Breakthrough Ideas (continued)
Direct Competitors
Slywotzky and Morrison (p. 305) argued that it is necessary to search beyond the traditional “companies that do what we do” to find “your true competitors are any companies that share your customers and/or your scope.” Although no company is comparable to Qualcomm in the leanness of its scope, there are direct and indirect competitors for its potential customers.
The standards war between CDMA2000 and GSM/WCDMA defines the direct competition. The battle lines were drawn based on allegiance to standards. The battles are over capturing operators and captivating end-users. The OEMs and infrastructure providers often sell arms to both sides. However, the battle lines demarcate the Royalty of Tele versus a New Wave of Brash Upstarts. To date, the propaganda war has been as important as the war waged between the Royal and the New Wave Pentagons, which represent the interplay of strategy, airlinks, spectrum, and technical armament. On the ground, the old guard attempts to hold its vast, but under armed, territory against the rapidly advancing onslaught of CDMA’s multi-pronged and technically superior attacks.
The three most commonly named direct competitors are Nokia, Ericsson, and Motorola. All three are vertically integrated mobile wireless (or more) companies, possessing a much broader scope, and having greater asset intensity. At the same time that all three are direct competitors within mobile wireless from the perspective of the standards war, they are also CDMA customers in the sense that all have licenses and pay royalties and also manufacture CDMA handsets or install CDMA infrastructure. This illustrates the broader scope of their business activities. What these three share is their collective presence as ringleaders of the GSM Big Five, plaintiffs in most of the patent lawsuits, and as sources of anti-Qualcomm expectations management, along with AT&T. NTT DoCoMo is a new ally of the UMTS camp and AT&T, but its form of WCDMA is not yet compatible with the recently “completed” release 99 for UMTS.
Also, there are distant drums, surging sounds from a rebel populous murmuring and muttering that there must be a better alternative to these failing telecom giants. Strange sounds of “WiFi for freedom,” or is it “WiFi for free?”
Given their stake in preserving the past, the Royalty of Tele want the world to believe that the present map, specifically GSM’s current market share of 80% and CDMA’s share of 20%, will also be the future map of the worldwide third-generation market. For the Royalists, the third-generation is to be business as usual for their European political oligopoly, not a New Wave, where first-mover dominance has shifted to CDMA and Asia. Let’s look at where the rival technologies and world markets stand today before we undertake the hazards inherent in predicting the future.
The Battlefields of Europe, Asia, and the USA. First, let’s listen to two correspondents covering the European 3G war front. On August 16, in a note entitled “Death Bell Tolls – 3G RIP,” Datamonitor analyst, Nick Greenway, who is annoyed that people are pussyfooting around, decided to tell it like it is. He said, “Within the next 12 months some license holders may shelve 3G aspirations, possibly having to treat the cost of licenses as a write-down.” He urged, “Give it away now – there’s more sense in abandoning the market altogether than rolling out a service on top of giant sunk costs.”
Given that the take-up of 2.5 G has been so abysmal, Greenway argued that abandoning the market altogether will prove cheaper than having to subsidize handsets, with subsidies ranging between E500 and E800 looking necessary, to make them affordable. “Do the maths,” said Greenway, “If a European operator with 12m subscribers wants even half of them to have a 3G handset, the cost of subsidy will exceed E3bn.
Telefonica and Sonera pulled out of their German joint venture, Quam, writing off over E8bn in the process. Orange, Vodaphone, and others lobby governments for delays in starting or extensions of time periods because of delayed or poorly functioning trial networks and bulky and balky dual mode handsets. Content providers lament their lousy margins. Dedicated start-ups are on their last legs. Greenway stated that all it takes are a couple of more shaky operators and licenses may become available at knockdown prices. According to Datamonitor:
Message 17885013
“…the industry is looking more than expectantly to Hutchinson 3G to blaze a trail out of the mire.
All this would be enough of a problem, even if anyone had made a successful 3G handset. But so far, no one has really managed to integrate the circuitry required for 2G (inc 2.5G) and 3G to exist on the same silicon wafer. In effect, all prototype 3G handsets at present comprise ‘two handset in one’ under the bonnet.
Nobody has even properly managed to make these double-chip 2G/3G combinations works successfully. Hutchinson 3G hopes to be the first major European 3G player to bring a service to market, hoping to launch in late 2002. It recently admitted, however, that its dual-mode handsets cannot currently hand-over calls between 2G and 3G networks, meaning that the user has to ring back. This is serious: if you pay E499 for a phone, you’ll be justifiably annoyed about having to re-dial your call every time you move from one network area to another. According to Hutchinson, the problem won’t be fixed until mid-2003, but they are by no means the only ones.
