I agree wholeheartedly it will be difficult for someone to knock QCOM off its CDMA pedestal. I also think control over CDMA is a pretty strong leverage point to work from when trying to establish a broader architecture. Here's what both interests and worries me:
In an earlier post on architecture, I described the desireability of investing in the technical equivalent of the heart, mind, or spine of an architecture and in another post I suggested CDMA may be the spine. What we need to worry about is who controls the heart and mind. (I know it's an imperfect analogy but bear with me.)
When you look at the functional diagram Eric forwarded to me from qualcomm.com. you can see QCOM is trying to control the whole cell phone architecture but at the same time you can see how someone might approach controlling the architecture from building on the ARM core or their own core, their own DSP, or perhaps even from their own MMC which could conceiveably control access to software and advanced media.
In these scenarios, any 3G competitor to QCOM would still pay QCOM for their CDMA IPR, or even better, pay QCOM for their CDMA IPR and ASIC. But for QCOM to be the biggest, baddest Gorilla in the wireless jungle it needs to control the architecture and not just be limited to the CDMA ASIC. Control the IPR and you get a 5% royalty, control the ASIC and you get substantially more, control the architecture and you get it all. The handset manufacturers literally become your resellers and they make money on the plastic case and the styrofoam peanuts they put in the shipping box.
My interest in TXN stems from the fact that DSP is a good point (the heart?) from which to control an architecture. Everybody else in the wireless world who is looking to contain QCOM in their CDMA corner will probably turn to TXN for help. Combine that with a potential play in DSL, cable modems, voice over IP, MP3, and digital cameras and I would say we have the makings of an interesting DSP Gorilla Game if TXN can find a tornado and hang on. |
