Bronco Webinar 800x100 1

RTL Design Restructuring Explained

RTL Design Restructuring Explained
by Daniel Payne on 09-22-2016 at 4:00 pm

Modern SoC designs can use billions of transistors where transistors are grouped into gates, then gates grouped into cells, then cells grouped into blocks, blocks grouped into modules, and so on, creating a complex hierarchy. What a front-end designer conceives of logically for a hierarchy will differ from how an optimized physical hierarchy appears in order to meet physical implementation constraints in the back-end of the design process. Reasons to use a different physical hierarchy include:

  • Reduce die size and therefore costs
  • Increase utilization
  • Improve power consumption

Floor planning is the step in a design flow where the physical hierarchy is controlled and realized, however making changes to the design hierarchy has some side effects like:

  • Iterations of re-design
  • Function verification required

Let’s define some terms first.

RTL Building
Organizing all of the blocks in a design is how levels of a hierarchy are created and defining what the connectivity between blocks should look like.

In this simple example we’ve created two levels of hierarchy using four blocks. New logic blocks can be added into your design because of connectivity re-routing, feedthrough wires, power-related logic, or DFT issues like clock-gating or on-chip test controllers.

RTL Restructuring
Restructuring happens when we want to change our cell instances and connectivity in order to optimize our design or just meet some physical design constraints. In the following example we move block C from underneath B to be under A:

Ungrouping is where we remove an existing level of hierarchy, like taking blocks C and D from underneath B and placing them at the same level as block A:


The opposite of ungroup is to group cells, so here we take instances C and D then combine them into a single group:

Related blog – A Versatile Design Platform with Multi-language APIs

Partitioning is where we combine group, ungroup and move actions during our optimization process, like partitioning the top level:

A final restructuring concept is known as Clean, where we remove some design logic like cell instances, connections or process statements.

Approaches to Restructuring
For each new SoC you could just cobble together some customized scripts per project, or just start making manual edits to source files for restructuring. This approach doesn’t force you to buy anything, however it will cost you time to develop and perform so many manual edits, likely introducing bugs along the way.

Automation sounds like a more powerful approach that would scale well into the millions of instances, save engineering time and allow for more floorplan iterations to reach something optimal.

Defacto Solutions
Fortunately for the SoC design community there is an EDA vendor named Defacto Solutions that does have a tool in this important space, and they’ve named their tool STAR. The design flow for using STAR is shown below where it has input files like your RTL code or even gate-level netlists, then within the tool you do all of the restructuring (group, ungroup, move, remove, add), finally it will output your restructured RTL code along with any gate-level netlists:

Popular RTL languages like SystemVerilog, Verilog and VHDL are accepted as inputs. Your restructured RTL code is automatically created, saving you time, plus there will be comments added that explain all of the automated edits. Original comments, indentation and pre-compilation directives are maintained so that you can still read the code.

Related blog – A Brief History of Defacto Technologies

When you restructure your RTL design then the UPF file for power intent is also automatically updated for you, and you can even do a coherency analysis to double-check that the RTL and UPF are consistent with each other. Restructuring RTL also updates SDC files that you have for timing constraints. So STAR is really part of a unified flow for your design, beyond just RTL.

STAR Usage
Moving from theory to practice, here’s what actual users of STAR are doing with the tool:

Notice how IP-XACT files can be used as inputs along with RTL and connectivity files. IP-XACT became a standard in 2009 as a way to enable automated configuration and integration using the XML file format. This usage flow has five major steps to it:

[LIST=1]

  • Parse the design to extract all hierarchy and connectivity
  • Create the top-level
  • Restructure or partition as needed to reach physical requirements
  • Review DRC reports to check for warnings or errors
  • Output the restructured design

    Automated Restructuring Benefits
    Now that we know about what restructuring is, and how it helps optimize the physical hierarchy, just how useful is it in the real world? Users of STAR have shared:

    • Saving 10% in smaller die area. SOCIONEXT (Japan)
    • Remove the need to manually edit code (DAC customer)
    • Reducing project development schedule by several man-months
    • Eliminate long loops, optimize loopbacks, and remove redundant ports and many equivalent features in minutes
    • Build “correct-by-construction” sub-system at RTL level, dramatically reducing debug time

    Summary
    Stop using custom scripts and manual editing of RTL files for your SoC design restructuring, and instead consider using an automated approach from a commercial EDA vendor that has lots of happy customers.


