What proved to be the most popular content on Semiconductor Engineering this year?
I like to look at what users find the most interesting topics, not because it directly influences what I write, but to get a sense of the subjects that are on most people’s minds. Some of it comes as no surprise. Content about new fabrication technologies tends to blow everything else away. While it directly affects very few of us, I think we all want to know the general direction of the industry and what we can expect five or ten years down the road. The problems of today are mundane by comparison. While there may be a little time bias, favoring articles published early in the year, the general readership numbers for Semiconductor Engineering continue to increase fairly rapidly which tends to balance things out quite nicely.
As far as my own articles, fifth place was taken by a relatively new article, Embedded FPGAs Going Mainstream? published October 27th. It looked at the growing attention being paid to FPGAs, especially those now available as IP that can be embedded into SoCs.
In fourth place was Decoding The Brain, an article inspired by a DAC keynote. How will decoding the brain of a fruit fly help us build better computers? Lou Scheffer says we have a long way to go before electronics catches up with image recognition. Other members of the industry added their opinions.
The bronze medal and quite a jump from fourth place was a roundtable titled Optimization Challenges For 10nm And 7nm that was published in three parts. Optimization used to be a simple timing against area tradeoff, but not anymore. As we go to each new node the tradeoffs become more complicated, involving additional aspects of the design that used to be dealt with in isolation. Part one on its own earned it the bronze medal.
Second place was a surprise as articles about software, IP and tools do not normally attract as many hits from our readership, but Will Open-Source Work For Chips? proved to be the exception. It looked at attempts to create open source semiconductor content and EDA tools. The industry is buzzing about that new possibility.
First place was total engineer porn. FinFET Scaling Reaches Thermal Limit looked at problems with finFETs that may imply a limit to the scaling that is possible. While somewhat ignored today, there is evidence that it could become a limiter.
My top blog was Abundant Change Ahead, which questioned if Moore’s Law would be the principal driver for the second half-century of semiconductor advancement. The whole notion of design could be transformed and even impact the role of the designer.
Turning our attention to the entire publication, we see the dominance of semiconductor technology articles.
In fifth place was The Future Of Memory, published June 16th.
It is unusual for a roundtable to get such high honors, even though they do have consistently good readership. This one looked at the DDR5 spec that was being defined and a new SRAM under development.
Fourth place went to Photonics Moves Closer To Chip, published June 20th.
Silicon photonics is resurfacing after more than a decade in the shadows, driven by demands to move larger quantities of data faster, using extremely low power and with minimal heat. Government, private funding ramps up as semiconductor industry looks for faster low-power solutions.
Third place and the bronze medal goes to 7nm Lithography Choices, published March 7th.
Chipmakers are ramping up their 16nm/14nm logic processes, with 10nm expected to move into early production later this year. Now, chipmakers are focusing on the lithography options for 7nm. The article lays out four possible scenarios for patterning the next generation of chips.
The silver belongs to 10nm Versus 7nm, published April 25th.
The silicon foundry business is heating up, as vendors continue to ramp their 16nm/14nm finFET processes. At the same time, they are racing each other to ship the next technologies on the roadmap—10nm and 7nm. The economics and benefits of moving to the next process node are not so obvious anymore.
And this year’s gold medal goes to What Transistors Will Look Like At 5nm, published August 18th.
As finFETs run out of steam after 7nm, what comes next? The debate is just beginning.