FRAMINGHAM (07/24/2000) - Thirty-five years ago, Gordon Moore, the co-founder of Intel Corp., observed that the transistor density of semiconductor chips doubles roughly every 18 months. This observation has proved accurate and became known as Moore's Law. Its net effect has been new PCs that are almost twice as powerful as previous-generation systems. As PCs became more powerful, their capabilities and functionality increased and their appeal broadened the market for potential adopters with each successive generation.
Today, Moore's Law remains in force and, by most accounts, will continue through at least 2017. But does it still benefit users and empower them in the way it has in the past? The answer is no. It now has a diminished impact on users. With Pentium III technology as the current standard and with system speeds of 1 GHz available, entry-level and trailing-edge systems are, for the first time, suitable for mainstream use. "Fast" has finally become "fast enough." For end users doing mainstream productivity work involving word processing, spreadsheets, e-mail and Web browsing, the performance difference in going from a 500-MHz Pentium III to an 850-MHz Pentium III just isn't readily apparent.
I can already hear some of you arguing that this line of reasoning is wrong.
For certain classes of users, you might even be correct. Sure, if you're an engineer working on the human genome project, or a 15-year-old kid trying to play Quake III across the Internet, fast might not be fast enough. If you're creating a massive engineering simulation or trying to create the next Toy Story sequel in real time, fast might not be fast enough. But if you're like me and primarily use a suite of office products for word processing, spreadsheets, presentation graphics and e-mail; run some version of 32-bit Windows (or some other comparable operating system); and occasionally use a few other productivity applications, then fast is definitely fast enough.
While fast will never be fast enough for some, a threshold has been crossed for most users. While relatively cheap PCs have always been available, no one ever wanted to purchase them because they couldn't run the latest generation of software. The current slowdown in the benefits of Moore's Law has made it reasonable to consider cheap PCs. With the advent of low-cost and more-than-capable systems, it's a good time to rethink buying patterns of the past.
In the old days (circa 1992), buying decisions were simple: You bought the fastest processor on the market and added as much memory and disk space as you could afford. That's no longer the best approach. You don't need the fastest processor that Intel or AMD has to offer. Based on user requirements, many of you should consider entry-level Pentium III systems with 128MB of RAM for most of your mainstream users. And it's time to think about longer depreciation cycles. Think about extending that desktop's life cycle from three years to four, and your laptop's from two years to three.
So is this the endgame for PCs? Is the revolution over? Hardly. Nothing lasts forever, and neither will this lull in the benefits of Moore's Law. Systems running into the multi-GHz range will once again broaden the appeal for leading-edge technology and reset the curve. Technologies that will include real speech recognition and cognition that are integrated into the core operating system will one day be possible, and user interfaces will one day advance beyond the graphical user interfaces that we have been saddled with for nearly 20 years. But that will take some time. Real advances will require not just faster processors, but also the software advances that go with them. The current reality is that for the first time since the PC was introduced, users can safely ride price curves down and extend the life of older systems. So enjoy the ride while it lasts.
Michael Gartenberg, former vice president and research area director at Gartner Group Inc., is a partner at Dellet LLC, a venture capital firm in Englewood, N.J. that focuses on the Israeli market. Contact him at firstname.lastname@example.org.