  • ESL Architectural Power Estimation Support from TSMC — yes, TSMC

    ESL Architectural Power Estimation Support from TSMC — yes, TSMC
    by Tom Dillinger on 09-22-2016 at 11:00 am

    Electronic system level (ESL) modeling for system architecture exploration is rapidly gaining momentum. The simulation performance requirements for hardware/software co-design are demanding — an abstract model for SoC IP cores is required. Typically, soft IP will include a number of model configuration parameters. The SoC architect needs to optimize performance, power, and area (PPA) through evaluation of various design alternatives. Some soft IP cores include the capability to define a configurable instruction set architecture (ISA), for optimum performance of specific algorithm code.

    ESL-based design is benefiting from several standardization activities. The SystemC language definition has become the norm for model description, driven by the Open SystemC Initiative (OSCI). The definition of a synthesizable SystemC language subset has also guided IP core release — with high-level synthesis support from EDA vendors, SoC designers can realize both efficient model simulation and optimized cell-based implementations. The emphasis on transaction-level modeling (TLM) for core interface abstraction has provided architects with performance insights, without requiring implementation detail. A set of SystemC libraries release by the OSCI as part of the TLM2.0 standard has facilitated SystemC IP model interoperability in a complex verification environment.

    Parenthetically, verification engineers approaching ESL model simulation from an RTL background are likely dealing with an unfamiliar time-base representation. SystemC descriptions may be untimed, loosely timed, or approximately timed. A loosely-timed model reflects non-pipelined transactions — e.g., a complete, atomic read/write access operation has a corresponding timing interval, applying a blocking communication interface. In a loosely-timed model, there are two timing points — i.e., start of transaction, end of transaction. An approximately-timed model breaks transactions into individual steps, with a non-blocking interface. For example, in an approximately-timed model, there would be start/end request and start/end response timing points for each operation, which enables pipelined transaction simulation detail.

    SoC architects are rapidly adopting ESL modeling for system performance analysis. Yet, power dissipation is also a crucial optimization objective. How does an SoC architect integrate power estimation into the design exploration phase (long before physical implementation), with technology-based accuracy?

    I recently had the opportunity to chat with the team at TSMC who are working on this problem. They described a unique and innovative project underway at TSMC with key partners to address the SoC architect’s dilemma. “As much as 50% of the power may be saved if optimization and analysis is done at the early system level, whereas barely 10% or less of the power can be saved through late gate-level optimization. Optimization at the system level gives the earliest opportunity and greatest gain in system low-power design.” they noted.

    “Our customers and IP partners approached us, requesting assistance to define an ESL-based power modeling methodology.”, they highlighted.

    Initially, I was admittedly a bit surprised at this initiative — however, as they described the TSMC System-PPAmethodology, it became evident to me that TSMC is an ideal innovator to spearhead this activity. TSMC has an extremely close relationship with IP vendors, who develop/qualify/release their designs on TSMC process shuttles.

    The TSMC team briefly described the System-PPA IP power model generation flow — please refer to the figure below.

    IP vendors typically release a SystemC model for their IP, using an approximate-timing reference. A set of power-state API’s into the model is written. (This is a relatively low resource effort, according to TSMC.) This code is incorporated into the TLM 2.0 wrapper template developed by TSMC. IP power characterization is executed, and a power data look-up table (LUT) with specified PVT conditions is generated. To support this flow, TSMC has developed a Baseline Virtual Platform(BVP), where IP vendors and system developers can plug-in ESL level power models and perform power analysis and optimization using the TSMC-developed Virtual Platform Analyzer.

    Cadence/Tensilica, source of configurable DSP cores, and Arteris, source of Network-on-Chip (NoC) IP, have teamed up with TSMC to collaborate on the early System-PPA implementation activity.

    The goal of the TSMC System-PPA methodology is to provide a general, extendible TLM2.0 framework, where individual SoC IP cores each include the API wrapper, and can collectively be presented to the Virtual Platform Analyzer application.

    TSMC will be collaborating with additional IP partners in the future, and will be working with EDA vendors to help build momentum for this approach.

