Thinking About Moore's Law
by
Randy Mears
In the business of Information Technology, there is a perception that Moore’s Law has been our salvation for a long time. After the recent turn of the century, there seemed to be a sentiment among some fellow IT aficionados that Moore’s Law would soon hit the proverbial wall and Information Technology advancement would freeze in its tracks. For some, though, the prevailing opinion was that there would be enough innovation to offset stagnation for a very long time.
I have often thought that focusing on Moore’s Law may be too single minded. When it comes to barriers in IT, it may turn out that the issues won’t be with hardware advancement but rather with fixing software technologies. Wirth’s Law may say it best, “Software gets slower faster than hardware gets faster.” In addition to drawing attention to how important software improvement is for IT, Wirth’s law also reinforces the idea that we still, more than ever, need to rapidly advance the hardware.
So, let’s get back to Moore’s Law. These days we tend to think in broader terms than IT, so the term digital technology may more appropriately designate the full range of impact for Moore’s Law. Over the past few years, a number of advancements have appeared that will ensure significant increases in transistor density for many years to come. Here is one I just encountered. I thought it was noteworthy because it merges traditional silicon chips with materials that are a product of nanotechnology (I love it when technologies come together). But there are more traditional approaches that will also buy us time. Here is one from a couple of years ago that addresses increasing density with new silicon laser technology.
Moore’s Law is fairly specific so there may come a time when it isn’t terribly important; a time when some new foundational technology, behaving by a different set of laws, may rule the day. After all, the density of transistors my not always be “It.”
