Moving to multicore processing is necessary to get the kind of performance required by new games. But before that happens, game developers must learn how to take advantage of these multiple cores.
The Xbox 360 has the simpler of the two architectures, since splitting an application among cores is relatively straightforward. (That's because the same code can run on any core.) Likewise, a straight port could use just one core or SMP-style (symmetric multiprocessing) multitasking.
The Cell processor in the PlayStation 3 is a more interesting case. The synergistic processing elements (SPEs) handle the bulk of the processing, with coordination between them and the single PowerPC core. Applications then will need to be split accordingly, which means that tasks running on an SPE will have to account for the per-SPE memory.
Partitioned systems like the Cell processor offer performance advantages, but the programming job can get complicated when tasks start bumping into the hard limits of hardware. As a result, games often won't use all of the SPEs. Overall, it shouldn't prove to be a big deal, though, as long as the ones that are used deliver a suitable gameplay environment.