This hybrid phone naturally incurs a correspondingly larger BOM (bill of materials) for manufacturers, but they are in no position to absorb this through additional efficiencies in production – they’re sailing as close the wind as possible already when it comes to making margins on handsets. …If you are an operator, do you launch with no handsets and a substandard network or wait? You’re between a rock and hard place.”
Even as Greenway tries to break through the denial of Europe over the probable success of their bastardized version of 3G, he remains blind to what Qualcomm already has accomplished, not only in CDMA2000, but also, in GSM/UMTS. Somehow, Greenway fails to mention the ITU, whose opinion counts most, has declared the unmentionable: both CDMA 1X and 1xEV-DO meet its 3G standard’s requirements. Thus, CDMA2000 represents the successes that contrast with European 3G failures.
Not only does Qualcomm’s MSM6200, which is currently sampling, tightly integrate GSM and UMTS on a single silicon wafer, it also reliably hands off calls between GSM and UMTS networks. Moreover, its ZIF-architecture reduced the BOM by 30% or more to create the missing winning margins. Qualcomm’s multimode solution calls for an MSM6300, GSM1x, to overlay existing GSM that, when coupled with the MSM6200 [and the new MSM6250] dual-mode GSM/UMTS, potentially solves many European problems. If Europe adds the MSM6500 and MSM6600, it can join the New Wave’s world in using a seamless world phone for voice and high data rates. This would solve Europe’s compatibility problem, letting GSM and UMTS converse.
However, because WCDMA vendors and carriers are members of its value web, friendly-dominant Qualcomm is working actively to expedite a solution. In fact, on September 4, at the SSB 2002 Tech Conference, Qualcomm COO Tony Thornley made the following remarkable comments (emphasis added):
siliconinvestor.com
“In Europe, with WCDMA being delayed – not having matured as we have been saying for some considerable time, the CDMA market in Europe is developing slowly. However, the good new is that our chipset is clearly taking the lead in terms of its capabilities. And, operators in Europe are recognizing the value of Qualcomm technology in deploying their networks. I think what’s going to happen is that WCDMA is going to be rolled out from a network point of view in ’03, and then in ’04 and ’05 we’re going to start seeing subscriber growth – and I would say 10 to15 million is kind of where we’ve been looking at significant volume for 2004. I hope we would begin to see that kind of level of investment in WCDMA. We are working very hard at the standards level and, of course, with operators and manufacturers, to make sure all of the interoperability issues are resolved as quickly as possible. There is nothing fundamentally wrong with WCDMA, but, just simply that it does take a given amount of time to stabilize any kind of technology.”
Returning to the theme of 3G battles, on August 20 2002, according to the respected Economist (emphasis added):
economist.com
“Was 3G the right technology to back? In the 1990’s, Europe sped ahead of the United States in mobile usage because all countries subscribed to the same technology, making it possible to ‘roam’ using one handset – something that is still not possible across America. European countries decided to follow the same approach in moving to the third generation of mobile technology. Spain’s Telefonica has said that Europe’s 3G standards do not allow a system that is stable or cheap enough to compete with existing technologies. Europe’s 3G is based on one broadband technology WCDMA. A rival technology is CDMA. Experience in Japan, the first country where 3G has been launched, suggests that CDMA may have the edge. Japan’s NTT DoCoMo’s 3G launch, using WCDMA, has been slow to take off, while its rival, KDDI, using CDMA, has been gaining customers quickly.”
On August 22 2002, according to Probe Research’s study of the U.S. 3G battle: siliconinvestor.com
“The whole world is watching to see which standard—GSM or CDMA—will succeed as the 3G battle revs up in the United States, but the seemingly inevitable deployment of third generation wireless networks is hitting a brick wall in some areas and rushing headlong in others. The difference between the winners and losers appears to be the technology chosen by the carriers, said David Chamberlain, research director. …The little secret amongst friends is that the path from GSM to W-CDMA is not evolutionary but will require replacement of the radio access network and, in many cases, additional spectrum, said Chamberlain. …At this point even the most biased onlookers may soon have to admit that CDMA may have taken the 3G lead on a global scale, not just in the United States, said Probe. Probe’s comments were reinforced by RBC Dain Rauscher Inc. study that found 1xRTT data throughputs are generally greater than those of GPRS. RBC tested Sprint PCS’ service against T-Mobile’s GPRS service in several U. S. cities. In Dallas, Sprint’s network achieved data speeds up to 89 kbps, while T-Mobile network speeds reached only 33 kbps. …Brash upstart Qualcomm, Inc., however, is hardly a favorite of European carriers and regulators, reducing the effectiveness of the company’s message there.”
