>How will TI's CDMA efforts dovetail with what NOK is doing?
>Will TI's efforts be confined to wCDMA and NOK's to CDMA? >Will TI try to produce a 1X chip and if so, how will this relate to NOK's chip efforts?
>In what way (if any) are the answers to these questions impactful to QCOM?
I wish had the answers to all of the above, especially since much of my portfolio is tied up in QCOM/TXN. This is my understanding....hopefully some of the engineers on the thread can comment if I have anything wrong. Much of my interpration comes from this post by W. Molloy....
Message 12093447
The GSM protocol stack won't be running on the DSP.
DSP's are optimised for fast algorithmic processing, not message and event passing. NOK will have a CP-DSP solution. They may, for example
source the CP core from ARMHY and the DSP core from TXN.
The two cores will be fabricated in a single ASIC.
QCOM MSM chips use an ARMHY CP with an in-house DSP.
I think that it is likely that Nokia's current CDMA phones use a similair configuration. The entire ASIC is probably fabricated by TXN, but since the IP resides with Nokia, TXN hasnt needed a CDMA license up until now. It is tough to tell how 1xrtt will be developed but I believe that it probably will continue along these lines. This would mean that TXN wouldnt be able to sell this solution to any other company except Nokia.
OTOH...W-CDMA development has taken a different path. TXN has developed the OMAP (Open Multimedia Application Platform) architecture to allow it to market a W-CDMA solution to all of its customers. I believe that TXN needed the W-CDMA license since it is developing the software for the baseband chip. Note the following passage from a white paper on TXN's website. The paper mostly talked about the applications which OMAP would allow.
www-s.ti.com;
The OMAP architecture provides a means to coordinate dual processors across the two basic components of the wireless appliance and to seamlessly take advantage of the unique capabilities of each. Nokia, Ericsson, Sony, Handspring, and others already have selected the OMAP hardware and software architecture as the development platform for their wireless appliances. The first samples of an OMAP-based chipset will be available in 4Q00.
I sort of got confused on weather OMAP would be doing the baseband processing since that paper mostly talked about applications. However, this passage from an article seems to state that OMAP will do both....
techweb.com
TI's OMAP will feature a TMS320C55x DSP and ARM9 hybrid in the baseband chip, and an identical 'C55x/ARM9 engine to handle computing applications. The company plans to eventually integrate all four computing elements on a single chip, and is also working on an ARM10 implementation.
It's not clear if the OMAP chipset which is sampling will include the baseband chip. This would be the portion of the architecture which needed Q's license. The timing seems like it might support the fact that TXN has the baseband chip ready to go.
In summary....I think that it is likely that TXN will not challenge Q's supremacy in the IS-95 arena. I have not heard of any plans to market a 1x chipset....but they are probably working with Nokia for a custom 1x chipset for Nokia only. With the OMAP chipset, I think that TXN will probably be SpinCo's biggest competitor in W-CDMA.
However the biggest thing that struck me, while looking through the TXN website, is the amount of extra computing power that all of these applications/bandwidth are going to need. These chipsets are going to be a lot more expensive than the current generation. I think that margins for both companies will be increasing over the next couple of years.
Slacker |
