The iPhone 5 Technology Rabbit Hole

| About: Micron Technology (MU)

What could possibly have caused Apple's (NASDAQ:AAPL) iPhone 5 to fall fifteen months behind schedule? A slip-up in a recent Apple patent application might provide the answer - and the investment opportunity of a lifetime.

The release of the iPhone 5 was initially expected on Apple's normal release schedule in June of last year. This is ideal timing for optimizing back-to-school sales. After all, students are Apple's largest customer base. But when the kids found themselves back in school without any new Apple gear, some speculated that the company was busy working on some groundbreaking new technology to further set them apart from the competition. And after what seemed like an eternity to the millions of frothing Apple enthusiasts, a new iPhone was finally announced in October. But it wasn't the expected fifth generation model.

Sure, the 4S had an upgraded processor, a retina display and apologists attempted to justify the otherwise unexplained delay on Siri, which was, admittedly, a very impressive feature. When the iPhone 3GS was introduced, everyone learned that Apple is able to command such premium prices because their products are more than just technology - they are fashion. Fewer people are going upgrade if it isn't outwardly apparent that they didn't spend a tidy sum in order to do so. Had Steve and Tim made the same mistake twice while burning up an extra three months in the process?

Due to information disclosed within a seemingly benign patent application and some very under-the-radar deals, it is my opinion that the iPhone 4S was a rushed, stop-gap cover-up for a botched iPhone 5 launch. This realization provokes but one big question: what technology is Apple bringing to market?

And down the hole we go...

Realize that fifteen months behind schedule doesn't just happen for something trivial. Especially in the context of the most valuable corporation on the planet. They have a stable of cash that only Dr. Evil can comprehend. They should be able to buy anything, right? I call US Patent Application 20120185797 to the stand. While the assignee of this patent is not listed, it should be obvious from the content that it is Apple (e.g. - the inventors work for Apple, there is an Apple copyright, several mentions of iPhone, iPad, Mac, etc).

This is a pretty boring patent app with one exception: the mention of "phase change memory" or PCM, a next-generation memory that has the potential to displace flash and DRAM memory types with dramatic power-efficiency. Now, it is important to note that PCM is listed in many of Apple's other patents. But these mentions always been in conjunction with the other leading next-gen memory types (which is a typical patent strategy). In this particular application, it is listed exclusively alongside the traditional memory technology incumbents - RAM, ROM and flash.

If the iPhone 5 uses PCM, then it will have breathtaking processing power with incredible energy efficiency - handheld panacea.

If you are still with me, connecting the dots is not too difficult. Most notably, Apple relies on Samsung in order to fabricate the custom microprocessors that power the iPhone and iPad. It should be no surprise that Samsung is the leader in PCM research, having produced a functional 8 gigabit PCM array on cutting-edge 20 nanometer technology. While Samsung has presented this as simple research, it is my opinion that nobody would invest such a substantial amount into just research. Tying up a brand new 20 nanometer fab cost them a fortune in both direct and opportunity costs. This was to be a commercial part - one that, potentially, was to sit in the core of the iPhone 5.

But something went awry with this part and, thus, the iPhone 5 was delayed. Samsung discussed this 8Gb PCM array at the International Solid-State Circuits Conference last February and the focus was on the need to redesign the chip in order to increase bandwidth. I submit that the engineers had proofed the chip using a larger fab process (say, on an old 45 nanometer fab) and anticipated certain performance after scaling it down to 20 nanometers. When these expectations were not met, they had to go back to the drawing board. And Apple had to scramble to churn out a stop-gap 4S.

What exactly does PCM technology facilitate? Before we can understand the answer to this question, it is important to understand the current technology. Briefly, when your smartphone notifies you of some activity (e.g. - an email, Facebook activity, text message, etc), understand that this notification is provided by way of a running application. When you power-on your phone, applications are copied from flash memory, which lacks the durability and speed to execute applications, to system memory which is provided by DRAM, a memory that is durable and fast but loses its state when power is removed ("volatility").