    With today’s system power design requirements, an ESL platform provides the most efficient and effective method for early system architecture exploration. It is essential that power optimization be an integral part of this analysis. TSMC’s System-PPA power modeling methodology enables effective and accurate power analysis during ESL definition.

    -chipguy


    Solutions for Variation Analysis at 16nm and Beyond

    Solutions for Variation Analysis at 16nm and Beyond
    by Tom Simon on 09-22-2016 at 7:00 am

    Variation is still the tough nut to crack for advanced process nodes. The familiar refrain of lower operating voltages and higher performance requirements make process variation an extremely important design consideration. As far back as the early 2000’s design teams have been looking for a better approach to model variation than simply adding margin. This just meant that you were trading performance for yield. Back then it was thought that statistical static timing analysis (SSTA) would provide a viable solution. However, that did not pan out right away. The sign-off tools available then simply were not up to the task.

    Another approach in use is called Advanced OCV, which attempted to take into consideration path lengths, by modeling chains of the cell in question with inserted parasitic elements. AOCV suffers from not including a number of significant design elements. Foremost amongst these is not looking at all of the arcs. Also it ignores side inputs. Compared to SSTA, AOCV tends to be either extremely optimistic or extremely pessimistic.

    According to Cadence, in a recent white paper they published, a Statistical OCV approach offers the best solution to modeling variation. Without the compute and data expense of SSTA, statistical OCV takes into consideration pin to related-pin dependencies, input slew, output load, and provides the variation information needed for the signoff flow. The Cadence paper is authored by Ahmed Elzeftawi, Sr. Principal Product Manager and Ken Tseng, Software EngineeringGroup Director at Cadence.

    The paper goes on to say that the Liberty Technical Advisory Board has created a unified Liberty Variance Format (LVF) document which includes OCV modeling coupled with timing, noise and power models. The Liberty Technical Advisory Board represents a broad consortium consisting of design tool providers, foundries and semiconductor companies. By taking advantage of statistical mean and Sigma values it’s possible for tools using this method to report timing values as probabilities or as discrete representations.

    Producing these models requires looking at every transistor in the cells and deciding which ones contribute most to variation. Each timing arc within a cell must be analyzed. Additionally, the impact of input slew and output load must be included in the resulting models.

    Cadence has extensive offerings for characterization and modeling that can be applied to standard cells, IO’s, memories and mixed signal blocks. Using Monte Carlo simulations as a reference they see very good correlation with their own technology for characterization. The Cadence Virtuoso Liberate characterization suite has many elements. The foundation tool in the suite is known as Virtuoso Liberate provides fast library characterization for standard cells and complex IO’s. The Virtuoso Liberate LV solution is useful for library validation providing functional equivalence and data consistency checking.

    To handle variation, they offer Virtuoso Variety which provides modeling of random and systematic process variation. Virtuoso Variety can generate Advanced OCV, Statistical OCV and LVF models. In addition, Virtuoso Liberate MX is useful for custom and compiled memories and Virtuoso Liberate AMS provides mixed signal characterization.

    The Cadence Innovus Implementation System can take advantage of these models to speed up timing verification and improve performance. At the end of the paper they provide as an example a 1 GHz design with the set up and hold slack for the top 200 paths. It’s pretty plain to see that there’s an average improvement of 150 picoseconds for set up and 200 picoseconds for hold.

    It was inevitable that statistical approaches would be used to deal with variation. I remember having discussions with design managers 10 years ago about the promise of statistical approaches. It’s nice to see now that they have come to fruition. Certainly at nodes beyond 16nm this technology will be more than a “nice to have”. If you are interested in reading the entire white paper, it can be found here.


    3 Small-Team Design Productivity Challenges Managed

    3 Small-Team Design Productivity Challenges Managed
    by Don Dingee on 09-21-2016 at 4:00 pm

    “Data management tools? We use small teams doing small designs. Each project only has two or three designers. Everyone uses the same EDA tools. Why do we need another tool for collaboration?” Good question. If you enjoy frequent meetings and redoing work because someone didn’t understand the status of IP blocks, the answers may not interest you. Continue reading “3 Small-Team Design Productivity Challenges Managed”


    eSilicon Revolutionizes Semiconductor IP Selection and Purchasing!

    eSilicon Revolutionizes Semiconductor IP Selection and Purchasing!
    by Daniel Nenni on 09-21-2016 at 10:00 am

    Design starts are the lifeblood of the semiconductor industry which is why we have been following the eSilicon STAR Platform since its introduction with great anticipation. The STAR platform was first launched about three years ago. Today, there are over 1,300 registered STAR users in 52 countries around the world.