On June 2 2002, from the perspective of a Qualcomm bull, Ben Garrett (his emphases) posted the best analysis of Europe’s problems: Message 17546045
“Europe’s wireless carriers have run themselves aground. Once perceived the best engine for future growth, poor judgment and mounting debt are ruining corporations, destroying wealth and undermining the foundations for the greater economy. …In a gamble to maintain its commanding position …carriers collaborated with the European Telecommunications Standards Institute (ETSI) and the 3rd Generation Partnership Project (3GPP) to develop and control a new technology standard, UMTS wCDMA.
…In their euphoria, carriers also paid billions for spectrum licenses, …which aren’t expected to generate significant revenues until 2007. Europe’s gambit has not paid off. 3G network schedules are in a state of perpetual delay, and proposed ‘commercial’ announcement are thinly disguised trials. The 3G standard – designed in a futile attempt to circumvent North American intellectual property – is hampered by chronic problems
…According to EMC Market Data, there are approximately 2 million global users of GPRS services at the end of May 2002 – 0.3% of GSM subscribers. …To date, GPRS has proven an unstable bridge and a technical disappointment. EMC Market Data has estimated average speeds on Europe’s networks at 17-27 Kbps, and increased usage would further degrade performance. …Vendors are now marketing the virtues of an interim upgrade called EDGE to compensate for GPRS shortcomings. Once again exercising their gift for hyperbole, vendors promote EDGE as providing nearly 500kbps peak data speeds and a tripling of capacity. Still unproven, once again touted as easy and inexpensive, and rumored to require substantially greater cell station density, EMC Market Data pegs likely performance at 50-80 kbps. Contrary to capacity claims, EDGE provides no additional voice capacity. The upgrade would only increase data capacity or throughput speed for the timeslots(s) not required for voice service at any given moment – rendering EDGE upgrades meaningless without 2G infrastructure investments of an even greater magnitude than those required for GPRS data capacity. …
In order to differentiate the wCDMA standard from competing variants, 3GPP integrated controversial, untested technology – including asynchronous handoff – that now presents problems that appear beyond the solutions of today’s science. Remarkably, over 100 3GPP members claim essential intellectual property, with the committee-based development process cultivating a culture where corporate interests trump best science.”
On August 20 2002, after 3GPP “completed” Release 5, Garrett (his emphasis) wrote, “Core problems with Release 99 are integral to the standard. And powerful vendor members who injected essential intellectual property into the standard do not want it disturbed. Catch 22. A contrived, manipulated committee-based development process threatens the very viability of its own standard.”
Nonetheless, Thornley seemed to say that Qualcomm sought and found a solution to these interoperability problems with its own chipset that may be chosen by the carriers as their best available solution. This implies that by setting a de facto standard atop the fractionating 3GPP standard, Qualcomm can overcome many limitations their political process of setting standards has created.
On the Asian front in June, Morgan Stanley analyzed high data speeds on 1X in Korea, WCDMA in Japan, and GPRS in Hong Kong. According to Morgan Stanley, 1X had the “upper hand” in network speed, providing day-to-day speeds between 100 to 120 kbps, compared to NTT DoCoMo’s 80-90 kbps, and GPRS speeds of about 42 kbps. Morgan Stanley gave the handset nod to 1X also, commenting that color handset with camera could drive data usage. Next, Morgan Stanley noted the less expensive migration path of CDMA 1X, estimating that Sprint PCS will spend about $8 per pop, whereas AT&T will spend three times as much. Remarkably, Morgan Stanley still believed GPRS would make up in volume for what it lacked in quality, positing forecasted economies of scale that would drop GPRS handset prices below those of 1X.
3GAmericas [the GSM/UMTS Ministry of Propaganda] released a white paper on its web site in July that prompted a pointed response from Deutsche Bank that it entitled, “Alice in GSM Fantasyland.” According to Deutsche Bank’s Brian Modoff, Daniel Kaplan, and Michael Thelander (emphasis added):
outlook4mobility.com
“The white paper immediately begins with several misleading statements that set the tone for the rest of the paper. GSM is called the “de facto standard” with “capabilities today that match or exceed competing technologies.” We would argue that at least 50 CDMA networks, 120+ million CDMA subscribers and the ITU would have a differing opinion. …The biggest fallacy with the paper is that it treats future GSM voice capacity enhancements as being already proven and commercially available today, presenting result that were made under the most optimistic scenarios.
