Revisiting My 3G iPhone Predictions

|
Includes: AAPL, IDCC
by: Vijay Nagarajan

TechOnline published a very detailed teardown analysis of the 3G iPhone that can be accessed here. Their analysis went a step beyond iFixit's disassembly as they scrutinized 'under the hood' of the various iPhone components to identify them more accurately. Their extraordinary effort also lets me review my iPhone predictions that were summarized in my June 2008 post here.

Here is the photo of the iPhone 3G chipset as dissected by TechOnline:

Prediction #1: As this photo indicates, Infineon (IFX) emerged the cellular chipset winner. The analysts are not sure if it is the PMB 8878 since they had no way of comparing the iPhone component with a known sample of the Infineon 3G-baseband offering. They found that the baseband was a two-chip, single package solution. They further identified it as the 2G chip and a 3G accelerator.

For the moment, I have no reason to suspect that this is anything but the PMB 8878. I don't believe that Infineon would have a chip customized for Apple. On the other hand, the company perhaps quickly cobbled together a 3G solution in the form of a two-chip PMB8878 to cater to the market needs (including those of Apple.)

Given the long cellular product cycle and the tight iPhone schedules, this would have been the path of least risk as compared to spinning single chip solutions. For the record, Infineon recently announced smaller and more sophisticated 3G solutions with its own software stack. These latest chips (that are not in the iPhone) are likely the more thought out, and more optimal single chip UMTS solution.

The two-chip solution inside the iPhone 3G further brings back the question I have been asking for a while now. Does the 3G IP in Infineon partially or wholly belong to InterDigital (NASDAQ:IDCC)? It could well be that the iPhone has many if not all elements of the SlimChip IP from InterDigital. Apple's license with InterDigital last year will then have deeper implications than the normal 3G license that the King of Prussia-based company seeks from handset vendors.

While the exact details will emerge in the coming days, my prediction #1, "Infineon will be at the heart of the iPhone," has turned out to be true.

Prediction #2: "Samsung will perhaps continue to own this part.” There is not much of a surprise here as the iFixit teardown revealed last Thursday. Prediction #2 was also true.

Prediction #3: Before we discuss WLAN and Bluetooth, here is another picture from TechOnline:

 

Therefore, as it turns out, Apple yet again went with Marvell (NASDAQ:MRVL) for WLAN. Prediction #3, "For platform stability issues, I will bet on Marvell grabbing this socket again." was right.

Prediction #4: "If WLAN belongs to Marvell, CSR, which is in the current iPhone, will likely own the Bluetooth socket again." A look at the figure above will make it four on four so far.

Prediction #5: The 3G iPhone carries the Hammerhead II GPS solution co-developed by Infineon and Global Locate, the company that was bought by Broadcom last year. Therefore, that makes my prediction #5 - "The next generation will have GPS and it will likely belong to Broadcom (BRCM)"  right too. I don't know the specifics of the licensing agreement between Global locate and Infineon. So, I will not be able to comment on whether BRCM will benefit from this component, if at all.

Prediction #6: The touch screen controller belongs to Broadcom as well.

In summary, all six iPhone predictions I have recorded in this blog have come out to be true. As you can see, I based my predictions about most of these components on the rationale that Apple will not want to hamper the stability of the 3G iPhone by testing out new components in a short time span. As it turns out, most of Apple's component decisions were based on this very logic, and the TechOnline article aptly points this out.

For a more thorough look into the component dissection, please visit TechOnline's site.

Disclosure: The author was long IDCC at the time of writing.

[Disclaimer: All thoughts expressed by Vijay Nagarajan in his articles are his and do not necessarily reflect those of either Atheros Communications or TensorComm Inc.]