    The ASIC business model is one of the reasons the fabless semiconductor ecosystem is what it is today, a force of nature. As an extension to the ASIC model, the STAR tools make it very easy to: Browse and try different IP, optimize your design, get quotes and compare with different foundry and IP options, and track your project online from start to finish. I guess the next step is a virtual reality interface so you can actually watch your design through the entire process?

    The eSilicon[SUP]®[/SUP] STAR platform is an automated online secure environment that provides a self-service, transparent, accurate, real-time experience from IC design through volume ASIC production. The STAR online design virtualization platform helps you manage complexity and make the right decisions on your ASIC journey from concept to volume production.

    Today eSilicon announced that STAR Navigator now includes automated, online quoting and purchasing capabilities for memory IP and IO libraries. I got a live demo last week and let me tell you I was seriously impressed! One click gets you IP silicon reports?!?!?!

    Given that embedded memories are a big part of chip design the ability do “what if” scenarios based on power, performance, area, and price give the average ASIC designer a HUGE advantage. Seriously, thoroughly evaluating all possible memory options is often skipped due to time constraints and that can jeopardize the success of your ASIC.

    “STAR Navigator simplifies the comparison of results across multiple technologies, architectures and other characteristics and takes the guesswork out of hitting PPA targets,” said Lisa Minwell, eSilicon’s senior director, IP marketing. “This goes much, much deeper than IP portals that serve as IP catalogs. Using STAR Navigator, designers can download front-end views, run simulations in their own environments and then purchase the back-end views of the IP and I/Os that best fit their design. The choice of optimized IP is now in the hands of the designer.”

    By the way, I worked with Lisa Minwell at Virage Logic back in the day. Before spending 8 years at Virage, she spent 15 years at Motorola and has been at eSilicon for the past 5 years so yes Lisa knows IP, absolutely. In fact, tomorrow Lisa will present on HBM/2.5D at the TSMC Open Innovation Platform Ecosystem Forum 2016:

    The next generation of high-performance computing, graphics and networking applications have increasing needs for bandwidth. High-bandwidth memory (HBM) combined with 2.5D technology offers a tremendous increase in capacity and performance. Increased capacity because of the stacked memory in a smaller area and increased performance because of the interposer and shorter signal routing. The interposer allows the integration of highly parallel connections to the memory stacks inside the package, therefore it is able to offer huge capacity and performance increases.

    This presentation will highlight the silicon characteristics of eSilicon’s HBM PHY in TSMC’s CLN28HPC technology. The presentation will also highlight TSMC CoWoS technology as well as complex ASICs that use high-bandwidth memory.

    The SemiWiki bloggers will be there in full force so stay tuned for in depth coverage. I hope to see you there!

    About TSMC 2016 OIP Ecosystem Forum
    The TSMC OIP Ecosystem Forum is a one-of-a-kind event that brings together the semiconductor design chain community and approximately 1,000 of director-level and above TSMC customer executives. The OIP Forum will feature a day-long, three-track technical conference along with an Ecosystem Pavilion that will host up to 80 member companies.

    About eSilicon
    eSilicon guides customers through a fast, accurate, transparent, low-risk ASIC journey, from concept to volume production. Explore your options online with eSilicon STAR tools, engage with eSilicon experts, and take advantage of eSilicon semiconductor design, custom IP and IC manufacturing solutions through a flexible engagement model. eSilicon serves a wide variety of markets including the automotive, communications, computer, consumer, industrial products and medical segments. Get the data, decision-making power and technology you need for first-time-right results. www.esilicon.com


    Low Power Design – a Server Perspective (Webinar)

    Low Power Design – a Server Perspective (Webinar)
    by Bernard Murphy on 09-21-2016 at 7:00 am

    Most of what you have read about design for low power has probably focused on mobile devices where power consumption constraints tend to outweigh performance objectives. These devices use aggressive power switching strategies, based on the reasonable assumption that parts or all of the device can be powered down at any given time and recovery times from power-down need only match reasonable human response times.

