Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Nokia Corp. (NOK) -- Ignore unavailable to you. Want to Upgrade?

To: Puck who wrote (1398)	10/1/2001 3:07:41 PM
From: carranza2	Read Replies (1) \| Respond to of 9255

I believe he claimed that CDMA technology in real world conditions was only 1 to 1.5 times as efficient as CDMA, which in his mind meant that very marginal spectral efficieny [sic] that CDMA might offer wasn't worth the price of deviating from GSM's worldwide economy. I've never quite been able to reconcile his statements I've tried but I can't reconcile those statements, either. I think CDMA is at least as efficient as CDMA, but I don't know. I'd be very interested in anything that's 1.5 times as efficient as itself. What does "GSM's worldwide economy" mean? Interesting choice of words.

To: Puck who wrote (1398)	10/1/2001 3:41:23 PM
From: Puck	Respond to of 9255

Addendum to post #1398. Sentence #3 should read:"I believe he claimed that CDMA technology in real world conditions was only 1 to 1.5 times as efficient as GSM..."

To: Puck who wrote (1398)	10/1/2001 5:00:59 PM
From: Eric L	Respond to of 9255

Puck, << Didn't Schmidt go on to run Omnipoint? >> Yes. George Schmitt (not Schmidt - my typo - sorry Puck and George) went from PrimeCo to Omnipoint in a matter of weeks back in early 1995. He is now Chairman and CEO of a DC area CLEC called e.spire Communications. espire.com << Are they still independent? >> Omnipoint was acquired and merged into VoiceStream. Announcement was made around March of 1999. George stayed on until the entities were formally merged in early 2000. VoiceStream isn't exactly tiny, but way too small to house two charismatic industry giants like George and John Stanton. << reading an interview of him a while after he left PrimeCo ... I believe he claimed that CDMA technology in real world conditions was only 1 to 1.5 times as efficient as CDMA, which in his mind meant that very marginal spectral efficiency that CDMA might offer wasn't worth the price of deviating from GSM's worldwide economy. I've never quite been able to reconcile his statements with the claims >> George is truly a GSM cheerleader, and is justified in being one. He is a wireless pioneer (although not from the tech side like Viterbi and Lee). He has always struck me as being larger than life. He launched Mannesmann D2 in Germany, way back when (92), and was extremely active in the GSM MoU before and after returning to the States. Prior to heading up PrimeCo George was AirTouch Communications, where he was EVP of AirTouch's (several) International Operations. What he might have said in any interview would be dependent upon the time in which the interview was granted, and maybe the mood he was in. When he left PrimeCo cdma was not yet fully commercialized despite what CDG says about BAM launching in March 96, they were planning no data services (would add later) and international roaming was far from their mind. GSM was already launched by at least 3 carriers in the States (Sprint Spectrum, BellSouth Mobility, and Western Wireless) and doing right nicely with data services (fax, SMS, OTA, et al) and international roaming was cranking up. I'm not sure that George would have said that CDMA had only 1.5 times the spectral efficiency of GSM (I heard him say back then 2 to 2.5 times and Hong Kong had launched trial the previous September at maybe 2x GSM. BooBird Frezza (who quoted or referenced George several times back then) might have attributed 1.5X to George). Time to market, roaming, economies of scale, data were all legitimate reasons that George did not want to be involved with IS-95 in early 1996. I don't know if you followed Frezza back then but he sure was irritating. One of the reasons he was irritating was that many of the things that he had to say were accurate. Eventually, however, IS-95 CDMA worked, and it worked darned well. I always think back on this when I hear the BooBirds chirping about how WCDMA isn't working. Eventually it will. I think Bill has been reincarnated, with a different technology bias, and now posts on Ramsey's Mod with Rules thread. By the time Omnipoint launched in NYC in November of 1996, as it turned out, PCS PrimeCo and Sprint PCS had already launched. Omnipoint had to launch NYC several times. It is one tough city to build out. Omnipoint became cash strapped along the way and a lot of the Omnipoint buildout needed to be redone with cash funneled in by Hutchinson Whampoa and moved over from VoiceStream even before the merger finalized. << Questions: In a practice, how efficient is CDMA compared with GSM, in your opinion? >> As for relative spectral efficiency, that's hard to pin down. network planning is tough science. As our favorite Finn points out CDMA works great in the desert but steel canyons and populated hillsides are another matter. It will different network to network to some degree because of terrain, real estate, et al. Generally accepted number has been that IS-95A is 3 to 3.5 x as efficient as GSM. That changes with new efficiencies incorporated into GSM. It changed dramatically when GSM introduced a new codec some time ago, so the 7 to 10 X GSM claims made early on were not real when CDMA was finally commercialized. Now 1xRTT claims to be 2x IS-95A (well, maybe 1.6x to 1.8x). Soon there will be a new GSM codec, frequency hopping, frequency reuse, smart antennas (and a whole batch of RF things that are honestly beyond my comprehension, so things will change again. This I know. George didn't have to tell me. Spectral efficiency is important. Its most certainly the only consideration when choosing technology. << And do you think that those wireless carriers who chose to go with CDMA have been noticeably advantaged or disadvantaged in their competitive position by their choice? >> In the USA we are most seriously spectrum constrained. Spectrum is in different frequency bands, not contiguous, chopped up geographically, and worst if all their is a spectrum cap of 45 MHz for any given piece of geography licensed by any given carrier. cdmaOne/cdma2000 as it has evolved is remarkably well suited to the USA, (the Americas). WCDMA is optimized initially for big chunks of (ideally) contiguous spectrum in frequency bands that are not allocated here for use by wireless carriers. I am of a mind that there is not a carrier in the USA that has sufficient spectrum today to deal with the amount of wireless data that will be coming down the pipe by the end of this decade - regardless of technology. On the other hand most have adequate spectrum for today and tomorrow. AWS and Verizon are in the catbird seat spectrum wise because they have more spectrum than the Cingulars, Sprint PCS, VoiceStream, and Nextel. I think this will eventually be solved by lifting of the spectrum cap, M&A, and technology advances. << Suppose that you were in charge of a new wireless carrier start-up in the U.S. whose plan was to build out a second generation network and then migrate to third generation technology along with the rest of the industry. Presume that financing was no issue. (You had all the cash you needed to fund your operations from the start and then for several years until your business was established and could tap conventional sources of corporate credit.) Which second generation technology and which third generation upgrade path would you choose: GSM/GPRS/WCDMA or CDMA/CDMA2000 or some non-conventional variant? >> Given your parameters ... cdma2000 hands down. 1xRTT for starters and I'd evaluate 1xEV-DO against the vaporware that today is 1xEV-DV, possibly opting never to implement 1xEV-DO. My answer would be different if I had a legacy network. I sure wouldn't want to have a legacy IS-136 network, or the mixed bag that Cingular has. - Eric -