October 8th, 2012 ~ by admin

Apple A6 vs Rockchip RK3066: 4 Years vs. 6 weeks of design

The introduction of the iPhone 5 was also the introduction of Apple’s first truly original Application Processor design.  The iPhone 2, 3G and 3GS all featured designs by Samsung.  The iPhone 4 introduced the A4, which was closely based on the Hummingbird Cortex-A8 core developed with Samsung and Intrinsity, again, not a truly Apple design.  The iPhone 4S introduced the A5 (and the A5X used in the iPad 2).  The A5 is based on the ARM Cortex-A9 MPCore, a standard ARM design, albeit with many added features, but architecturally, the processor is not original, just customized.

ARM provides cores designs for use by developers, such as the Cortex-A9, A8, etc.  These are complete designs of processors that you can drop into your system design as a block, add your own functions, such as a graphics system, audio processing, image handling, radio control, etc and you have your processor.  This is the way many processor vendors go about things.  They do not have to spend the time and effort to design a processor core, just pick one that meets their needs (power budget, speed, die area) and add any peripherals   Many of these peripherals are also licensed as Intellectual Property (IP) blocks making building a processor in some ways similar to construction with Legos.  This is not to say that this is easy, or the wrong way to go about things, it is in fact the only way to get a design to market in a matter of weeks, rather then years.  It allows for a wide product portfolio that can meet many customers needs.  The blocks are often offered for a specific process, so not only can you purchase a license to a Cortex-A9 MPCore, you can purchase one that is hardware ready for a TSMC 32nm High-k Metal Gate process, or a 28nm Global Foundries process.  This greatly reduces the amount of work needed to make a design work with a chosen process. This is what ARM calls the Processor Foundry Program.

That is exactly the approach Rockchip (who also makes Application Processors by the million) does.  It is what MediaTek does, and literally dozens of other companies.  A recent Rockchip processor, the RK3066, which is a dual core ARM Cortex-A9 with  quad core Mali-400 graphics.  From concept, to sample silicon was 6 weeks.  That is blinding fast product cycle, but required in their market of low cost Android devices.

Apple used a similar method or the A4 and A5 processors.  A standard IP core, customized and surrounded with Apple specific additions.  They did so because they had to, they didn’t yet have their own ARM design, but alas they were working on it, and had been for some time.

In addition to licensing individual cores you may also purchase an ARM Architecture license.  This gives unlimited access (for 3 years, or a perpetual license) to build anything using the ARM instruction set.  This is the type of license companies like Qualcomm, TI, Cavium, and Freescale have.  They are not restricted to a type of core as they can build the physical and architectural elements anyway they want.  This is expensive, time consuming and requires a very specific skill set that Apple did not originally have.

That changed in 2008 when Apple purchased P.A. Semi for $278 million.  PA Semi designed the first compatible Power Architecture processor outside of the AIM alliance.  It was started in 2003 by Daniel W. Dobberpuhl, who had worked on the DEC Alpha 21064.  PA Semi was excellent at designing high performance, efficient processors. Exactly what Apple would need to design their own custom processor. However the iPhone used an ARM type processor, not PowerPC, but the Alpha is not all Dobberpuhl designed, he also worked on the DEC StrongARM, which later was sold to Intel, and then later to Marvell.  The connection is now clear.  Steve Jobs saw very early on the need to be able to design their own cores, and not have to wait for someone else.  This could give them a lead of months on performance.

By 2010 the architecture was complete, Apple now needed to lay out the architecture in a manner that it could be manufactured most efficiently and with the least power drain.  The performance of a processor has as much to do with its architecture as it does with how that architecture is laid out and implemented in silicon.  This is something Apple had no experience with, and P.A. Semi only had limited experience with.  Apple however had been working with Intrinsity, designing the A5, and hardware layout is exactly what Intrinsity did, and did very well. So, in typical Steve jobs fashion, Apple bought Intrinsity, for an estimated $120 million.  The formula was now complete, Apple had all it needed to complete a truly original ARM design.

The result?  The A6 has performance roughly equal to the ARM Cortex-A15, which is not going to be shipping until December 2012.  4 years of work, and nearly a half billion dollars bought Apple a 3+ month lead in performance.  A lead that means they have one of the most powerful Application processor designs going into the holiday buying season, a season they can easily make $25 billion in profit, an amount that would allow Apple to buy Rockchip, or any other company they wanted. not to mention several small countries.

Sources:
Linley on Mobile
Anandtech – Apple iPhone 5 Performance Preview
ifixit – A6 Teardown
EETimes: China fabless: Rockchip Rocked by Android Tablet Wars

Posted in:
Research

Leave a Reply