SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Qualcomm Moderated Thread - please read rules before posting -- Ignore unavailable to you. Want to Upgrade?


To: 100cfm who wrote (42217)9/13/2004 11:18:37 AM
From: slacker711  Read Replies (1) | Respond to of 197011
 
So even if NOK/EMP/NEC gets a chipset out for sampling by end of this year. You still have the typical 9 months from then before commercial production of the chipset. That puts HSPDA out to 4th Qtr of 05. Isn't that a more realistic timeframe for Cingular.

I dont think that anybody has an HSDPA chipset sampling yet, but it is impossible to tell for sure with companies like Ericsson/Nokia. I believe that the only time that Ericsson has provided a chipset roadmap was 18 months ago....and I have never seen it updated.

3gamericas.org

They dont even mention an HSDPA capability for any of the chipsets on that roadmap.

I see a real opportunity for Qualcomm with HSDPA. The current list of customers for their W-CDMA chipsets is long but they need more tier-one manufacturers. I would think that handset manufacturers like Siemens and Panasonic must be realizing that their chance to gain market share with the move to W-CDMA is already slipping. HSDPA will give them another chance....if they can get handsets to the market ahead of the competition.

Slacker



To: 100cfm who wrote (42217)9/13/2004 12:39:17 PM
From: Eric L  Read Replies (1) | Respond to of 197011
 
The HSDPA Reality ...

<< So even if NOK/EMP/NEC gets a chipset out for sampling by end of this year. >>

Nokia doesn't sample chipsets. They are not a merchant chipmaker, and before they deliver HSDPA handsets they will deliver base station cards with HSDPA modem chips. They probably have prototype handsets in their own networks division labs now, and in Ericsson's and Siemens' IOT labs now.

EMP who is a merchant chip maker, will probably sample 1st to Sony Ericsson and perhaps Sharp, but again, Ericsson networks will deliver Base station cards with HSDPA modem chips, and they probably. They probably have prototype handsets in their own networks division labs now, and in Nokia's, and Siemen's,IOT labs now.

Remember this, Ericsson and Nokia are delivering about 50% of the network gear in the world. It puts them in an enviable position competitively.

Chipsets from Motorola/Freescale who is a merchant chip maker, will probably 1st appear in Motorola handsets.

NEC/Agere, hard to say, but I do expect NECEL to be early on in trialing in Japan.

Most assuredly TI has prototype HSDPA base station modems populating labs right now.

Qualcomm may do the "trial handset" trick again, but they are dependent on their handset (and more likely modem card) manufacturers to incorporate their chipsets in UE. Their value chain is currently less involved in HSDPA maturation than Ericsson, Nokia, Motorola, NEC, Matsushita, Fujitsu, and Mitsubishi, so the priorities of that value chain will dictate their development based around Qualcomm chipsets.

<< You still have the typical 9 months from then before commercial production of the chipset. >>

Probably more. Nine months has not been 'typical' for Qualcomm WCDMA chipsets. You can backtrack all 3 generations of Qualcomm chipsets, on this, and factor in the fact that moving from an MSM6200 to an MSM6250 is a shorter evolutionary step then moving from an MSM6250 to a commercial grade MSM6275.

In order to take advantage of the data transmission capabilities of HSDPA, on the short haul an integrated ARM9 and DSP powered baseband/applications processor isn't going to hack it so look for an ARM7/ARM11 C/A processor combo in the early going and bear in mind that Qualcomm won't be sampling an ARM11 multi-engine chipset till Q1 2005 at earliest, and wont be sampling a 3GSM UMTS WCDMA/HSDPA chipset till Q4 2005 earliest.

There is a big edge here to TI, and to their value chain which has been developing the software platform that resides between the hardware level and the keyboard/display around the rapidly evolving OMAP platform. This particularly applies to the 3 leading Japanese vendors who also have mastered FOMA's specific software requirements.

<< That puts HSPDA out to 4th Qtr of 05. Isn't that a more realistic timeframe for Cingular. >>

Realistically, HSDPA is further out than that for anybody, commercial handset wise, and probably even further out for Cingular, although they could possibly be doing commercial trialing by that time, but more realistically will be trialing in lab. What Cingular needs now, is delivery of 3GPP R5 compliant network hardware (base stations, cards, IMS subsystem) that at least is software upgradable to HSDPA, and their RFP probably focuses on that.

Best,

- Eric -



To: 100cfm who wrote (42217)9/13/2004 1:19:46 PM
From: Eric L  Respond to of 197011
 
Lynette Loony on HSDPA ...

