Wireless Bill's Admission ...
... before he faded off into the sunset just as the sun started to rise on CDMA.
mQ,
<< From L M Ericsson inventing CDMA in the 1890s to Bill Frezza's rantings in the 1990s and "It's coming any time now and gee it's going to whiz-bang" ... >>
Frezza v. Brodsky and Brodsky v. Frezza. Fun wasn't it. Waiting for CDMA back then, was every bit as painful as waiting for VW-40© more recently. Technology sometimes evolves at its own pace.
This blast from the past was written about 3 weeks before I retired AMPS and immediately became a very satisfied CDMA subscriber on BAM using a QCP-800. Amazingly, the problems of the previous BAM NYNEX summer when I was a friendly user in Trenton, with analog to digital to analog handoff, were virtually non-existent.
When the article was written, CDMA had ~1½ million global subscribers, most in Korea. Today we are nudging 250 million with another 16 million WCDMA customers.
>> CDMA: The Revenge Of The Nerds
Bill Frezza
Network Computing
April 8, 1997
networkcomputing.com
They said it couldn't be done. Actually, I said it couldn't be done--but I had a lot of company. It looks like I was wrong. Nearly a decade since it was first conceptualized as a means to expand the capacity of analog cellular systems, Code Division Multiple Access (CDMA) digital cellular technology has clawed its way to the starting line in the global wireless sweepstakes.
Although the lead enjoyed by rival Time Division Multiple Access (TDMA) technologies--with more than 50 million users worldwide--continues to widen, even the most adamant skeptics have to admit CDMA is here to stay. With the war of the wh ite papers over, the battle has moved to the market. The outcome remains uncertain but, win or lose, the creative energy exhibited by feisty San Diego start-up QUALCOMM to prove its critics wrong is a testament to the motivating force of a powerful vision. The fact that CDMA never delivered on its most aggressive promises was lost in the hoopla of system launches. It finally delivered something, and the evidence is in that this something will not keel over the first time it's loaded up. It is now up to QUALCOMM and its army of licensees to demonstrate that CDMA delivers sufficient competitive advantage to earn a return on the massive investment required to get it off the ground.
We've been following the fortunes of the CDMA camp for several years QUALCOMM's still unmet capacity claims, as well as the incestuous relationship between the a nalysts and investment bankers that have helped QUALCOMM and its partners raise billions, have provided zesty column fodder. This has fueled a lively debate in the industry and at my online forum where impassioned CDMA advocates have been beating me about the head and shoulders trying to get me to eat my words. (Note that I am dutifully munching a few right here.)
Truth be told, now that the arguments have shifted from analyzing suspect technical claims to tracking commercial launches, CDMA does look a lot more promising. Yes, the consensus from operators is that CDMA delivers at best an 8x-to-10x capacity improvement over analog, not the 10x-to-20x still brazenly touted by QUALCOMM. But nobody seems to care that QUALCOMM fibbed a little. CDMA still delivers more capacity than TDMA and, with one notable exception, this capacity has met the prudent expectations of the large network operators. And, yes, thanks to the vexing issue of power-balancing, engineering and installing CDMA systems turned out to be significantly more complex than TDMA. Operators feel that the added up-front effort will be repaid in lower system maintenance costs. Most important, key features that actually matter to users--in particular, voice quality and battery life--have proven to be first rate, particularly since the CDMA community abandoned its inferior first-generation 8-Kbps vocoder. In the final analysis, it is consumer features, as well as the marketing savvy and financial strength of the operators, that will determine who will prosper and who will bleed in head-to-head competition.
Back From the Brink
For a time, matters looked rather desperate as cellular operators overlaying CDMA channels on their existing analog systems got bogged down in a morass of schedule delays, technical glitches and inferior first-generation products. Putting a bold face on things, California cellular operator AirTouch announced a "strategy" to make digital service available only to its highest volume customers. The stated plan is to mig rate 10 percent to 15 percent of total billable minutes to CDMA. With an analog subscriber base growing many times faster than the paltry rate at which subscribers are switching to digital, this looks a lot like thinly veiled face-saving. AirTouch defends its 170 m illion dollar plus investment in CDMA--and its 5,000 or so digital subscribers--as positioning for the future.
Bell Atlantic NYNEX Mobile has similarly soft-pedaled the technology in the market, finally launching service in four cities at the beginning of 1997. The so-called commercial CDMA system in Trenton, N.J., where I've been a test user for almost a year now, has long been touted as CDMA's first U.S. success story. Yet the system is still not ready for prime time. Although the voice quality of the 13-Kbps vocoder is superb in the core coverage area, reliability is extremely ragged at the digital/analog boundaries, leading to repeated call-attempt failures and dropped calls.