…The paper also tries to lump UMTS/WCDMA and GSM technology together, calling WCDMA an “evolution” from GSM and a “compelling upgrade from GSM.” If we believe the paper, “UMTS does not involve a complete replacement of user equipment and infrastructure.” UMTS, in reality is 100% different from GSM in the handset, the radio access network, and the core network. It is not a compelling upgrade, it is a completely new technology…
Further, all of our research and discussion with industry contacts suggest that those operators who initially deploy Release ’99 of the 3GPP standard, including EDGE, will require an entirely new core network (perhaps a new RAN as well) if they want to offer Release 5 services (real-time multimedia services, VOIP).”…
…Readers of our June 17th S2N may recall that we measured rates on Cingular’s network that were as high as the mid-20kbps, but that we were frequently being dropped due to congestion. Last time we checked, we were consistently measuring average rates above 60kbps and as high as 112kbps on Verizon’s 1X network…
In addition to presenting unrealistic GSM capacity claims based upon simulations and theoretical models, the paper understates the voice capacity of commercial CDMA2000 networks today and assumes that CDMA is a static technology that will not introduce its own improvements to further boost capacity and/or data rates. …
Finally, the paper inaccurately lumps WCDMA and GSM together, suggesting that WCDMA is a natural evolution from GSM. WCDMA is not a natural evolution from GSM, it is an entirely new technology that may even require an additional swap out of equipment to support Release 5 of the standard.”
On August 23 2002, according to wireless industry analyst Andrew Seybold:
Message 17912934
“It has taken the European press a very long time to see the cracks in the European plan to add 3G networks and services to their existing GSM/GPRS stable of wireless communication solutions. Now it appears as though every analyst, technology reporter and the business press has decided that the European wireless operators are in a lot of trouble because of their heavy investments in 3G spectrum, technical problems with the infrastructure, lack of a final standard and devices that won’t work on WCDMA let along WCDMA and GSM/GPRS.
Alas, the problems that need to be solved will not be solved, at least in Europe... There is too much PRIDE in the way, too much “not invented here” attitude and too many egos tied up…
The solution is this: Give the government back their 2.1-GHz spectrum, tell them what they can do with it, and at the same time get them to relax their “GSM only” stance on the existing bands. Then add in a little magic: Take a block of the existing GSM/GPRS spectrum and overlay CDMA2000 1X. (See, I told you the European wouldn’t like this.) GSM1X, as it is called, works today. It can be used on the same spectrum, cell sites and back-end and provide the same coverage with better voice capacity and much faster data speeds. If SIMs were used in the devices, a SIM could be taken out of a GSM phone and installed in a CDMA2000 1X phone.”
Qualcomm used it superior knowledge of spread spectrum to commercialize CDMA and to become the first-mover in 3G-spread spectrum, which naturally evolved from cdmaOne. Using its exceptional spread-spectrum integrated learning base, Qualcomm next added a data optimized mobile high data rate solution that is unequalled, possessing a multiyear lead over it competitors who have no such solution or even a plan for data optimization. Next, Qualcomm simplified its MSMs using ZIF to produce the radioOne multimode ASICs to do today what Europe cannot yet do, integrate its GSM with its UMTS. But also, Qualcomm will introduce a world phone next year that harmonizes all access modes, bands of spectrum, and network configurations. Qualcomm invests its license fees and royalties earned for its mastery of a proprietary spread spectrum architecture into further research, development, and investment into the CDMA 3G future that its shapes and advances.
By January of 1998, ETSI recognized that the modulation theory of spread spectrum promised a solution to the voice and data capacity criteria that the ITU determined must characterize third-generation wireless. Wanting to extend the hegemony of their politically proprietary GSM 2G digital standard and to profit from their installed base, they decided to set out on a similar strategic path: To establish their own 3GPP wideband standard to dominate 3G mobile wireless.
For this strategy to succeed, Europe needed an advanced spread spectrum technology that their GSM integrated learning base has yet to produce. Having failed in the patent courts when challenging Qualcomm, Europe tried to invoke its own unique blend of “GSM/WCDMA” intellectual property to force Qualcomm to cross license. This failed too. In the interim, they tried to agree upon a standard, but this very political process also failed. 3GPP hoped to develop a unique and superior WCDMA that would become a Universal Mobile Telecommunication Solution. They failed; instead creating a flawed form that is continually delayed and will under perform on several dimensions. To hold their installed base, 3GPP invented a faux GSM to UMTS Migration Path that let the vendors make more money at the expense of apparently gullible carriers.
