Category: Chips

Samsung makes record chip earnings

scrooge-mcduckSamsung Electronics saw record chip earnings which meant it could gloss over the Note 7 smartphone fiasco in the fourth quarter.

The outfit, which is involved in an influence-peddling scandal surrounding President Park Geun-hye, said it expects profit growth in 2017, despite challenges arising from political uncertainty.

The political uncertainty appears to be whether the South Korean government will make Samsung executives accountantable for the scam. So far, the smart money is that it will probably not do that.

Samsung Electronics said in a statement that all this was creating an “uncertain business environment” although it did hint that there was also a changing political landscape abroad too.

All this poses a challenge to the execution of mid- to long-term business strategies, such as M&A and investment decisions and developing new growth engines.

Even so, it flagged higher earnings this year after a slow first quarter, when steeper marketing costs will eat into its bottom line as it tries to rebuild its reputation from the failure of its latest flagship phone.

The world’s top manufacturer of smartphones, memory chips and flat-screen televisions is counting on the booming chip market to continue driving growth and give the mobile business breathing space to rebuild its premium lineup.

The company forecast “stable demand” in 2017 for memory chips, which hit an all-time earnings high in the October-December period.

Fourth-quarter operating profit jumped 50 percent $7.93 billion which was its highest in over three years and matching its prior guidance. Earnings from the chips business soared 77 percent year-on-year. Revenue were flat.

In its mobile business, operating profit rose 12 percent in the fourth quarter as models such as the Galaxy S filled the void following the discontinuation of the fire-prone Note 7 in October.

Samsung claimed that defective batteries caused the Note 7 handsets to overheat and catch fire, and indicated that it may delay the launch of its next premium Galaxy S smartphone as it overhauls its product safety systems.

While the mobile business is struggling, the positive outlook for memory chips used in mobile devices and OLED televisions propelled Samsung’s shares to a series of record-highs this month.

A brain does not work like a computer chip

mybrainhurtsAccording to the BBC, a processor is the brain of a computer, but it seems that the hardware has neuroscientists baffled.

A paper published in PLOS Computational Biology wondered if more information is the same thing as more understanding. Eric Jonas of the University of California, Berkeley, and Konrad Kording of Northwestern University, in Chicago, who both have backgrounds in neuroscience and electronic engineering, reasoned that a computer was a good way to test the analytical toolkit used by modern neuroscience. However they had to admit that they were wrong.

They took an MOS Technology 6502 chip which was first produced in 1975 and famous for powering, among other things, early Atari, Apple and Commodore computers. It has 3,510 transistors and is simple enough to create simulation that can model the electrical state of every transistor, and the voltage on every one of the thousands of wires connecting those transistors to each other, as the virtual chip runs a particular program.

The simulation produces about 1.5 gigabytes of data a second—a large amount, but well within the capabilities of the algorithms currently employed to probe the mysteries of biological brains.

But brain science and electronic science started to diverge in the test. For example if you damage part of the brain you know what is going to be stuffed up. A chip though comes up with false positives.

Disabling one particular group of transistors prevented the chip from running the boot-up sequence of “Donkey Kong but allowed it to run other games.

If it were a brain you would think that transistors were thus uniquely responsible for “Donkey Kong” but the reality is that it is just part of a circuit which implements a much more basic computing function that is crucial for loading one piece of software, but not some others.

The bofins  looked for correlations between the activity of groups of nerve cells and a particular behavior but when they tried to apply this to the chip, the researchers’ algorithms found five transistors whose activity was strongly correlated with the brightness of the most recently displayed pixel on the screen.

Jonas and Kording know that these transistors are not directly involved in drawing pictures on the screen and they are only involved in the trivial sense that they are used by some part of the program which is ultimately deciding what goes on the screen.

Jonas said that neuroscience techniques failed to find many chip structures that the researchers knew were there, and which are vital for comprehending what is actually going on in it.

In fact, all the neuroscientists’ algorithms could detect in the chip was the master clock signal, which co-ordinates the operations of different parts of the chip.
In short, computers and brains have got as much in common as a packet of crisps has with the Empire State Building. This means that the BBC will have to find a new simile.

FTC sues Qualcomm for antitrust antics

monopoly (1)The US Federal Trade Commission filed a lawsuit against Qualcomm saying that the outfit had used “anticompetitive” tactics to maintain its monopoly on a key semiconductor used in mobile phones.

