TIRIAS RESEARCH
WIRELESS/DSP NEWSLETTER
December 8, 2017

Qualcomm and Microsoft Announce the Always Connected PC 
Leave your power cord home.
At the Qualcomm Snapdragon Technology Summit in Hawaii, Microsoft and Qualcomm jointly announced the introduction of the Always-Connected PC, or AC-PC for short. They were joined by ASUS announcing the NovaGo, HP announcing the Envy X2, and Lenovowith an announcement of a new AC-PC to be made in January at CES. ASUS was the only OEM to announce pricing – a 64GB/4GB storage/memory configuration for $599 and a 256GB/8GB configuration for $799. The “Always Connected PC” is a term coined by Microsoft and Qualcomm to essentially represent a Windows PC using an Arm-based smartphone SoC, the Snapdragon 835, instead of an x86 processor from AMD or Intel. The advantages of such a PC are longer battery life by using a low-power processor that offers performance comparable to low- to mid-range traditional x86 processor and connectivity anywhere with an integrated cellular modem plus wi-fi.
I am surprised that Microsoft and its partners did not differentiate the platform in terms of form, function or even the business models with the carriers to drive more disruption into the PC market, but I am excited to get my hands on one to test for a range of usage models. Smaller, lighter, and longer battery life does come at a price.
Read the whole story on Forbes.

Broadcom’s Bid for Qualcomm Heats Up 
Broadcom’s hostile takeover bid for Qualcomm has shifted into overdrive. We still believe the acquisition is bad for the industry and recommend that Qualcomm accept more risk in its bid to remain independent. Here’s why.
After the first rebuff by Qualcomm, Broadcom pushed for Qualcomm to drop the acquisition of NXP in exchange for a higher offer. This would allow Broadcom to leverage the $35 billion in cash and cash equivalents that Qualcomm has available for the $105 billion acquisition. But with another rebuff by Qualcomm, Broadcom nominated 11 directors for Qualcomm’s board in hopes that it can pressure the company into accepting the acquisition.
Qualcomm officially rebuffed Broadcom on the basis that the offer undervalues the company. Given that Qualcomm is the largest patent holder of wireless and mobile technology, which affects just about every electronics market segment, and it is well positioned to continue this technology leadership with 5G, the company has a valid point. But, the other aspect of the rebuff is the potential damage to Qualcomm. As detailed in my previous articles, Broadcom does not value long-term or what might be deemed “risky” investments in innovation.
The fate of the US as a wireless technology leader hangs in the balance. Qualcomm’s investment in future technologies, its investment in wireless ecosystems, and its participation in the broader US business community as a corporate citizen are all at risk. Qualcomm, for the most part, is a conservative company that prides itself on being a technology leader. The acquisition of NXP is the largest acquisition and one of the riskiest move the company has ever embarked upon.
Rather than fighting an acquisition, a better strategy for Qualcomm might be to become even more aggressive and try to acquire other companies that would further its ambitions. This may appease shareholders and better fend off the Broadcom hostile bid.
Read the whole story on Forbes.

RISC-V Turns a Corner 
Many of you probably have not heard of RISC-V. It’s a new instruction set intellectual property  (IP) that is open sourced and offers an alternative to licensed IP from Arm and MIPS. As instruction sets go, RISC-V is relatively new, having just exited the University of California, Berkeley and entered the market in 2014 and is now managed by the RISC-V Foundation. But in those last 3 years, the instruction set has gained momentum largely due to its simplicity of design and business model.
RISC-V the with a fresh new instruction set that promises scalability from microcontrollers to supercomputers and offers user-defined extensibility, allowing companies to differentiate with custom instructions. In that regard, RISC-V is closest to the Tensilicainstruction set from Cadence, which also allows the addition of custom instructions. Custom instructions are useful for accelerating certain functions, often digital signal processing functions such as those found in audio and video functions and cellular modems.
With this open source approach, the goal for RISC-V is that the collaboration between research, academic and industry can work together to iterate the architecture faster than traditional business models.
To promote the RISC-V instruction set and associated ecosystem, a foundation was set up that allows companies to cooperate and provided a unified approach to managing the instruction set ecosystem. However the instruction set was still missing a key corporate commitment. At the 7th RISC-V Workshop held in November 2017, RISC-V finally got that commitment from Western Digital.
Western Digital is known for its storage products. What many do not realize is that the company ships over 1 billion processor cores within its products each year and is moving toward 2 billion per year. At the workshop, Western Digital CTO Martin Fink announced that over the next few years those billion plus processors will be transitioned over to RISC-V.
Read the whole story on Forbes.

