Sign up for our Technology Newsletter 
Is Samsung Readying A 10nm SoC?
Of course, it is that time of the year. Apple, Qualcomm, MediaTek and now Samsung will have 10nm SoCs ready for phones in early 2017. Of course Samsung wants to use its own 10nm SoC in the Galaxy S8 that is expected in late February 2017, but probably with a mix of 10nm Snapdragon too.
Samsung’s next generation Exynos’ name is very uninspired. You don’t call your much better chip just the Exynos 8895, but that might not be the final name.
The Korean giant went from Exynos 7420 for Galaxy S5 and first 14nm for Android followed a year after with Exynos 8890 still 14nm but witha custom Exynos M1 “Mongoose” plus Cortex-A53eight core combination.
The new SoC is rumored to come with a 4GHz clock. The same leak suggests that the Snapdragon 830 can reach 3.6 GHz which would be quite an increase from the 2.15Ghz that the company gets with the Snapdragon 820. Samsung’s Exynos 8890 stops at 2.6GHz with one or two cores running while it drops to 2.3 GHz when three of four cores from the main cluster run. Calls us sceptics for this 4GHz number as it sounds like quite a leap from the previous generation.
Let us remind ourselves that the clock speed is quite irrelevant as it doesn’t mean anything, and is almost as irrelevant as an Antutu score. It tells you the maximal clock of a SoC but you really want to know the performance per watt or how much TFlops you can expect in the best case. A clock speed without knowing the architecture is insufficient to make any analysis. We’ve seen in the past that 4GHz processors were slower than 2.5GHz processors.
The fact that Samsung continued to use Snapdragon 820 for its latest greatest Galaxy Note 7 means that the company still needs Qualcomm and we don’t think this is going to change anytime soon. Qualcomm traditionally has a better quality modem tailored well for USA, China, Japan and even the complex Europe or the rest of the world.
Courtesy-Fud
ARM Shows Off 10nm Chip
ARM’s collaboration with TSMC has finally born some fruit with the tapeout of a 10nm test chip to show off the company’s readiness for the new manufacturing process.
The new test chip contains ARM’s yet-to-be-announced “Artemis” CPU core which is named after a goddess who will turn you into deer and tear you apart with wild dogs if you ever see her. [The NDA must have been pretty tough on this chip.ed]
In fact things have been ticking along on this project for ages. ARM discloses that tapeout actually took place back in December last year and is expecting silicon to come back from the foundry in the following weeks.
ARM actually implemented a full four-core Artemis cluster on the test chip which should show vendors what is possible for their production designs. The test chip has a current generation Mali GPU implementation with 1 shader core to show vendors what they will get when they use ARM’s POP IP in conjunction with its GPU IP. There is also a range of other IP blocks and I/O interfaces that are used to validation of the new manufacturing process.
TSMC’s 10FF manufacturing process is supposed to increase density with scaling’s of up to 2.1x compared to the previous 16nm manufacturing node. It also brings about 11-12 per cent higher performance at each process’ respective nominal voltage, or a 30 per cent reduction in power.
ARM siad that comparing a current Cortex A72 design on 16FF+ and an Artemis core on 10FF on the new CPU and process can halve the dynamic power consumption. Currently clock frequencies on the new design are still behind the older more mature process and IP, but ARM expects this to improve as it optimizes its POP and the process stabilizes.
Courtesy-Fud
Chipmakers Advocating MIPS Open Source Moves
Comments Off on Chipmakers Advocating MIPS Open Source Moves
Qualcomm Atheros, Lantiq (part of Intel) and Broadcom have joined the Prpl Foundation.
For those who came in late Prple is the organisation set-up by Imagination Technologies to support open-source software on the MIPS architecture.
The big names follow CUPP Computing, Elliptic Technologies, Imperas Software, Kernkonzept and Seltech joined the foundation at lower levels.
In a statement the Foundation said that the newcomers to the prpl Foundation’s board of directors will participate at the executive level and appoint representatives to the technical steering committee and to engineering groups including the security.
So in other words the key players will be advocating an open source approach to MIPS.
Prpl, is open to other architectures, and focuses on “datacenter-to-device portable software and visualized architectures”, it said. Initial domains oem its hit list are: datacenter, networking, storage, connected consumer, embedded and IoT.
AMD To Power Samsung’s Digital Media
AMD’s Embedded R-Series accelerated processing unit, previously codenamed “Bald Eagle,” is powering Samsung’s latest set-back-box digital media players.
Bald Eagle was designed for high performance at low power with broad connectivity but mostly for digital signage.
It seems that new Samsung SBB-B64DV4 is intended for demanding signage applications that transform Samsung SMART Signage Displays into digital tools for a wide range of business needs.
The chipmaker claimed that by using its Embedded R-Series APUs, Samsung SBB media players for digital signage can manage HD graphics performance and support multivideo stream capabilities up to two displays, in a power efficient and ultra-compact form factor.
Scott Aylor, corporate vice president and general manager, AMD Embedded Solutions said that digital signage is a key vertical for the AMD Embedded business.