The FTC said that Qualcomm used its dominant position to impose “onerous” supply and licensing terms on smartphone manufacturers and to weaken competitors.

Qualcomm said in a statement that it would “vigorously contest” the complaint and denied FTC allegations that it threatened to withhold chips to collect unreasonable licensing fees.

The action is the last under current Democratic Chairwoman Edith Ramirez, who will step down soon after President Donald (Prince of Orange) Trump takes office.

Trump will name Republican Commissioner Maureen Ohlhausen as acting FTC chairwoman and will fill three vacancies that will reshape the agency. She has previously said that the lawsuit was based on a “flawed legal theory … that lacks economic and evidentiary support”.

In its complaint, the FTC said the patents that Qualcomm sought to license are standard essential patents, which means that the industry uses them widely and they are supposed to be licensed on fair, reasonable and non-discriminatory terms.

The FTC complaint also accused Qualcomm of refusing to license some standard essential patents to rival chipmakers, and entering an exclusive deal with Apple.

“Qualcomm’s customers have accepted elevated royalties and other license terms that do not reflect an assessment of terms that a court or other neutral arbiter would determine to be fair and reasonable,” the FTC said in its complaint.

In February 2015, Qualcomm paid a $975 million fine in China following a 14 month probe, while the European Union in December 2015 accused it of abusing its market power to thwart rivals.

 

Qualcomm details Snapdragon 835 processor

game-of-thrones-daenerys-dragonQualcomm has detailed its new Snapdragon 835 processor, which will power most of the leading Android smartphones this year.

Qualcomm SVP Keith Kressin said that the 835 was powerful enough to give AR and VR markets a kick up the bottom and could cause some serious problems for Chipzilla.

It has a Kryo 280 CPU under the bonnet but Qualcomm really is not talking much about that as it is the rest of the gubbins on the chip.

Qualcomm claims it is the first 10nm mobile processor, which will improve efficiency, and also by saying the CPU is “tightly integrated” with other components using the new Symphony system manager, which operates automatically yet can be customized by application developers.

This distributes work across the CPU, GPU, DSP, and more exotic components, letting the Snapdragon 835 work better than it would with CPU alone.

What will put the fear of Jehovah into Intel is that this is the first Qualcomm chip which can run Windows 10 on a mobile phone.

The Snapdragon 835 consumes 25 percent less power than the 820, according to Qualcomm. That means seven hours of 4K streaming video and two hours of VR gaming on a typical device, the company said.

Qualcomm can only do so much on power efficiency, it’s also introducing Quick Charge 4, which supposedly charges a phone to five hours of use in five minutes and is USB-C power delivery compliant.

The new Adreno 540 graphics chip improves 3D performance by 25 percent over the previous generation, Qualcomm said. But it enables features like HDR10, which improves colours; foveated rendering, which most clearly renders what you’re looking at rather than elements in the periphery of a scene; and low latency, which allows you to move your head smoothly around VR scenes. With one 32MP or two 16MP cameras running at the same time, the Snapdragon 835 supports various dual-camera functions.

The Snapdragon 835 will feature the X16 modem, which Qualcomm announced earlier this year and will be able to boost LTE to gigabit speeds. It has triple 20MHz carrier aggregation with 256-QAM encoding and 4×4 MIMO antennas.

We are predicting a rather good year for Qualcomm, when Vole gets its new Surface phones out later in the year.  This phones will be basically portable Windows 10 PCs and play nice with corporate networks.

Qualcomm faces $854 million for anti-trust

monopoly (1)South Korea’s antitrust watchdog has bitten a $854 million chunk from Qualcomm’s bottom line claiming the chipmaker used unfair business practices in patent licensing and modem chip sales.

Qualcomm said that it will appeal the decision which would be the largest ever levied in South Korea.

The Korea Fair Trade Commission which is very nearly the unfortunate acronym of a US fried chicken outfit  (KFTC) ruled  Qualcomm abused its dominant market position and forced handset makers to pay royalties for an unnecessarily broad set of patents as part of sales of its modem chips.

Qualcomm also restricted competition by refusing or limiting licensing of its standard essential patents related to modem chips to rival chipmakers such as Intel, Samsung  and MediaTek , the regulator said, hindering their sales and leaving their products vulnerable to lawsuits.

