Put a 12-core ARM chip in a MBP and you may not notice the difference. I started out programming on the Acorn Archimedes (ARM4 - 8Mhz), and it wiped the floor w

Put a 12-core ARM chip in a MBP and you may not notice the difference. I started out programming on the Acorn Archimedes (ARM4 - 8Mhz), and it wiped the floor with the PC-class 12MHz 80286 hardware available at the time ('89 ish). Maybe we'll come full circle and multi-core ARM chips will once again outperform the Intel competition. With NEON, floating point performance is coming along, and Apple have invested heavily in getting Grand Central (multi-threading) accepted by app(lication) developers. Cocoa apis are moving heavily towards blocks, and towards being architected with efficient exploitation of multiple cores. There are likely further changes, such as tagging enumerations with 'can-parallelize', that will further increase app performance. I believe there is a WWDC 2010 session on how to make your code run into the future. I will watch that with interest to see if there are any more clues, but in truth, we probably already have them.