“The AMD Embedded R-Series APU enables leading digital signage providers to harness high levels of compute and graphics performance within a low-power design envelope. AMD Embedded Solutions help designers at Samsung achieve aggressive form factor goals and drive down system costs while providing the rich multimedia their digital signage customers’ demand,” he said.
The AMD Embedded RX-425BB APU combines an x86 CPU with an integrated, discrete-class AMD Radeon R6 graphics processing unit in a low-power configuration to minimize heat dissipation constraints and meet energy efficiency requirements.
The processor uses AMD’s latest Graphics Core Next architecture, created for advanced graphics applications and parallel processing capabilities.
Will AMD’s Kaveri Launch As Godavari?
Comments Off on Will AMD’s Kaveri Launch As Godavari?
The new desktop refresh according to SweClockers is going to end up with A10-8850K branding. The new processor will get a 100MHz faster turbo clock and is based on the same 28nm manufacturing process. The base CPU clock for the A10-8850K is 3.7GHz, the same speed as the AMD A10-7850K, but the Turbo clock will jump to 4.1GHz with the new one. The A10-7850K has 4.0 GHz top turbo clock and 720 MHz GPU speed for its GCN Sea Island GPU.
The new A10-8850K will get the GPU to 856MHz. The memory speed supported stays at 2133MHz and the socket of choice remains FM2+. The TDP stays at 95W.
As you can see this is a small evolution and you can expect some cool parts for AMD on the desktop side in the latter part of 2016, some eighteen months from now, in 14nm.
TSMC’s FinFet Coming In 2015?
TSMC has announced that it will begin volume production of 16nm FinFET products in the second half of 2015, in late Q2 or early Q3.
For consumers, this means products based on TSMC 16nm FinFET silicon should appear in late 2015 and early 2016. The first TSMC 16nm FinFET product was announced a few weeks ago.
TSMC executive CC Wei said sales of 16nm FinFET products should account for 7-9% of the foundry’s total revenue in Q4 2015. The company already has more than 60 clients lined up for the new process and it expects 16nm FinFET to be its fastest growing process ever.
Although TSMC is not talking about the actual clients, we already know the roster looks like the who’s who of tech, with Qualcomm, AMD, Nvidia and Apple on board.
This also means the 20nm node will have a limited shelf life. The first 20nm products are rolling out as we speak, but the transition is slow and if TSMC sticks to its schedule, 20nm will be its top node for roughly a year, giving it much less time on top than earlier 28nm and 40nm nodes.
The road to 10nm
TSMC’s 16nm FinFET, or 16FinFET, is just part of the story. The company hopes to tape out the first 10nm products in 2015, but there is no clear timeframe yet.
Volume production of 10nm products is slated for 2016, most likely late 2016. As transitions speed up, TSMC capex will go up. The company expects to invest more than $10bn in 2015, up from $9.6bn this year.
TSMC expects global smartphone shipments to reach 1.5bn units next year, up 19 percent year-on-year. Needless to say, TSMC silicon will power the majority of them.
Is The EU Going After Qualcomm
Qualcomm faces an antitrust investigation in Europe, even as it seeks to end a probe of its alleged monopoly practices in China.
Reuters reported that Qualcomm is looking for an amicable resolution of an investigation conducted by China’s National Development and Reform Commission (NDRC) over suspicions that it holds a monopoly in the Chinese telecoms market.
The investigation involves allegations that Qualcomm’s China subsidiary has been overcharging and exploiting its position in the wireless communications sector.
The antitrust probe of Qualcomm has been ongoing since last November, when the firm revealed that it was under investigation by the NDRC, though at the time it said the NDRC had not revealed the substance of the investigation.
In February, the NDRC declared it had received complaints against Qualcomm from the China Communications Industry Association, regarding its market position and patent fees it charged Chinese mobile phone manufacturers.
While the NDRC has ruled that Qualcomm does hold a monopoly in China, it has yet to decide whether the company has abused its position in the market.
Under China’s 2008 anti-monopoly laws, Qualcomm could face high fines, potentially topping $1bn.
In a statement to Reuters, Qualcomm said that it is seeking an amicable conclusion to the investigation. “Qualcomm executives discussed with NDRC officials several topics in an effort to reach a comprehensive resolution. We are continuing to cooperate with NDRC and cannot comment further,” the firm said.
Given that the NDRC is targeting at least another 30 foreign firms with antitrust investigations, including Microsoft and Volkswagen, critics have suggested that the monopoly law is being used to unfairly target overseas firms so that China can protect its native businesses.
Even if the China case is settled Qualcomm is now facing the prospect of a monopoly probe in Europe. Reuters has also reported the company could face a European Commission antitrust investigation following a complaint made four years ago by British software defined modem company Icera, a subsidiary of Nvidia.
Icera alleged that Qualcomm had engaged in anti-competitive behaviour by discouraging customers from doing businesses with Icera through patent related incentives and exclusionary pricing of chipsets.
