One of the most nagging questions asked by young children on a long trip is, “Are we there yet?” Parents usually plan trips with overnight accommodations, sights to see, and timed end goals in mind, but children have a much shorter perspective on time.
In packaging, there’s an eagerness to know where we’re going, followed by a “how long will it take” that is almost child-like in its impatience. Take, for instance, memory chips. The industry has searched for universal memory that can retain data without a continual power supply – one that’s cheap, fast, and sturdy enough to replace flash, DRAM, SRAM, and others. Without it, system designers use types of multi-chip packages. Samsung, for example, offers four types of flash and DRAM in an 8-chip package for cell phone storage.
When a memory company can merge the best of many memory types, space and time savings will be the reward. At present, estimates of more than 70 companies are looking into new memory devices – an area of high growth for the packaging industry.
It might be that a one-stop approach to memory will not fit what the industry requires. If the problem requires speed, then one type will fit. If it requires a non-volatile memory, another will fit. If cost control rules, then the best that fits wins. SRAM requires low energy. Then there’s the NOR vs. NAND argument (see the archived webcast on this at www.apmag.com). All in all, according to many analysts, the memory revenue forecast shows healthy growth for the next few years.
Taking the need for increased density one step further, Georgia Tech’s Packaging Research Center (PRC) in Atlanta, GA, is once again planning an open house to present its microelectronics vision that it calls “System-on-a-Package” (SOP) on Wednesday, March 13th. PRC pioneered the SOP concept, which it sees as an emerging and fundamental microelectronics and microsystems paradigm for the portable and desktop convergence in the near future. The SOP is based on the electronics industry’s need to decrease size, increase portability, and converge functions. For instance, it combines the best of on-chip integration with the best of packaging integration. Imagine microelectronics, photonics, RF, and MEMS in a single component system. In the process, these researchers aim to bring packaging far beyond MCM, flip chip, CSP, and SIP approaches.
We attended the open house last year, and will again this year just to see the progress that PRC’s research has made to integrate functions such as high-speed digital, high-bandwidth optical, analog, RF, and MEMS in a microminiaturized board/SOP.
Are we there yet? Not really. The journey is long, but the trip is fascinating.
 |
Gail Flower
Editor-in-Chief