri), Core: 533 MHz, DDR3L, shared memory, bus: 64 Bit

Display: 15.6 inch 16:9, 1366×768 pixel, AU Optronics AUO47EC, TN LED

Connections2 USB 2.0, 1 USB 3.0, 1 VGA, 1 HDMI, 1 Kensington Lock, Audio Connections: 1 combo audio in/out, 3.5 mm jack, Card Reader: SD/SDHC/SDXC,

Networking: Realtek RTL8168/8111 Gigabit-LAN (10/100/1000MBit), Atheros Communications AR9565 Wireless Network Adapter (b g n ), 4.0 Bluetooth

Optical drive: Matshita DVD-RAM UJ8E2Q

Operating System: Microsoft Windows 8.1 64 Bit

Weight: 2.5 kg

Battery: 56 Wh Lithium-Ion

 Acer E5-551_10Design

The Aspire A5 design is, bluntly, a bit of a snooze and not its strongest feature. Certainly, it is thinner than many laptops in this range but it is not that thin. It is lighter and easy to carry than others we have had. Certainly in comparison to the bricks that are available in this price range it is certainly not bad. Besides, for £409 design is something that happens to other people’s computer.

The plastic case also seems to pick up every type of grease that a hand can generate and make the machine slimy.

 The chip

The important part of this computer is an AMD Quad-Core A8-7100 which is a quad core with eight GPU cores to carry out the AMD Radeon R7 series graphics. The idea of the Heterogeneous System Architecture (HSA) is that it allows the CPU and GPU to work together by quickly dividing and directing the right tasks toward the appropriate cores to improve performance and efficiency. The chip managed to make running software a breeze and it even passed our Firefox running trillions of opens tabs trick without any problems.

 The Graphics

The graphics on the Kaveri platform is far better than Intel’s equivalent HD graphics cores integrated into the Haswell chips. This makes Kaveri armed laptop owners able to play reasonably demanding games and videos. Playing Civ 5 was not only possible it was a quality game on Acer E5-551, something that I have only seen on our other home laptop, which has a dedicated card, when there was nothing else running. HD films came out very well, limited only by the screen which for a budget machine was still not too bad.

Connectivity

The machine has an HDMI socket, Ethernet connection, two USB 2.0 sockets, a USB 3.0 connection and a VGA out.  Not bad not great.

Communication

The Atheros AR9565 Wi-Fi module supports the IEEE 901.11 b, g and n standards,  Bluetooth, version 4.0. The signal connections were fairly standard and nothing to right home about

Battery

The 56 Wh Lithium-Ion managed 4.5 half hours before needing a recharge, which was not too bad given the thrashing we were giving it.  Again, this is due to the lower power draw of the AMD A8 chip.

Noise and heat

There was no noise coming from the machine at all. This was unheard of in any laptop in this price range, particularly one with an ordinary hard drive. One could almost believe it is fanless. The heat levels were also not to bad.  The laptop did get warm, but not hot and was comfortable to put on your lap.

Keyboard

The flat, roughened keys were not bad and were capable of fast touch typing.

Touchpad

The ClickPad which is a large, single-surface which you press when you want to ‘click’ was reasonable in comparison to the alternatives we have used.  Generally, if you are not going to have a touch screen on a laptop you are always going to be better with a mouse. When such use was impossible though once you got used to it was fine. Most of the touchpad surface can be pressed, which is interpreted as a left click except for in the lower right of the Clickpad.

 Hard-drive

The hard drive is 1TB which is more than enough for any portable music or movie collection.  It is not an SSD but we never heard it in operation.

Summery

The Acer E5-551 would be a normal budget laptop where it not for the Kaveri chip making it into something special. It enabled speed, enhanced graphics, extended battery life, and quiet operation something which we have not seen from Intel equivalents in the market.  It is still a budget machine, limited by the lack of a quality screen, but for that price it it is jolly useful.

The Eyes have it

four-stars

 

 

 

Good value for money.

 

AMD reads Synopsys

AMDlogoAMD has signed a deal with Synopsys, which gives it access to a range of designs and intellectual property on advanced 16nm/14nm and 10nm FinFET process technologies.

According to Kitguru, AMD will give Synopsys IP and engineering resources.

AMD gets interface, memory compiler, logic library and analogue intellectual property from Synopsys and will use it to develop future generations of chips to be made using 14nm/16nm as well as 10nm FinFET manufacturing processes.

Synopsys hires approximately 150 AMD IP R&D engineers and gains access to AMD’s interface and foundation IP. The move clearly saves AMD money although it is not so good in terms of resources, whereas Synopsys becomes stronger.

Synopsys provides chip designers a broad range of high-quality IP for integration into system-on-chips (SoCs) and delivering expert technical support. It makes its cash effectively developing non-critical areas of chips.