While it was thought that the allegations had dropped from the European Commission’s agenda, the issue has resurfaced. It could even be fast-tracked following a similar monopoly case and subsequent fine made against Intel, which cost the company €1.1bn.
As yet, no official investigation has been opened by the European Comission. Qualcomm was contacted for a statement on both antitrust investigations, but the company has not yet responded.
Patents and their subsequent enforcement tend to play a major part in the technology industry as companies vie for market shares or state their supremacy. Qualcomm is no different, with the company having snapped up 2,400 patents from HP, including one for the now-defunct Palm technology, earlier this year.
Is Snapdragon A Security Flaw?
Security researcher Dan Rosenberg has told a Black Hat conference how it is possible to permanently unlock the bootloader on Android phones – provided they use a Qualcomm Snapdragon chip.
Rosenberg said that the flaw is in ARM’s TrustZone technology, which runs a trusted operating system and another for normal apps. This is supposed to improve device security, but in Qualcomm’s implementation, they cocked it up. It means that if a hacker gets access to the trusted operation part of the chip, it can run whatever application he or she likes.
This affects all known Android devices with a Qualcomm Snapdragon SoC, including the Nexus 5, the HTC One, and Samsung’s Galaxy Note 3, as well as the Moto X. The Samsung Galaxy S5 and the HTC One M8 have already been patched.
Can Qualcomm Move Forward?
Qualcomm has posted its smallest quarterly revenue increase since 2010, which saw its share price plummeting five percent in after hours trading.
Qualcomm reported its second quarter earnings on Wednesday for the three months to 30 March, and its revenue rose to $6.37bn during the period, up four percent from a year ago, with net profit up five percent to $1.97bn.
However, that was the smallest year over year percentage increase since the June quarter of 2010, when revenue declined by two percent, and was far lower than the quarterly growth rates of over 20 percent that Qualcomm investors have seen previously.
“We delivered another solid quarter, driven by demand for our leading multimode 3G/LTE chipset solutions and record licensing revenues,” said Qualcomm CEO Steve Mollenkopf in the earnings report, not mentioning that earnings reflected a much lower increase than seen in recent quarters.
“Looking forward, we are pleased to be raising our earnings per share guidance for the fiscal year. We continue to see increasing demand for our industry-leading chipsets and strong growth in calendar year 2014 of 3G/4G smartphones around the world.”
Qualcomm also forecast sales of between $6.2bn and $6.8bn for the April to June quarter, with the low end of that estimate representing a decline of one percent from a year ago.
It’s probable that while growing smartphone penetration in emerging markets is helping to keep the firm’s unit sales high, it’s also having an negative effect on Qualcomm’s average selling price (ASP) levels of mobile chipsets and devices.
Following Qualcomm’s earnings report, analysts said that the dip in revenue was attributable to a decline in sales in China as the country’s biggest network, China Mobile, prepares to launch a faster network with 4G, or LTE, technology, and customers are anticipating the launch before buying new smartphones.
Qualcomm now expects to make a profit of between $5 and $5.25 per share, five cents above its earlier projection, the firm said.
AMD Changes Kaveri
Since AMD officially launched its 4th generation A-Series Kaveri APUs and lifted the NDA veil from all press materials, we noticed that it has started to use a new term to define the structure of its new Kaveri APUs. As we reported last week, AMD is now talking about Compute Cores, which practically puts CPU and GPU cores on an equal footing, suggesting that there should not be any difference between them and that some tasks, previously limited to the CPU, can be done by the GPU as well.
If you take a look at the official AMD slide below which details the three new Kaveri APUs, the A10-7850K, A10-7700K and the A8-7600, you will notice that AMD lists the flagship as the APU with 12 Compute Cores or simply four CPU and eight GPU cores. Since the Kaveri APU is actually the first APU with HSA (Heterogeneous System Architecture) support, with hUMA, or equal memory access by both CPU and the GPU, heterogeneous queuing, which allows the GPU and CPU to have equal flexibility to create/dispatch work and an ability to talk about APU GFLOPS, or combined compute power of the entire APU, it makes sense for AMD to also talk about Compute Cores.
Of course, there are still some application specific tasks where the CPU or the GPU are much better, but, according to AMD, Kaveri is the first true APU, where the GPU is not just for gaming, it can actually do much more.
AMD Senior Manager Sasa Marinkovic, Technology lead for the Client Business Unit, said: “At AMD, we recognize that our customers often think of processors (CPUs) and graphics cards (GPUs) in terms of the number of cores that each product has. We have established a definition of the term “Compute Core” so that we are taking a consistent and transparent approach to describing the number of cores in our HSA-enabled APUs. A Compute Core can be either a CPU core or GPU core i.e. Kaveri can have up to 12 Compute Cores (4 CPU and 8 GPU).”
Although it does sound like a marketing gimmick, but actually is not due to HSA, it will definitely mark a new way for AMD to market/sell its APUs and it will definitely simplify the shopping experience for many casual buyers, more Compute Cores, more performance.