To date, AirTouch, Bell Atlantic, U S West and a smattering of other cellular operators have gotten a few thousand subscribers to switch to CDMA. If this were the entire story, CDMA would have been written off as an unmitigated disaster. But two factors emerged to save CDMA's bacon: deep-pocketed Korean industrial planners and the advent of personal communications services (PCS) in the United States.
Welcome Korea Inc.
TDMA owes its position as the world's predominant digital cellular technology to European industrial planners, who mandated the TDMA-based Global System for Mobile Communication (GSM) standard continent-wide. Since then, GSM has been adopted in more than 100 countries, reaping the benefits of massive scale. Figuring two can play at this game, Korean telecommunications authorities, working h and-in-glove with major Korean consumer electronics firms, bet that backing CDMA could propel them into the exploding international telecom market. Mandating CDMA as the only digital standard in Korea and simultaneously choking off analog technology through regulatory restrictions, they created a protected domestic market where wobbly first-generation CDMA systems could take root. Despite initial quality and reliability problems, this has been a numerical success. More than 1 million CDMA phones have been sold in Korea, representing almost 90 percent of the current worldwide CDMA market. In the years ahead, we can expect to see the likes of Samsung, Goldstar, Hyundai and Maxon sally forth from their protected niche and attack worldwide markets with a vengeance.
Korea kept the wolf from the door during QUALCOMM's darkest hours, but a new day dawned when the shining star of the CDMA universe, PrimeCo Personal Communications, took the field. PrimeCo is a joint venture of AirTouch, Bell Atlantic, NYNEX and U S West--the same companies that have done such a poor job introducing CDMA into their existing analog cellular systems. But unlike the cellular incumbents who face the daunting task of upgrading existing analog systems--a problem that CDMA has proven it is poorly suited to solve--PrimeCo started with a clean sheet. Paying $1.1 billion for PCS frequencies in 16 cities, PrimeCo stunned the world by claiming it would have all 16 markets up and running with pure digital service by the end of 1996. To the chagrin of the naysayers, it actually did it in the largest simultaneous launch in wireless history. Preliminary results are sketchy, but by the time this column runs, PrimeCo hopes to have crossed the 100,000 subscriber mark.
Fate has not been quite as kind to Sprint PCS--a partnership of Sprint Corp. and cable operators Tele-Communications Inc. (TCI), Cox Communications and Comcast. Promising to launch 15 to 20 CDMA systems by mid-1996, Sprint quietly limped into nine cities beginning this past December. Much of the blame was laid at the feet of its major systems vendor, Northern Telecom. Sprint still claims it will launch all 65 of its markets by this summer, but I wouldn't hold my breath. This bumbling and stumbling does have a silver lining for the consumer. The later Sprint comes to market, the more it is going to have to rely on aggressive pricing to get attention. In the first of what may be a sign of things to come, Sprint has offered 5 cents per minute airtime in Pittsburgh. Holy spherical cow! Let the price wars begin.
The NextWave of the Future?
The prize for creating the most outrageous embodiment of CDMA hubris clearly goes to former QUALCOMM executive Allen Salmasi--the kamikaze of PCS. While all the other operators took QUALCOMM's most extreme capacity and coverage claims with a grain of salt, Salmasi conjured up a company, called NextWave Telecom, that built these promises into the heart of its business plan. Bidding up the price of PCS spectrum beyond all reasonable bounds--paying a whopping $4.7 billion for frequency rights that would allow it to deploy a nationwide footprint--NextWave boasted it would undercut all competition by selling only bulk airtime to resellers like MCI, thereby incurring no marketing, sales and customer support expenses. Salmasi certainly fulfilled his mission of keeping spectrum out of the hands of potential TDMA operators, but the hangover from this "irrational exuberance" is going to be a doozy. Nothing will turn PCS into a price-driven commodity faster than this "minutes of use" factory, assuming NextWave actually completes its financing and builds a network. I fervently hope NextWave survives until next year's annual CDMA column. You couldn't ask for more entertaining material.
