New details have emerged which suggest chip maker Palo Alto Semiconductor (PA Semi) might not be the hive mind behind the iPad’s “A4” processor as was widely expected. In case you missed it, the A4 is the diminutive custom silicon that lies at the heart of Apple’s new iPad. It’s the wee beasty that has the raw power to make Magic Move work so smoothly in the upcoming Keynote app, while providing the intelligence to manage energy efficiently enough to squeeze 10 hours of actual use out of the iPad between charges. Oh, and it runs at 1Ghz and is fuelled by unicorn tears, or something. Anyone with an iPhone (and everyone who has ever relied on laptops to do a days work) knows that there’s usually a big difference between a mobile device’s advertised and actual battery life. So, unless Steve Jobs is lying through his teeth, how exactly does the iPad’s A4 processor manage to deliver its number-crunching goods over such a long period of time? Certainly Apple’s developments in battery design help a lot, but it’s thought that the real magic happens in the custom-designed processor itself. Venturebeat.com’s Paul Boutin has been investigating the A4, and pushing chip engineers for answers. In an article published on the weekend, he offers the following (possible) explanation ; Apple has invested heavily in OpenCL and LLVM, which are technologies to distribute work across multiple CPUs and multiple GPUs. In this Apple is different than other mobile devices: other vendors want video decoding and 3D games at a good rate, but often leave the GPU mostly idle. Apple is looking to drive a lot of work through the GPU all the time, as part of any application. For Apple, it makes sense to put a lot of GPU cores in the chip
Read the original post:
The Tangled Web: PA Semi, Processors, and Magic