Datacenter Processors Will Shed Power In 2018
Processor core performance stopped being a differentiator in the datacenter in 2017.
Compute accelerators are now the differentiator.
TIRIAS Research believes that in 2018, chip- and system-level low-power design will be an added differentiator in the datacenter.
The slowing of Moore’s Law changes both chip architecture and manufacturing economics, forcing chip and system designers to figure out new ways to get more performance without having to build new power plants.
Processor cores, once the center of the computing universe, have succumbed to offloading key performance-sensitive tasks. Network packet processing is best done on network accelerators. Switching and advanced interconnect topologies are best implemented on dedicated silicon. Matrix math and complex algorithm acceleration can be offloaded to DSPs, GPUs, FPGAs or dedicated accelerators.
The purchase habits of only a dozen or so cloud giant customers (like Amazon, Google, Microsoft, Alibaba, Tencent, Baidu, etc.) are now driving datacenter architecture evolution quickly and on a worldwide scale.
Read the whole story on Forbes.

Arm Server Processors Have Become Interesting Again
Marvell Technology Group and Cavium’s merger makes sense. The uncertainty here is whether Cavium shareholders should hold out for a better offer. Why might that be the case?
Read the whole story on Forbes.
[https://www.forbes.com/sites/tiriasresearch/2017/11/21/marvell-and-cavium/]
Qualcomm launched its Centriq server system-on-chip (SoC) a few weeks ago. The event filled in Centriq’s tech specs and pricing, and disclosed a wide range of ecosystem partners and customers.
We believe that Qualcomm’s strategy of focusing its ecosystem efforts on high value workloads and applications is sound. There were more 64-bit Arm datacenter workloads demonstrated at Centriq 2400 launch than we have seen gathered in one place, ever.
Read the whole story at The Next Platform.

IBM Launched an AI Platform Instead of a POWER9 Chip
IBM chose to ship its new POWER9 processor into the market this week by launching a single platform intended for AI applications, the IBM Power System AC922, instead of making a lot of noise about the POWER9 chip itself. Was it a good decision for IBM to launch POWER9 this way?
Read the whole story on Forbes.

AMD’s EPYC Return to the Datacenter Continues 
Microsoft Azure announced an advanced preview of its new AMD EPYC-based L-Series storage instances. ‘L’ stands for Linux Virtual Machines, and these instances are virtual CPUs with no offload acceleration (no GPU, FPGA, etc.).
TIRIAS Research doesn’t see any challenges to Microsoft broadening its relationship with AMD. The key to this Azure advanced preview of an AMD EPYC-based instance is that it starts Microsoft’s purchasing engine for a very flexible EPYC motherboard. Given that AMD is the only x86 instruction set alternative to Intel, we believe it is likely that Microsoft will eventually deploy EPYC will to run cloud instances of Windows Server applications.
HPE decided to resurrect its ProLiant DL385 brand for AMD EPYC. This is a huge statement of HPE support for AMD and EPYC in HPE’s core enterprise market. We did not think that AMD could win back a design in one of the very highest volume enterprise server brands, after HPE discontinued the AMD Opteron version of the design years ago.
HPE’s support for EPYC in its ProLiant DL385 server and Microsoft’s support for EPYC in its Azure public cloud L-Series storage instances show that AMD is getting customer traction with EPYC. More on EPYC’s market traction will become clear in 2018.
Read both of these stories on Forbes and more Forbes.