Realize that, even if you can't see them, these applications are always running in the background in order to provide such notifications or other functionality. Since they need to be in DRAM in order to be running, there is a requisite power consumption associated with any active functionality. This is why battery life decreases as more active applications are installed - a problem that has become the bane of the smartphone industry. Uphill. Both ways. In the snow.

PCM, however, has the non-volatile characteristics of flash memory and the effective speed of DRAM memory. So, once an application is loaded into PCM, it never needs to be loaded again - even after power has been removed from the underlying memory. The application will always be in the state that it was in when it was last executed. So instead of supplying continuous battery power to the Facebook application, for example, PCM allows the phone to simply power-up the underlying memory once or twice a minute so that it can check for new activity. PCM-enabled phones will sip battery instead of guzzle. This is applicable for the operating system as well. After PCM, "booting up" will be a story that we tell to the grandchildren.

The technical types will be quick to point out that PCM bandwidth is not high enough to serve as main memory in a high-performance smartphone application. This issue has been addressed in the form of "hybrid memory", which leverages a small DRAM cache to mitigate the performance penalties of PCM (Side note: what is "hybrid" in the "hybrid memory cube"?).

With eight bits in a byte, the aforementioned 8 gigabit Samsung PCM array works out to a gigabyte which, not surprisingly, coincides with the rumored specification of the iPhone 5. Again, this specification has also puzzled industry analysts because 2GB RAM is the current high-end standard. Just realize that 1GB was the standard 15 months ago.

The hypothesis of a PCM-enabled iPhone is further reinforced by the leaked battery specification, which is fully expected to be entirely insufficient in the face of the power-hungry LTE network requirements.

"Either way, Apple will likely have something up its sleeve if these leaked pictures really are the final battery, as it's tough to imagine the company will launch its first-ever LTE iPhone with poor battery life." -- Brad Reed,

Furthermore, to the befuddlement of industry analysts, Toshiba (one of Apple's largest suppliers) recently announced a dramatic cut in flash memory production capacity. Still, the PCM-enabled iPhone idea is certainly a stretch and I will be the first to admit that. But events like this can produce the investment of a lifetime. If the iPhone 5 does show up with PCM, how can one capitalize?

The base PCM technology is owned by a company called Ovonyx, which was established by Intel (NASDAQ:INTC), some ex-Micron employees and Energy Conversion Devices (ECD). ECD was put into bankruptcy this past February and the stock has gone dark as of last week. It is my opinion that this bankruptcy was orchestrated with the sole purpose of quietly extracting ECD's 38.6 percent interest in Ovonyx. This transaction was recently completed - Micron purchased the interest for just $12 million.

Ethical (and speculative) rants aside, Micron (NASDAQ:MU) is now positioned to capitalize on a hypothetical PCM-enabled iPhone. Even if this speculation proves to be false, the PCM technology is now real and expected to be a foregone conclusion in displacing a broad part of the $70+ billion DRAM and flash markets. IBM, for example, expects PCM to play a role in dramatically reducing the power consumption of data centers.

Intel owns 45 percent of Ovonyx with the remaining 16.4 percent being held by private interests. Because Intel's market cap is much higher, the impact on share price will be less dramatic. I do, however, see Intel acquiring Micron, which owns Intel's old flash memory business Numonyx (not to be confused with Ovonyx) at this point since PCM is the solution to Intel's inability to penetrate the mobile markets. A relationship is already in place - with PCM fruit.

With PCM, your phone can have an Intel processor just like your laptop or desktop. And so it can run the same operating system. Your PC will always be with you - in your pocket or purse. When you go home, a wireless monitor, keyboard, mouse and speakers will allow you to use it as a regular PC (or Mac, which also leverage Intel processors).

This is a fundamental shift in the way that we use computing. Disruptive.

Disclosure: I am long MU. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Disclaimer: This article is entirely speculation that may or may not turn out to be true. I have lost a substantial amount of money chasing the idea (by way of investment in the now extinguished Energy Conversion Devices) that phase change memory would turn up in the iPhone. I am long in Micron with hopes that it is true.

About this article:

Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500. Become a contributor »
Tagged: , , , Semiconductor - Memory Chips
Problem with this article? Please tell us. Disagree with this article? .