AMD has a huge library of various complex IP used in advanced microprocessors and graphics processing units.  AMD gets silicon-proven IP for the chips it will make in the next several years in exchange for interface and foundation IP as well as engineers. AMD claims that it will give it ability to “focus its valuable engineering resources on its ongoing product differentiation and IP reuse strategy.”

Mark Papermaster, AMD senior vice president and chief technology officer said that it will allow AMD to focus internal teams on designing the 64-bit processor, graphics and peripheral IP that makes the difference between AMD and its competition.

Synopsys can deal with AMD’s future SoCs.  The two companies have been working together for more than a decade.

 

Micron releases super dense SSD

mircon ssdMicron announced  a new SSD that uses its densest process and has an onboard chip that can program the memory to act as high performance SLC or high-capacity MLC flash.

Dubbed the M600 SSD, the drive uses Micron’s new 16 nanometer (nm) lithography with 128Gb NAND density.

Thanks to the greater density, the company could drop the cost per gigabyte to as little as 45 cents. The fact you can program the flash also reduces power use and improves write performance as much as 2.8 times over models without the feature.

Jon Tanguy, Micron’s senior technical marketing engineer said the M600 flash drive draws less than two milliwatts of power in sleep mode and averages 150mW during active use.

It has a sequential read rate of 560 MBps and can write at 510MBps. Its random read rate is up to 100,000 I/Os per second (IOPS) and it can write at 88,000 IOPS.

The SSD is based on an eight-channel Marvell controller that comes with government-grade hardware encryption using the 256-bit AES protocol.

Micron is selling the drive to manufacturers of corporate notebooks and ultra-thin netbooks, workstations and desktop PCs.

It comes in three form factors, a 2.5-in. SSD, an mSATA card and an M.2 memory stick. The mSATA and M.2 form factors come in capacities of 128GB for $80, 256GB for $140 and 512GB for $260. The 2.5-in. SSD comes in all those capacities and an additional 1TB version which will set you back $450.

 

Apple’s iPhone 6 chip is a lemon

CD153It looks like Apple’s new iPhone 6 will have the same performance of its older gizmos according to Hot Hardware benchmarking 

Normally one of the few things that is different about the new model iPhone is a that it comes with a better chip.  Last time it was the A7 System-on-Chip (SoC) which was the world’s first 64-bit smartphone processor.

Even Apple naysayers said that the A7 chip was rather good and dominated benchmark runs and consistently outperformed previous generation iPhone models.

However if Apple fanboys were hoping for a performance bump from the iPhone 6 and iPhone 6 Plus, both of which sport a custom A8 SoC they are going to be disappointed.

Hot hardware noted only modest gains compared to the iPhone 5s. The dual-core 1.4GHz Cyclone CPU and A8 GPU, the iPhone 6 scored 21,204.26 and a earned a place at the top of the chart, though not by much. The iPhone 5s scored 20,253.80 in the same benchmark so the iPhone 6 is less than 5 percent faster than the iPhone 5s.

What is strange is that not only was everyone expecting a better performance gain the iPhone 6 launch live stream implied that the difference would be huge.

Apple said that the all-new A8 chip is our fastest yet. Its CPU and graphics performance are faster than on the A7 chip, even while powering a larger display and incredible new features. And because it’s designed to be so power efficient, the A8 chip can sustain higher performance.

Well it is sort of true – the chip is faster, but not by enough for anyone to notice.

According to Apple, it offers 84x faster graphics performance than the original iPhone and is up to 50x faster in CPU performance.

Hang on a minute, Apple is comparing its current chip with that of the first iPhone which was released in  June 29, 2007. Of course, the iPhone 6 is going to be faster. However, this means that Apple is aware that its new chip is disappointing and it is trying to pretend it is hot.

Qualcomm releases Snapdragon 210

qualcomm-snapdragonQualcomm has released its new Snapdragon 210  chip which should mean more LTE smartphones running high-quality video.

Sticking two fingers up at its rival MediaTek, Qualcomm has built a 28-nm chip for the entry-level market. It has based around a 1.1 GHz quad-core Cortex-A7 CPU, Adreno 300-series GPU, supports up to 8 megapixel cameras, and has a resolution of up to 720p.  Not huge but you are trying to watch a movie on a screen the size of a beer mat.

It supports HD Video with high efficiency video coding support and supports the usual USB, Bluetooth, single-stream 802.11n WiFi, and NFC standards.

Qualcomm claims this is the first LTE-Advanced chipset to target the sub-US$100 phone category.

It is clear that Qualcomm wants to be the first of the starting blocks with this sort of technology.  The world is slowly moving to LTE but most of it is happy with its 3G phones.