Quotable Quotes which beats her tradititional "sorces say":

"It’s doable from an infrastructure standpoint [HSDPA meeting a mid-2005 goal], but we always seem to forget about handsets. To have perfect timing between infrastructure and handsets, where it has never been accomplished before, would set a new world record." Lars Nilsson, Ericsson -

"HSDPA is under a microscope. ... We believe this technology will face many operation and integration challenges like any other new technology." Brian Modoff, Deutsche Bank USA -

"While operators want HSDPA today, HSDPA is not technically feasible and probably won’t be feasible outside of Japan until 2006 or 2007. [Thelander’s research has led him to conclude that several of the key original equipment manufacturers will not have an internally developed HSDPA solution until mid-2006. Mobile chipsets may also be lacking. Qualcomm is the furthest along in its development stage, with plans to introduce its first HSDPA chipset by the end of the year, but it will only support up to 1.8 Mbps]. Qualcomm will need to deliver a more advanced HSDPA-enabled chip to attract the interest of an operator." - Michael Thelander, Signals Research Group LLC -

"The purpose of the RFP is to get side-by-side comparisons of timelines of various issues, one of which is HSDPA and its timeline for the 850 band. Our commitment to EDGE has not changed. ... We still see EDGE as being a high-speed technology that will meet most customer demands, and we are requiring EDGE to be the backup solution for a WCDMA/HSDPA environment where quality is not acceptable. We’re continuing to look at the business case in terms of how deeply to deploy [WCDMA/HSDPA]." - Kris Rinne, Cingular acting CTO -

Lynette Loony on HSDPA ...

Quotable quotes rather than her normal "sources say." ...

"It’s doable from an infrastructure standpoint [HSDPA meeting a mid-2005 goal], but we always seem to forget about handsets. To have perfect timing between infrastructure and handsets, where it has never been accomplished before, would set a new world record." Lars Nilsson, Ericsson -

"HSDPA is under a microscope. ... We believe this technology will face many operation and integration challenges like any other new technology." Brian Modoff, Deutsche Bank USA -

"While operators want HSDPA today, HSDPA is not technically feasible and probably won’t be feasible outside of Japan until 2006 or 2007. [Thelander’s research has led him to conclude that several of the key original equipment manufacturers will not have an internally developed HSDPA solution until mid-2006. Mobile chipsets may also be lacking. Qualcomm is the furthest along in its development stage, with plans to introduce its first HSDPA chipset by the end of the year, but it will only support up to 1.8 Mbps]. Qualcomm will need to deliver a more advanced HSDPA-enabled chip to attract the interest of an operator." - Michael Thelander, Signals Research Group LLC -

"The purpose of the RFP is to get side-by-side comparisons of timelines of various issues, one of which is HSDPA and its timeline for the 850 band. Our commitment to EDGE has not changed. ... We still see EDGE as being a high-speed technology that will meet most customer demands, and we are requiring EDGE to be the backup solution for a WCDMA/HSDPA environment where quality is not acceptable. We’re continuing to look at the business case in terms of how deeply to deploy [WCDMA/HSDPA]." - Kris Rinne, Cingular acting CTO -

Lynette's July article in its entirety ...

>> GSM Operators Need HSDPA to Grow Up Fast

Lynnette Luna
Broadband Edge
July 9, 2004

bbedge.mblast.com

The wireless high-speed data revolution sparked by CDMA 2000 1x EVDO technology and other emerging wireless broadband technologies has created a sense of urgency among global competing GSM operators who are in the midst of deploying third-generation WCDMA technology, say infrastructure vendors and industry analysts. Now the race is on to meet this competitive threat with a better version of WCDMA—and there’s not much time.

WCDMA barely left the starting gate when players such as Japan’s KDDI, SK Telecom and KT Freetel in Korea and Nextel Communications, Sprint PCS and Verizon Wireless in the United States emerged with service offerings that offer substantially faster data rates than today’s version of WCDMA, known as Release 99, which offers typical downlink data rates of 100 kilobits per second to EVDO’s 300 to 500 Kbps.

That has left WCDMA operators to pin their hopes on HSDPA (High Speed Downlink Packet Access), the next generation of WCDMA that promises downlink rates anywhere between 2 Megabits per second and 14.4 Mbps, and its market debut in mid-2005.

