Article
SAN FRANCISCO — At an ISSCC session yesterday afternoon, IBM announced details of a smaller, lower-power version of the Cell BE processor that powers Sony's PlayStation 3. The Cell BE is currently fabricated on IBM's 65nm SOI process, but IBM will soon move the console chip onto the company's much-ballyhooed, next-generation 45nm high-k process.
The 45nm Cell will use about 40 percent less power than its 65nm predecessor, and its die area will be reduced by 34 percent. The greatly reduced power budget will cut down on the amount of active cooling required by the console, which in turn will make it cheaper to produce and more reliable (this means fewer warrantied returns). Also affecting Sony's per-unit cost is the reduction in overall die size. A smaller die means a smaller, cheaper package; it also means that yields will be better and that each chip will cost less overall.
All of these chip- and unit-level savings may or may not get passed on to gamers in the form of price cuts any time soon. It all depends on whether Sony wants to boost its margins and show a profit in its gaming unit, or attract new gamers to the console by lowering the price. Eventually, the cost savings will get passed on to users; it's just a question of when.
Speaking of Cell and sales, the presentation suggests that, despite IBM's promise that Cell could see widespread adoption outside of the console realm, Sony is still far and away IBM's main customer of Cell. Specifically, IBM states the following in the paper digest: "To guarantee the proper operation of existing gaming software, the exact cycle-by-cycle machine behavior, including operating frequency, must be preserved."
In other words, IBM's Cell shrink was made with Sony in mind; the chipmaker didn't take advantage of the shrink to make any performance-enhancing tweaks, opting instead to preserve the exact performance characteristics of the 65nm version, which itself preserved the performance characteristics of the 90nm version.
Now, I'll admit that a member of Intel's Itanium team is the person who highlighted this part of the paper for me, but I still think he has a point. This process shrink is all about making cheaper PlayStation 3's, with IBM's narrow but profitable Cell-based blade business taking a back seat to the needs of the volume console market. So IBM may have suckered Sony into buying a supercomputing coprocessor disguised as a gaming chip, but it looks like Sony could get the last laugh.
