Various teardowns and magnified images have revealed something rather uncommon in the System-on-a-Chip field: Apple’s A6 SoC has a hand-designed dual-core CPU.
This is unique in that the typical approach is to feed the parameters in to a very sophisticated computer program, which then designs the transistors to meet the required specifications (the results of this process can be seen with the tri-core GPU on the A6). These designs look symmetrical, compact, and overall very robotic. The human-designed CPU, while still symmetrical to a certain extent, also has some advantages, according to iFixIt:
When compared to the rigid, efficient layout of the GPU cores directly below it, the layout of the ARM cores looks a little homespun—at first.
- Generally, logic blocks are automagically laid out with the use of advanced computer software. However, it looks like the ARM core blocks were laid out manually—as in, by hand.
- A manual layout will usually result in faster processing speeds, but it is much more expensive and time consuming.
- The manual layout of the ARM processors lends much credence to the rumor that Apple designed a custom processor of the same caliber as the all-new Cortex-A15, and it just might be the only manual layout in a chip to hit the market in several years.
So not only does this account for the noticeably faster speeds (though the higher-clocking speeds also help), it’s possible that this custom design also keeps the system from overheating as easily as previous devices and chips have. These changes also make sense, given the significantly higher benchmark scores that the iPhone 5 is able to achieve.
This A6 does seem to be the first legitimate product of the various talent acquisitions Apple has made over the years in the CPU/GPU field, and also highlights one of their key advantages over nearly every other competitor: they have the talent and cash to produce a chip such as this, even if it is more expensive to do so and flew in the face of the common practice, at a massive scale.