“There is a need to get HSDPA out in the market very quickly,” said Roger Derrien, vice president for UMTS product management with Lucent Technologies. “There is clearly a strong competitive pressure ... Release 99 is not able to compete in markets where you have EVDO.”

Nor can it compete with other emerging broadband technologies such as Flash-OFDM, which offers downlink speeds of 1.5 Mbps sustained, and TD-CDMA that offers speeds between 4.58 Mbps to 5.16 Mbps. Nextel Communications recently expanded its Flash-OFDM service in North Carolina to paying customers. TD-CDMA has gained momentum with entrepreneurial companies worldwide that are successfully competing for a share of the broadband market.

As a result, “HSDPA is under a microscope,” said Deutsche Bank analyst Brian Modoff in a recent report. The technology, standardized as Release 5, is a software upgrade to WCDMA base stations that not only promises theoretical peak downlink speeds of 14.4 Mbps, but increased capacity, a ten-fold improvement in spectral efficiency and support for Voice over IP (VoIP), leaving WCDMA operators clamoring for deployments. But analysts say the technology is hard-pressed to be commercially ready by mid-2005.

“We believe this technology will face many operation and integration challenges like any other new technology,” Modoff said.

Japan’s largest operator, NTT DoCoMo, pressured by competition from EVDO competitor KDDI, is investing $345 million to expedite commercial availability of HSDPA, and may be the only operator to deploy the technology by 2005. The carrier is already trialing HSDPA in its network.

“While operators want HSDPA today, HSDPA is not technically feasible and probably won’t be feasible outside of Japan until 2006 or 2007,” said Michael Thelander, chief executive officer with Signals Research Group LLC, a wireless telecommunications research and consulting company.

Thelander’s research has led him to conclude that several of the key original equipment manufacturers will not have an internally developed HSDPA solution until mid-2006. Mobile chipsets may also be lacking. Qualcomm is the furthest along in its development stage, with plans to introduce its first HSDPA chipset by the end of the year, but it will only support up to 1.8 Mbps, noted Thelander.

“Qualcomm will need to deliver a more advanced HSDPA-enabled chip to attract the interest of an operator,” said Thelander.

Infrastructure vendors claim HSDPA will meet the mid-2005 goal, but the gating factor will be handset availability. “It’s doable from an infrastructure standpoint, but we always seem to forget about handsets,” said Lars Nilsson, director of business strategy with Ericsson. “To have perfect timing between infrastructure and handsets, where it has never been accomplished before, would set a new world record.”

As a result, early deployments will support PC cards, but to achieve a broader uptake of services, an ample supply of devices is necessary. Handset availability has already crippled rollouts of WCDMA Release 99.

Ultimately, Cingular Wireless, set to become the largest wireless operator in the U.S. after its merger with AT&T Wireless is completed, could play a key role in how fast HSDPA comes to market. Sparked by competitive pressure stemming from announced EVDO rollouts from Verizon Wireless and Sprint PCS, Cingular announced it has issued a request for proposals to equipment vendors to develop and deliver UMTS and subsequently HSDPA. Cingular plans to begin UMTS/HSDPA trials in Atlanta this summer with Lucent with the intention of rolling out UMTS services in 2005.

Kris Rinne, acting chief technical officer with Cingular, said the carrier anticipates rolling out HSDPA in the second half of 2005, but many unanswered questions remain. “The purpose of the RFP is to get side-by-side comparisons of timelines of various issues, one of which is HSDPA and its timeline for the 850 band,” she said.

Cingular will likely be the first global operator to launch UMTS and HSDPA at the 850 MHz band, which presents testing challenges and other technical issues that could result in deployment delays. While HSDPA is a software upgrade, another level of interoperability between devices and the infrastructure will be required. In addition, Cingular has yet to determine critical decisions such as the network’s total cost and where it will deploy the technology, answers it hopes to glean from the RFP.

However, Cingular has downplayed any competitive urgency to deploy HSDPA, indicating its nationwide EDGE launch, set to be completed this summer in connection with its acquisition of AT&T Wireless, which already offers a nationwide EDGE network, would offer a solid solution until WCDMA is deployed. EDGE offers average throughput speeds of 100 Kbps.

“Our commitment to EDGE has not changed,” said Rinne. “We still see EDGE as being a high-speed technology that will meet most customer demands, and we are requiring EDGE to be the backup solution for a WCDMA/HSDPA environment where quality is not acceptable. We’re continuing to look at the business case in terms of how deeply to deploy [WCDMA/HSDPA].” <<

- Eric -