Once Europe recognized that the Science said, “Code division multiple access is a superior modulation theory for increasing voice capacity because of universal frequency reuse and for high data rates because the processing gain simultaneously reduces interference, which is averaged as background noise, and increases accurate signal detection,” this logically implied that Europe should go full bore to develop a 3G solution.
Instead, Europe displayed what psychologists call functional fixedness: Give a boy a hammer, and suddenly everything needs hammering. Even though GSM’s TDMA can never evolve into CDMA, they fixated on further developing GSM-functions for data, creating unproven GPRS and EDGE to use multiple time-slots instead of moving immediately to spread spectrum. This “migration path” not only failed, it reduced their limited voice capacity, messed up their RAN plans, and made their phones hot. That it was likely to fail was the very reason that a shift to spread spectrum was required for 3G. As this diversion failed, they blamed each other: “It’s the handsets.” “No, it’s the infrastructure.” “No, it’s no demand for data; the customer isn’t interested.”
We must assume that Europe’s delay in introducing their “own” spread-spectrum architecture is a function of technical problems, not just a stratagem to gain short-term profit at ruinous expense to European carriers. It was necessary because of insufficient transfer of training from GSM technology to spread-spectrum technology. Simply put, I believe that Europe did not have the necessary integrated technical learning base, and it still may not have developed it, just as NTT is falling short of demonstrating this distinctive spread spectrum competence in FOMA. Fortunately, Qualcomm will assist them.
Recall how “engineer” described using 16 PhDs trained in spread spectrum modulation theory and the specifics workings and requirement of the cdmaOne Radio Access Network to ensure that his software and ASIC engineers correctly translated the nuances of DSP RF algorithms into code and hardwiring. “Engineer” also explained that Nokia’s engineers must first focus on performance. Only then, can they turn attention to the problems inherent in the complex mastery of power control and power drain that is so central to CDMA. No company other than Qualcomm appears to have this capability. San Diego has become a path dependent locus of spread spectrum engineers and communication companies just as Silicon Valley was for computer technology.
As a natural evolution, it was a small step for Qualcomm to move from 2G to 3G spread-spectrum. But, for Europe, this changeover was a giant leap into the abyss. Complexity always demands intricate solutions, which are often developed painfully, bit by bit. But when a complex problem is handicapped by an unsound definition of the necessary and sufficient conditions for its solution, it becomes virtually impossible for under-experienced engineers, simply beyond their means. Even CDMA experts can only make-do with an inelegant patch over for unnecessary problems created by the inclusion of politically contrived “GSM” standards, such as asynchrony or insufficient bandwidth in the guardbands for currently available filters.
The move from scientific theory to practical applications is a long and perilous journey. It requires translation fidelity. The abstract theory must be faithfully translated into a set of operations that transform it into a concrete embodiment. This translation is embodied as a set of standards. A political process that turns its back on science to embrace a politics of proprietary advantage substitutes economic motives for the required scientific impartiality. UMTS has built in the seeds of inefficiency into the very standards that sprout the glitches in its technology that the brash upstart, Qualcomm, seeks to patch to make WCDMA viable.
Therefore, Europe’s GSM Bridge was an apparition claiming to be a migration path, and a bastardized poser claiming to be Wideband-CDMA, a committee-created conundrum so puzzling as to mystify, resist, and defy a first-rate engineering solution. This is necessarily so because the politics of the 3GPP’s standard-setting process flouted central theoretical propositions that were crucial to the scientific theory of spread spectrum and lacked the engineering experience required to select appropriate architectural design rules, errors that came from a futile and botched effort to replicate the strategy inherent in GSM-as-politically-proprietary to the Big Five. The politically flawed 3GPP process failed to duplicate the GSM claim of “universal roaming” and, hence, it fails to deliver a winning strategy for world dominion. Thus, Europe’s hubris essentially stranded their GSM installed base. Today, Europe’s least-worst alternative choice is to adapt by substituting Qualcomm’s complete standardized solutions for their own cherished GSM/UMTS. Adaptation is always the best economic solution for losers in a standards war.
In summary: Qualcomm has strong and sustainable competitive advantage resulting from its strategic architectural control of third-generation spectral efficiency, high data rate generation, and seamless worldwide roaming. |
