Ever since the iPhone's announcement at MacWorld a few weeks ago, there's been a lot of speculation as to what sort of CPU will power the device. Reuters initially confirmed that Intel would not be providing the processor (as many suspected) and FBR research quickly declared that Samsung was the culprit. A more reliable source, an Italian Intel executive, has now stepped forward and claims that Marvell, owners of the XScale line of ARM processors, was awarded the contract. Sound familiar? That's because an ARM processor was the brains behind Apple's ill-fated entry to the PDA market, the Newton (see today's trivia answer). A post made to the LLVM-dev mailing list by Apple employee Chris Lattner, stating that Apple would contribute code to the ARM branch of the open-source project, seemed to confirm that an ARM chip will indeed power the iPhone.
But is it really so simple? The sleuths at Ars Technica looked a little more closely at Lattner's comments and determined that while the code in question pointed toward running a port of Darwin on an ARM processor (!), it did not seem to suggest the XScale line of chips, due to that platform's lack of support for VFP and ARM v4/v6. At the end of the day, the signs seem to be pointing back in the direction of Samsung and unreleased ARM chips in their pipeline though at this point, it's still anyone's guess.
For more speculation on the iPhone's CPU:
- see this Ars Technica article