New Research
TIRIAS Research announces the publication of two research briefs on technologies intersecting AR/VR as part of the new AR/VR research service. In The Cloud is the New Training Ground for Virtual Reality, we look at the potential for VR to move to the cloud to enable mobile clients to deliver high-performance experiences otherwise inaccessible for the next decade. In Virtual Reality is the New Training Ground for Artificial Intelligence, we examine the intersection of VR and AI in the cloud to advance AI training through advanced visual and sensor simulation.

The TIRIAS Research AR/VR service is led by Simon Solotko, an accomplished VR and AR pioneer who serves as a mentor at the German Accelerator. With two decades of experience spanning HPC, semiconductors, and recent work with dozens of AR and VR companies, Simon is regarded as an AR and VR leader and subject matter expert. Recent articles in the media published by Simon include The Social Matrix: How The Cloud Can Activate A Billion Virtual Reality Users on UploadVR and Virtual Reality is the Next Training Ground for Artificial Intelligence on Forbes.

The TIRIAS Research AR/VR Practice employs a forward-looking framework for forecasting the evolution of AR/VR systems and experiences. We segment the field of emerging systems into tiers, tracking performance, capability, and user experiences by tier. We also look at intersections with adjacent technologies including machine learning, computer graphics, and computer vision to predict major inflection points that shape market transformation. Expert in understanding the market and technical challenges, TIRIAS Research provides direct support to teams seeking strategy guidance for technology investment, product development, and go-to-market.
For more information on the TIRIAS Research AR/VR service, please refer to the attached information or contact Simon Solotko at simon@tiriasresearch.com or by phone at 512-508-1054.

TIRIAS Research adds new analysts and services
TIRIAS Research is proud to announce dedicated coverage for Augmented Reality and Virtual Reality (AR/VR) and Machine Learning (ML) with the addition of leading industry experts to the TIRIAS Research team. Each expert comes to TIRIAS Research with extensive high-tech industry experience raising the cumulative experience of the TIRIAS Research staff to almost 200 years. With the introduction of these new research services, TIRIAS Research will be able to provide a full calendar of published reports and custom research in the near future.
The new team members include:

Scott McCutcheon, Research Analyst & Managing Editor 
With over 20 years of experience in the technology industry, Scott has analyzed data to uncover information to deliver insight for some of the world’s biggest IT brands. Scott has worked for market research firms Moor Insights & Strategy, Ipsos, and Kantar Millward Brown. Scott also chaired the Statistics Committee of the Semiconductor Industry Association (SIA) while working at AMD.

Simon Solotko, AR/VR Analyst 
Simon is an accomplished VR and AR pioneer who serves as a mentor at the German Accelerator. Simon brings a unique combination of technical and marketing expertise which he employs to help clients understand and navigate fast technology markets. Simon has two decades of experience spanning HPC, semiconductors, and recent work with dozens of AR and VR companies.

David Teich, Machine Learning Analyst
With more than 35 years of high-tech industry experience focused on B2B software, David’s experience spans development, systems consulting, sales, and marketing on platforms spanning mainframes, mid-sized systems (UNIX, Linux, and other), Apple and Wintel PCs, and mobile devices.

TIRIAS Research can be found in public!
You’ll also spot TIRIAS Research’s Principal Analysts attending these events over the next month:
·       Consumer Electronics Show (CES) – Las Vegas
Say Hi when you see us!

Research White Papers
All our white papers can be found here.
https://www.tiriasresearch.com/product-category/free-papers/

As always, we encourage your feedback
Kevin Krewell, Jim McGregor, Paul Teich
TIRIAS Research is a high-tech research and advisory firm, an independent third-party resource to high-tech companies. We provide custom research and advisory services on technologies, markets, and ecosystems to a select group of technology industry leaders. Our analysts have decades of in-depth expertise in silicon, software, and systems specification, design, and deployment.
About us

****************************************************
If this is a pass-along copy and you want to be added to our free newsletter mail list email us at newsletter (at) tiriasresearch.com.
We respect your privacy.  To remove your name from future mailings please reply to this message with the word BYE as the subject.

.

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list