Bill Frezza is the president of Wireless Computing Associates. <<
The only article I'm aware of that Wireless Bill wrote on the CDMA subject after the one above is this one:
>> Telecosmic Punditry: The World Through Gilder-Colored Glasses
Bill Frezza
Plugging In
CMP InternetWeek
November 29, 2000
internetweek.com
Few professional futurists have as devout a following as the self-appointed prophet of the grand trends in technology, George Gilder. Founder and majordomo of Forbes ASAP, publisher of the closely watched Gilder Technology Report and author of the recently published book "Telecosm," Gilder uses his probing intellect and passionate wit to bring to life stories of the key companies and entrepreneurs shaping the contemporary telecommunications scene.
Gilder is a delightful writer. It's hard not to admire his artful prose as you turn each page wondering what he might say next. Unfortunately, his analysis is infused with such persistent intellectual capriciousness and selective disregard for the facts that critics are often left bewildered as to his criteria for picking winners and losers. In a recent flap reported by The Wall Street Journal, James Crowe, the CEO of Level 3, suggested that Gilder dropped his company from favor after Level 3 declined to pay $100,000 to help sponsor a three-day Gilder-hosted conference. This was particularly confusing as Gilder had just a few months earlier declared that Level 3 was "poised to change the world."
Gilder first came to the attention of the telecom industry as an early booster of Qualcomm and its Code Division Multiple Access wireless technology. Bucking the established wisdom, Gilder correctly predicted that engineers could overcome the substantial technical obstacles required before CDMA could be made a commercial success. Succeed it did, in the process propelling Qualcomm into the first ranks of global telecommunications companies. Skeptics such as myself were forced to eat crow, especially after Qualcomm's stock finally caught fire last year and zoomed into the stratosphere.
Partisans celebrated, predicting the demise of rival Time Division Multiple Access technologies, foundation of the GSM standard, often characterized by Gilder in his speeches as the work of a fascist European conspiracy.
But facts are stubborn things. Not only are TDMA products as commercially viable as ever, but they continue to outsell CDMA, not just in the rest of the world, where there are about 375 million GSM subscribers compared with 75 million CDMA subscribers, but in the United States as well. Though you'd never know it by talking to the pundits, less than half of the digital wireless subscribers in the states currently use CDMA. The majority use one variant or another of TDMA. And the promised quality superiority of CDMA somehow isn't reflected in customer satisfaction figures. Sprint PCS, a vigorous proponent of CDMA, continues to experience churn rates near 3 percent per month. That means each year Sprint has to replace one third of its subscriber base just to stay even! Can this have anything to do with the so-called "breathing" phenomenon that causes CDMA cells to shrink under heavy load, dropping customers at the edge? Don't expect a straight answer from Gilder.
The part that really rankles, though, is the claim of vastly superior capacity. Yes, CDMA is more spectrally efficient than TDMA and on this merit has become part of the Third Generation standard toward which GSM is now evolving. But did the marginal capacity advantage warrant fragmenting the U.S. wireless market into so many incompatible standards that Europe and Japan have left us behind in the move to data?
Here's a fact challenge for Mr. Gilder. If you can show me a real metropolitan CDMA system (not a lab demo or a lone high-site out in the desert) that today supports 20 to 40 times the capacity of analog, I'll eat this column. If not, you can eat page 91 in your book where you repeat this discredited claim prior to mocking critics for daring to question its validity. (What do you say to that, George?)
Forget about Gilder's arcane technology arguments, supporters say. This man can sure pick stocks! Let's look at a few of the companies Gilder has touted. Does anyone remember Steinbrecher? Steinbrecher set out to build a broadband software-defined base station that promised to obsolete hardware-based channelization schemes. Great concept, but failed execution. I leave it as an exercise for the reader to find out which company ultimately solved this problem. (Said company will remain unnamed as my firm is an investor, and I don't use this column to tout stocks. By the way, its award-winning products are based on GSM.)
What about Globalstar, the Qualcomm-backed satellite-phone venture? Do you think CDMA will save it from going the way of Iridium? Anyone want to bet how far free-space optics darling Terrabeam will get before it crashes and burns?
But enough Gilder baiting. I'll save the rest for the next time we share a podium.
Bill Frezza is a general partner at Adams Capital Management. <<
As for Bill's comment above ...
"If you can show me a real metropolitan CDMA system (not a lab demo or a lone high-site out in the desert) that today supports 20 to 40 times the capacity of analog, I'll eat this column."
... today, with 1xRTT and the EVRC codec optimized for voice quality, Verizon wireless is experiencing ~25x AMPS, and voice quality is darned good.
Technology evolves.
Time for Bill to eat his column.
- Eric - |