The regulator ordered Qualcomm to negotiate in good faith with rival chipmakers on patent licensing and renegotiate chip supply agreements with handset makers if requested – measures that would affect the U.S. firm’s dealings with major tech companies including Apple, Intel, Samsung and Huawei if upheld.

The KFTC said it began its investigations into Qualcomm’s practices in 2014 following complaints from industry participants, but did not name specific companies.

Foreign companies including Apple, Intel, MediaTek and Huawei expressed their views during the regulator’s deliberation process, KFTC Secretary General Shin Young-son told a media briefing in the country’s administrative capital.

Qualcomm said it will file for an immediate stay of the corrective order and appeal the decision to the Seoul High Court. The firm will also appeal the amount of the fine and the method used to calculate it.

“Qualcomm strongly disagrees with the KFTC’s announced decision,” it said in a statement.

While the fine is big, analysts said that the KFTC’s orders for Qualcomm to alter its business practices have bigger future implications for the chipmaker. The ruling forces the company to license patents for some of its chips to rivals such as Intel, which has been competing hard to land its modem chips in mobile phones.

 

SK Hynix expands memory chip operations

OLYMPUS DIGITAL CAMERA

SK Hynix is investing $2.7 billion in its home country and China to boost memory chip production.

There is a bit of a boom in demand for memory chips and Hynix wants to take advantage of that.

Most of the investment will build a new plant to make NAND flash chips used for long-term data storage.

A spokes Hynix said that to grow further, it is important to secure production facilities in advance to deal with NAND Flash market growth to be led by 3D NAND solutions.

Cash will be spent on the NAND chip plant which will be in South Korea, and a big chunk to boost DRAM capacity at its existing facilities at Wuxi, China.
Analysts think strong demand for memory chips will likely continue in 2017 as it will take several years for capacity spending announced by key players to yield meaningful production growth.

Supply will also be constrained as chipmakers shift to new production methods. In the NAND market, many manufacturers are converting existing production lines to high-end 3D NAND technology.

AMD’s Zen will be branded Ryzen

rizen-300x300AMD’s Zen will have a brand which sounds suspiciously like a drug for erectile dysfunction, a chemical weapon extracted from caster beans or a gospel group

The Ryzen brand will be applied to desktop and mobile variants. It had been formerly known as Summit Ridge. AMD has been telling the world that Ryzen is pronounced Rye-Zen (like horizon, without the “ho”). It has not said what the Rye means, but tired and emotional hacks suggested that it might be inspired by Jack Daniels – as indeed so are they.

Zen-based processors will target desktops, servers, and mobiles device, but the first wave of products will be targeted at the performance desktop market, where gamers and VR continue to spur growth. PC gaming hardware is predicted for 35 percent growth from 2015 to 2020, and VR-capable PCs are predicted to grow by a factor of 10 by 2020. Interest in eSports continues to skyrocket as well.

AMD has disclosed that one of the high-end options in the initial Ryzen line-up will feature eight cores (16 threads with SMT) and at minimum a 3.4 GHz base clock, with higher turbo frequencies. That processor will have 20MB of cache – 4MB of L2 and 16MB of L3. AMD is also hyping up what it is calling SenseMI technology. AMD’s SenseMI is a set of sensing and adapting technologies, including what AMD is calling “an artificial intelligence network” inside every Zen processor. There are five main features of SenseMI that include Pure Power, Precision Boost, and Extended Frequency Range (or XFR), along with the neural net prediction algos and smart prefetcher.

In fact, there is not that much exciting in SenseMI – it is just an updated branch predictor, prefetcher, and power and control logic in Zen. It might have some sexed up algorithms but it is not the ground breaking technology it is claimed.

AMD said that the AM4 platform for Ryzen will be have all the usual gubbins you expect to see on a modern PC enthusiast platform. AM4 motherboards will use DDR4 memory and feature PCIe Gen 3 connectivity, and support for USB 3.1 Gen 2, NVMe, and SATA Express.

Another thing that Ryzen will have is something called Pure Power which senses what’s going on with the processor’s current workload and provides a closed-loop control system to scale performance and power, over what AMD is calling its Infinity Fabric.

This monitors temperatures, frequencies, and voltage, and adaptively controls each element to optimize performance and power usage.

Another thing that Ryzen will have is something called Infinity Fabric which is interconnect technology used to link the individual bits of IP within a processor and the processor to other IO.

