Advertisement

Blog

Putting the Customer in Analog Integration

Analog integration holds tremendous potential for transformation of our world. Just look at the emerging applications in areas like the smart grid, where smart meters now automatically measure electricity usage and communicate that usage on a unique network.

Fifteen years ago, the cost of such a network was completely prohibitive. Now, with advances and integration in both the metrology and communication, these networks are not only possible, they’re preferable and cost effective. Smart meters are just the tip of the iceberg, as machine-to-machine networks, or the Internet of Things, emerge. But who really pushes analog integration?

Let’s look at what pushed the first revolution in silicon technology. We all grew up in a world in which Moore’s Law dominated the advances in digital process technology and performance. Amazingly, my current iPhone is orders of magnitude more powerful than the 90MHz Pentium desktop I bought in college. I thought that was a pretty cool machine — actually it was, but times have changed.

Who really pushed Moore’s Law? Despite the fact that Moore’s Law seemed to proceed unimpeded for many, many years, there still needed to be customer demand fueling the consumption of high-performance digital chips. The same must be the case for analog integration, but now the opportunity is orders of magnitude greater.

Moore’s Law proceeded in the advent of the computing age. Each year, faster computers came out that could handle larger, more complex software. Business and consumer demand for computers was the single greatest force pushing Moore’s Law.

Now, analog integration offers the potential for multiple communication and sensing applications, such as smart grid, LED lighting, mobile communications, and more. All industries are pushing analog integration from different directions, with different needs. Furthermore, digital integration progressed under the quest of reducing one variable, the critical dimension. Analog integration progresses with the challenge and opportunity to shrink, pull together, and optimize multiple circuits from power management to signal processing to communications. Comparing digital integration to analog integration is like comparing black-and-white television to 3D color TV. The transformation is mind boggling.

Who’s going to drive analog integration? The creative ones will do it. There are no limits to the industries, applications, and new frontiers yet to be discovered. There are no limits to the number of tools and pieces that might fit together in a different way, performing beautifully in new, unique applications.

So, grab your paintbrush, sketch pad, tablet, or smart meter. Let us know your vision, which pieces you want to pull together, how you envision your next system operating — and not only operating, but operating better, faster, in new situations. We look forward to hearing from you.

5 comments on “Putting the Customer in Analog Integration

  1. Brad Albing
    January 25, 2013

    David – I personally have a strong interest in smart grid and LED lighting, so I'm quite intrigued (or I will be intrigued) as developments from the semiconductor manufacturers show up on the market. And yes, the creative guys will lead, so I'll expect amazing things from these folks with regard to integrated analog functions pertaining to LED lighting and smart grid. Should be exciting if they pull it off.

  2. eafpres
    January 27, 2013

    Hi David–great thoughts.  A couple years ago McKinsey had an article that showed a trend of computations/kwh vs. time.  The trend was linear upwards with a log scale for the computations/kwh axis.  The article can be found here:

    http://www.economist.com/blogs/dailychart/2011/10/computing-power

    The line of reasoning was that what had been happening since 1945 (Eniac) was decreasing the power per flop and that was the real underlying trend, not size or density.

    What I find interesting is deeper analog integration should conform to a trend like that as well; we have already seen wireless technologies rapidly dropping in power consumption.  As the sensing side (analog) becomes more integrated with the rest of the electronics, that should also help lower power/performance.  What do you think about analog integration vs. power consumption?

  3. David Andeen
    January 27, 2013

    Fascinating!  This appears to be right on, and yet another reason we'll see more applicaitons, like distributed, energy harvesting sensors.  I also loved seeing old names on that graph, like Commodore 64.

     

    You're absolutely right about RF energy consumption.  Here's a cool project…put together modern smart phone functionality with technologies circa 1995.  I wonder what that thing would look like, and how much power it would consume.

     

    Many analog sensing functions add very little power consumption, so I see this trend continuing, especially as microcontrollers get embedded into highly integrated analog parts.  I haven't done an actual computation, it seems like you would want to somehow couple in analog functions, like battery charging, while aren't going to involve so many calculations, but greatly improve battery life when integrated and moved to switch mode.

  4. Brad Albing
    March 20, 2013

    David, never thought of it from that perspective – assembling a smart phone from mid-90s technology. And not just from the perspective of PCB square inches, but power draw.

  5. Brad Albing
    March 28, 2013

    Actually (just to pursue that a little bit), you'd end up with something that looke like one of those cell phones (or even wireless residential phones) that was the size of a brick. Ah… what memories.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.