    REGISTER FOR THE WEBINAR

    But what about the other end of the mobile or IoT ecosystem – the cloud? Servers have to deliver high performance, they have unpredictable loads and the economics of that business pushes to maximize utilization to the greatest extent possible. As far as power-down strategies are concerned, the minimum switchable unit is typically a whole server. Some datacenters power down servers during light loading (packing virtual machines onto a subset of available servers), but this doesn’t help reduce peak power, which is where a viable datacenter wants to be operating most of the time.

    Historically, processor design teams have looked at power very late in design, when they can use gate-level netlists with accurate parasitics to get within ~5% of silicon measurements. That’s in part because power has not been a primary metric for processors (reasonable was OK) and in part because processor teams do a lot of hand-tuning for performance and power-saving techniques don’t usually help performance. The need for hand-tuning, by the way, applies even if you are using ARM cores if you’re shooting for GHz performance; check out what has gone into the high-performance ARM cores provided by the foundries.

    But that’s changed. In servers, low power is not about battery life. It’s about:

    • Heating which leads to increased leakage, which can lead to thermal runaway and at minimum compounds all the other problems
    • Cooling costs as mentioned earlier, for the datacenter as a whole and for the device because heat sinks increase server costs and active cooling increases maintenance costs
    • Performance problems because a device running too hot has to slow down and that makes customers unhappy
    • Reliability problems because increased heating increases delays and voltage drops in power lines which may tip some timing paths from passing to critical/failing
    • More reliability problems because increased heating increases resistance in power rails which can lead to electromigration in marginally-sized rails

    For all these reasons, power has become increasingly important for servers but getting an estimate late in design obviously isn’t very helpful, especially if it can take 6-8 weeks to figure it out.

    So what can be done if power-down strategies are off the table? Low-level clock-gating, manual or automated, is useful but many of those methods are used for second-order improvements and we want first-order help. That means we need to look at macro-level power-saving options with an understanding of architecture and intended usage. And that requires detailed use-case power analysis, over lots of use-cases. AMD has just published an article on how they went about reducing power in one of their server-class designs using ANSYS PowerArtist for power estimation at RTL where it was still practical to refine the design microarchitecture.


    Estimating power while the design is still at RTL and being able to generate estimates in minutes is essential to making this practical. Also important to AMD was very fast turn-around and multiple types of visualization for what-if analysis. Again this is because the big savings won’t come from automated gating, they come from designers figuring out where and how to make improvements in line with typical usage. So isolating and understanding power hotspots, followed by fast iteration (minutes) to experiment with power-saving scenarios is critical. (I expect ANSYS’ Big Data analytics in Seascape to further enhance this value proposition in the future.)

    It’s always interesting to hear a customer’s estimates of the impact of a tool. AMD said that using PowerArtist, they estimated that in idle-mode the design was using only 16% less power than when in full bandwidth mode and that more than 50% of this power was consumed by the clock distribution network alone. Hello opportunity for a first-order power reduction. Based on RTL analysis they were able to isolate areas where many cones of logic could be clock-gated. They also found, thanks to what they felt was quite accurate early RTL modeling of clock power distribution in PowerArtist, that they could in many cases move gating closer to the root in the clock tree, saving significant power consumption in the tree alone. In addition they found that flexibly adjusting the size of the queue based on queue utilization could reduce power. Between these factors, AMD was able to reduce idle power by 70% and also saw a significant improvement in active power.

    Though not mentioned in the AMD article, you should also know that PowerArtist generates an RTL Power Model (RPM) which can be read directly into RedHawk or SeaHawk for power-aware power integrity, thermal and EM analysis. So ANSYS has you covered for everything you need in server power analysis – for power consumption and optimization, for integrity, for heating and for reliability. Pretty cool solution (pun intended).

    REGISTER FOR THE WEBINAR


    Shakeup in Analog Rankings

    Shakeup in Analog Rankings
    by Bill Jewell on 09-20-2016 at 4:00 pm

    Last week Renesas Electronics announced an agreement to acquire Intersil Corporation for US$3.22 billion. This follows July’s announcement that Analog Devices Inc. (ADI) will acquire Linear Technology Corp. (LTC) for $14.8 billion. These deals will cause a shakeup in the analog IC market. According to IC Insights ranking of analog IC suppliers for 2015, ADI was number four, LTC was number eight and Renesas was number ten. Comparing 2015 analog rankings with 1995, 20 years ago, reveals some interesting changes. The table below shows analog rankings from 1995 from Gartner (then known as Dataquest), 2015 rankings from IC Insights, and our Semiconductor Intelligence (SC-IQ) forecast of 2016 rankings.


