In the 50s, they called it keeping up with the Joneses—purchasing new appliances, cars, or furniture to match or exceed the spending of your neighbors. Today, we usually call it bragging rights if you can claim to have the newest, fastest, or smallest of something. Living in the heart of Silicon Valley puts me at risk for driving over to the local electronics stores and at least window shopping to view the latest electronic appliances and enhancements. These devices promise to make my work easier, entertain me, reduce the number of gadgets that I must carry, or keep me in touch with the world.
Although I've purchased my share of items, I now realize that the initial purchase is just the tip of the proverbial iceberg of never-ending upgrades, or perhaps the first step into debtors' hell. Once that first device becomes yours, the next inclination is to make it as functional as possible, and thus, upgrade fever starts.
For instance, my PCs at home are in a constant state of flux as I add a graphics card here, a larger disk drive there, a new motherboard in another, and on it goes in a never-ending circle. Term it a hobby or an obsession—I'm not sure how to define it. Cell phones, PDAs, MP3 players, and other gadgets continue to tempt me as well. I use a cell phone, but for now, I've drawn the line at that level. I don't yet own a PDA, GPS system, MP3 player, or pager, so a limited number of items can distract me.
Software, however, is another area that requires constant vigilance over the wallet. The many programs that offer everything from games to help with preparing stories are always beckoning us to get the latest version to provide still more functionality or a new round of game play.
The same temptations to have the latest and the greatest pull at us at work, too. We all would like to have the fastest machine, the latest software, and so forth. On the engineering side, we typically experience the same temptations when it comes to our tools. We want the fastest workstation, the latest design software, and the most precise instruments. But do we really need the additional features or performance?
In many cases, the answer is a resounding yes! The new features will provide improved throughput and the ability to handle more-complex designs. This is especially true in the ASIC design arena, where tools that handle more-complex designs or rip through the analysis faster give the designer a strategic advantage.
In many more situations, though, the upgrades tend to be more cosmetic, or offer minimal improvements in performance. Such "upgrades" are much harder to justify. But where does one draw the line? What criteria should we use to determine when to upgrade? Should the new product offer at least a 50% improvement, a 100% boost, or more to justify an upgrade? I'd like to hear your thoughts.