AMD wants this tech to scale its products, beyond moving to new process nodes, increase performance and efficiency, scale bandwidth, improve latency and the chip’s QoS.

Infinity Fabric consists of two key elements, a scalable control fabric and a scalable data fabric. The scalable control fabric has all the central control elements, with small remote elements that are dispersed in each different block of the SoC.

However the early indications suggest that Ryzen is rather good and should put the fear of Jehovah into Intel – particularly if AMD keeps the price down.

AMD releases AI based Radeons with basic instinct

BasicInstinct002AMD is announcing a new series of Radeon-branded products today, targeted at machine intelligence and deep learning enterprise applications.

Dubbed the Radeon Instinct, the chip is a GPU-based solution for deep learning, inference and training. AMD has also issued a new free, open-source library and framework for GPU accelerators, dubbed MIOpen.

MIOpen is made for high-performance machine intelligence applications and is optimized for deep learning frameworks in AMD’s ROCm software suite.

The first products are the Radeon Instinct MI6, the MI8, and the MI25. The 150W Radeon Instinct MI6 accelerator is powered by a Polaris-based GPU, packs 16GB of memory (224GB/s peak bandwidth), and can manage 5.7 TFLOPS of peak FP16 performance when the wind is behind it and it is going downhill.

It also includes the Fiji-based Radeon Instinct MI8. Like the Radeon R9 Nano, the Radeon Instinct MI8 features 4GB of High-Bandwidth Memory (HBM) with peak bandwidth of 512GB/s. AMD claims the MI8 will offer up to 8.2 TFLOPS of peak FP16 compute performance, with a board power that typical falls below 175W.

The Radeon Instinct MI25 accelerator uses AMD’s next-generation Vega GPU architecture and has a board power of approximately 300W. All the Radeon Instinct accelerators are passively cooled but when installed into a server chassis you can bet there will be plenty of air flow.

Like the recently released Radeon Pro WX series of professional graphics cards for workstations, Radeon Instinct accelerators will be built by AMD. All the Radeon Instinct cards will also support AMD MultiGPU (MxGPU) hardware virtualisation.

Intel puts in new planning application for Irish plant

juteworks_tcm4-572826Intel Ireland is lodging a fresh planning application with Kildare County Council for an estimated $4 billion new chip manufacturing facility at its headquarters in Leixlip.

It is not clear if the project will go ahead as Ireland is competing with other locations, most notably Israel, to land the investment. It received a 10-year permission for the plant in 2013, which its local management said “sits ready to be used when the corporation needs it”.

Since then the standard design of Intel’s “fabs” has changed and new planning approval would be needed.

The latest application is for a smaller facility than the one for which it got permission in 2013, which at the time was estimated could create more than 3,000 jobs during construction and fit out, and would cost $4 billion.

Intel said the new application, as well as reducing the footprint, would also site the proposed manufacturing plant further back from the N4 road. It would also be lower in height than the 2013 version.

The latest application also seeks permission for a car park to hold 2,200 cars, indicating that, if the project proceeded, it would likely provide a significant permanent employment boost to the area.

Apparently Chipzilla Ireland is still waiting on Intel HQ to give the project the nod.

Never mind AI, Intel has Nervana

3d2a123ddc312423225755a14fe7db2dChipzilla’s billion dollar investment in Nervana might be the key to making its server chips more intelligent.

Intel is laying out its roadmap to advance artificial intelligence performance across the board and Nervana technology appears to be everywhere.

The high-performance silicon market is dominated by GPUs. However, with Nervana inside, Intel hopes its new corporate tech with its a fully-optimized software and hardware stack will give that business model a good kicking.

Nervana hardware will initially be available as an add-in card that plugs into a PCIe slot. The first Nervana silicon, codenamed Lake Crest, will make its way to select Intel customers in H1 2017.

Intel is also talking about Knights Mill, which is the next generation of the Xeon Phi processor family. Intel said that Knights Mill will deliver a 4x increase in deep learning performance compared to existing Xeon Phi processors and the combined solution with Nervana will offer orders of magnitude gains in deep learning performance.

Diane Bryant, Executive VP of Intel’s Data Center Group said that the Intel Nervana platform to produce breakthrough performance and dramatic reductions in the time to train complex neural networks.

Intel CEO Brian Krzanich said that Nervana’s technologies will produce a 100-fold increase in performance in the next three years to train complex neural networks, enabling data scientists to solve their biggest AI challenges faster.