    Only three names from the 1995 list are on the 2015 list: STMicroelectronics (ST), Texas Instruments (TI), and Analog Devices (ADI). However, several of the analog businesses from 1995 are represented in 2015 under different names through various spin-offs, acquisitions and mergers.

    Philips, number two in 1995, spun off its semiconductor business as NXP Semiconductors in 2006. NXP is number seven on the 2015 list.

    National Semiconductor, number three in 1995, was acquired by TI in 2011, solidifying TI’s number one ranking. In 1995 National included Fairchild Semiconductor, which was spun off in 1997. TI also bolstered its analog position with acquisitions of Silicon Systems in 1996, Unitrode and Power Trends in 1999 and Burr-Brown in 2000.

    Motorola (number four in 1995) divested its semiconductor businesses as ON Semiconductor in 1999 and Freescale Semiconductor in 2003. In 2015 ON Semiconductor agreed to acquire Fairchild, with completion expected within the few months. ON’s number nine 2015 ranking does not include Fairchild. NXP acquired Freescale in December 2015 (NXP’s 2015 number seven ranking includes Freescale).

    Toshiba and Sanyo are both still in the analog business, but have dropped out of the top ten.

    Siemens, number nine in 1995 spun off its semiconductor business as Infineon Technologies in 1999. Infineon was number two in 2015. Some of Infineon’s growth was due to the acquisition of International Rectifier in January 2015.

    NEC (number ten in 1995) combined its semiconductor business, NEC Electronics, with Renesas Technology in 2010 to form Renesas Electronics. Renesas Technology was formed in 2003 as a joint venture of the semiconductor business of Mitsubishi (number 13 in the 1995 analog rankings) and Hitachi (number 16). Renesas was number ten in 2015.

    Thus only three companies in the 2015 ranking do not have ties to companies in the 1995 ranking. Skyworks Solutions was formed in 2002 with the merger of Alpha Industries and Conexant’s wireless division. Maxim Integrated was not even in the top 20 in the 1995 analog rankings, but moved up to number six in 2015 with over 10 times the revenue of 1995. Much of Maxim’s growth was driven by acquisitions including Dallas Semiconductor and Volterra and product lines from Vitesse and Zilog. LTC was number 18 in 1995 and number 8 in 2015. LTC’s growth was primarily organic, with few acquisitions.

    Maxim has been mentioned as both a potential acquisition target and a potential acquirer. Bloomberg reported in January 2016 that TI and ADI each investigated an acquisition of Maxim, but each company decided the price was too high. EE Times revealed in August that Maxim was in the bidding for Intersil before Renesas closed the deal.

    What will the analog rankings look like when 2016 is over? TI will certainly remain number one with over $8 billion in analog revenue. ADI will move up to number two with over $4 billion in revenue including LTC. The ADI and LTC merger will not be completed until 2017, but we have combined their revenues in 2016 for comparison. Infineon and Skyworks should be three and four, but there is a chance Skyworks could pass Infineon. Maxim and NXP should remain six and seven. ON will add over $300 million in analog revenue with the Fairchild acquisition and move up from number nine to number eight. However, ON will not pass anybody, just move up one with elimination of LTC. Renesas will add over $500 million in analog revenue with the Intersil acquisition and move up from number 10 to number 9 (Renesas and Intersil are combined for comparison in 2016 even though the deal will not be complete until 2017).

    The ADI and LTC combination will open up a spot in the top 10 in 2016. That spot should be taken by MediaTek, which added about $400 million in analog revenue with the acquisition of Richtek Technology in October 2015.

    Intersil: the remnants of semiconductor pioneers.

    The proposed acquisition of Intersil by Renesas will lead to the further disappearance of two pioneers in the semiconductor industry. Intersil was formed in 1999 when Harris Corporation spun off its semiconductor business. Harris had previously purchased the General Electric (GE) semiconductor business in 1988. The GE semiconductor business included RCA Solid State, which GE purchased from RCA in 1986. Harris began its semiconductor operation in 1967 with the purchase of Radiation Inc. Also in 1967 the original Intersil was founded by Jean Hoerni, one of the founders of Fairchild and the inventor of the planar process. GE bought Intersil in 1981. The Intersil name was revived with the 1999 spinoff from Harris.

    GE and RCA were early leaders in semiconductors. GE was one of the original licensees of the AT&T transistor patent and RCA was an early licensee. The two companies also did significant semiconductor research on their own. The history of early GE and RCA research is detailed in the excellent Transistor History website created by Mark P D Burgess.

    GE and RCA were leaders in consumer electronics and primarily used many of their semiconductors internally. They also sold on the merchant market. GE was a major supplier of discrete semiconductors in the 1960s. RCA was a top ten merchant semiconductor supplier in the mid-1970s.

    After the Intersil acquisition, Renesas Electronics will contain the remnants of five companies which began semiconductor research in the 1940s and 1950s: GE, RCA, NEC, Hitachi and Mitsubishi.


    Next Book Signing: Linley Processor Conference 2016!

    Next Book Signing: Linley Processor Conference 2016!
    by Daniel Nenni on 09-20-2016 at 12:00 pm

    It is a busy month for book signings but it is a pleasure to do it for the greater good of the semiconductor industry. It really is an honor to meet the people who keep our electronic devices on the leading edge of technology, absolutely.

    The Linley Processor Conference is on September 27[SUP]th[/SUP]and 28[SUP]th[/SUP] at the Hyatt Regency Hotel in Santa Clara, right across the street from Levi Stadium for you sports fans. This is an in-depth embedded processor conference for communications, IoT, and advanced automotive systems.
    Here are the keynotes:

    Day One – September 27
    Specialization Drives Processor Innovation
    Linley Gwennap, Principal Analyst, The Linley Group

    Day Two – September 28
    How Virtualization is Changing Networking
    Bruce Davie, CTP Networking, VMware

    I’m sure we all know Linley so let’s check Bruce out:

    Bruce Davie is CTO for Networking at VMware, and a Principal Engineer in the Networking and Security BU. He joined VMware as part of the Nicira acquisition, and focuses on network virtualization. He has over 25 years of networking industry experience, and was a Cisco Fellow prior to joining Nicira. At Cisco, he worked closely with leading service providers to enhance the capabilities of their networks. He led the team that developed multi-protocol label switching (MPLS) and contributed to the standards on IP quality of service. He has written over a dozen Internet RFCs and several networking textbooks. Bruce received his Ph. D. in computer science from the University of Edinburgh in 1988 and is an ACM Fellow.

    SemiWiki bloggers Tom Simon, Don Dingee, and I will be covering this event so stay tuned for a more in-depth look at some of the presentations. In fact, quite a few of the companies SemiWiki has worked with over the years are presenting:

    Session 1: IoT Edge
    Intelligent edge devices create a buffer between numerous IoT clients and the rest of the Internet. To minimize the cost and power consumption of client devices, these IoT edge systems may offload complex protocols such as security. This session, moderated by The Linley Group senior analyst Loyd Case, will discuss potential threats to IoT security and provide hardware and software solutions to defend against these threats.

    Protecting IoT Edge Devices from Malicious Physical and Software Attacks

    Fergus Casey, Senior R&D Manager, ARC Processors, Synopsys

    Session 5: SoC Connectivity

    The number of IP blocks in a processor continues to rise. Many of these blocks perform some sort of processing, including CPUs, GPUs, DSPs, and image processors (ISPs). The newest trend is to connect these heterogeneous IP cores using a cache-coherent interconnect, simplifying data sharing. This session, moderated by The Linley Group senior analyst Tom Halfhill, will discuss network-on-a-chip (NoC) and other interconnect IP for complex SoC designs.

    Building More Powerful Infrastructure SoCs from Edge to Cloud
    Jeff Defilippi, Senior Product Manager, ARM

    Coherency: The New Normal in SoCs
    Anush Mohandass, Vice President, Marketing and Business Development, NetSpeed Systems

    Implementing Cache-Coherent Hardware Acceleration for ADAS and Machine Learning
    Matthew Mangan, Corporate Applications Engineer, Arteris

    Session 9: Automotive and Vision
    Advanced driver assistance systems (ADAS) are evolving from delivering hazard warnings, to active collision-avoidance, to fully autonomous driving. Specialized computer-vision IP cores coupled with deep-learning processors will provide the eyes and brains for future smart cars. This session, moderated by Mike Demler, senior analyst at The Linley Group, discusses new IP and SoC architectures for ADAS and other automotive applications.

    A Neural-Network Oriented Vision DSP with Customized Hardware and Software Framework

    Liran Bar, Director of Product Marketing, CEVA

    High-Performance Vision Processors for HD Resolutions at Embedded Power Levels
    Mike Thompson, Sr. Product Marketing Manager, ARC Processors, Synopsys

    As Embedded Floating Point Becomes Ubiquitous, What Are Your Options?

    Dror Maydan, Senior Group Director, Tensilica Software Group, Cadence

    Following the sessions is a Q&A which is definitely worth your time. For more information: Linley Processor Conference 2016


    Taxi Industry – Survival by Near Death Experience

    Taxi Industry – Survival by Near Death Experience
    by Kevin Kostiner on 09-20-2016 at 7:00 am

    The Past We Lived Through
    The taxi industry has been a part of city and community landscapes since the “modern” taxicab first appeared on the streets of London in the late 1800’s. Since then, taxis have grown into a massive worldwide industry with strong regulation and protection in most jurisdictions. Such “rules” have limited competition while also minimizing any need to evolve, improve and innovate. Then along came something called Uber bringing with it an entirely new industry called “ride-sharing” and the taxi industry suddenly saw it’s protective glass house shattered to pieces overnight.

    Shielded from competition with regulations limiting the number of cabs on the street while dictating fee structures, taxi company owners concentrated on revenue. This took its toll on service and fleet quality. Outside of hailing a taxi on the street, cab response times were unpredictable with long customer wait times being the norm. And often the cab itself was a poorly maintained vehicle with torn and dirty seats, squeaky breaks and a driver with no cab side manner.

    The Present We’re Living With
    Uber shows up with a simple app where customers are picked up in minutes by a clean, recent model and well-maintained car. Service is the focus, with well-dressed, polite drivers who gladly help with bags. Some even provide water, candy, and other extras to make the ride more pleasurable.
    Wake up horn for the taxi industry!

    They say disruption is often the most effective catalyst to force positive change, especially in an old, entrenched industry. You could say Uber drove in and massively disrupted the taxi business!

    The Future As It Can Be
    Yes, taxis still have a LARGE and powerful presence. The taxi industry IS NOT rolling over and falling into the scrap yard. We’re talking about an immense global business estimated to generate over $ 100 billion/year in revenue. Just in the United States, taxi industry numbers are staggering:

    • Annual revenue in excess of $ 20 billion
    • Over 230,000 taxi drivers
    • Over 413,000 employees
    • Over 338,000 businesses

    In my market alone, the Las Vegas taxi industry generated over $ 425 million in revenue in 2015 representing almost 28 million trips! Taxis are definitely sticking around!

    Reinvention

    Yes, it’s been hurt by ride-sharing. In some markets, ridership and driver income are down by up to 50% over the prior year. But fleet owners are wide awake, recognizing mistakes made, opportunities overlooked and are exploring ways to not only level the playing field with the ride-sharing companies but drive technologies that will help them gain the lead.

    As Senior Vice President for IoE Connectivity at beamSmart, I have been working closely with taxi companies and regulatory bodies around the country. I can tell you this industry IS focused on bringing true technology into their members. While much of the buzz in this area has been around basic apps that look to deliver Uber-like functionality, the real solution for the taxi industry is a total end-to-end system that integrates dispatching, response, communication and safety into one platform. A product that beamSmart has already perfected.

    beamAtaxi, the “Uber” of Licensed Taxis


    Up front phases improve CDC analysis

    Up front phases improve CDC analysis
    by Don Dingee on 09-19-2016 at 4:00 pm

    Many tools find clock domain crossings (CDCs) in FPGA designs. Some don’t find the right ones since they don’t comprehend things like in-house synchronizer constructs. Some find too many based on misunderstanding intent, inaccurate constraints, and other factors that lead to noise. Continue reading “Up front phases improve CDC analysis”