Page 4 of 10 FirstFirst 1 2 3 4 5 6 7 8 9 10 LastLast
Results 46 to 60 of 149

Thread: Why did Sprint and Verizon choose CDMA when the rest of the world chose GSM?

  1. #46
    Join Date
    Feb 2008
    Location
    right behind u! (made u look!)
    Posts
    980
    Device(s)
    bb tour
    Carrier(s)
    alltel/ATN
    Feedback Score
    0
    Quote Originally Posted by Steven Hurdle View Post
    I've read this thread with great interest, as people debated whether GSM or CDMA2000 was the smarter choice at various points.

    My question to all of you is: given the respective arguments, why do you think the Canadian carriers all switched to GSM by the HSPA stage, whereas Verizon and Sprint waited until the LTE stage to do so?

    Both countries had a similar situation with the carriers at the point we've been discussing (what 3G technology to roll out). In both countries most/all of the IS-136 providers had switched to GSM already, leaving the market divided approximately down the middle between GSM and CDMA2000. Why did it apparently make sense to switch to GSM at the 3G stage to Canadian carriers, but not until LTE for American carriers?

    I doubt it was roaming revenues from the 2010 Vancouver Olympics, as some have suggested, as it's hard to believe a couple of weeks of roaming revenue would be worth the capital expenditure of *nationwide* HSPA rollouts across (geographically) the second-largest country on Earth. The Olympics could have been the cherry on the top for something they already planned to do, not the reason to do it, I suspect. Besides the differing international frequencies, and international roaming agreements (or the lack thereof) would have limited the benefits of this anyway. It's also worth noting that Bell and Telus (AKA "Bellus") rolled out an HSPA network, they didn't also roll out a 2G GSM network, so only Americans with 3G handsets, and Europeans with tri-band and quad-band HSPA handsets, could have roamed on Bellus, a relatively small number of visitors to Vancouver in 2010.

    So if it wasn't the Olympics, what was it? Handset selection? Consumer demand? *Something* forced the issue in Canada sooner than in the U.S. Identifying that may help inform the rest of the discussion,
    I think some of the network flip flopping that happened was due to a buyout or merger of two companies running the different types of networks and was an effort to consolidate under one technology. I'm sure if you look at the time frame of those decisions gsm was chosen as its future looked brighter than cdma/evdo and probably allowed for an easier choice of handset availability as well. Look at the problems other cdma operators in the U.S. had with handset choices like uscellular and alltel. Vzw didn't have those types of issues because of the scale they purchased them on but I know from personal experience that alltel always lagged behind in newer model phones and often became the reason many customers left. Going with gsm left a much larger number of handsets that could be sold to customers and didn't require paying the manufacturer to build in custom settings that cdma phones required. Alltels excellent evdo coverge aat the time waas the reason I stayed but if at&t would have had hspa at the time in my area I would have switched just for the handset choices.

    via the HoFo App
    what is this new feature, and how can I manipulate it in order to make myself seem superior to those around me?

  2. #47
    Join Date
    May 2007
    Posts
    614
    Carrier(s)
    at&t
    Feedback Score
    0
    From what I understand. The main reason sprint choose CDMA over GSM was qualcomm gave them a killer deal on the network equipment in return sprint gave qualcomm a large greenfield spectrum CDMA network to show off to investors.


    via the HoFo App

  3. #48
    Join Date
    Feb 2006
    Posts
    22
    Feedback Score
    0
    Been using Verizon VOLTE for a few days now. I can say that from my current location Verizon is superior to AT&T (so is T-Mobile). Small data point but VOLTE at home works great.

  4. #49
    Join Date
    Dec 2012
    Location
    New Hampshire
    Posts
    2,648
    Device(s)
    S9+
    Carrier(s)
    Verizon
    Feedback Score
    0
    Quote Originally Posted by wolfs View Post
    Been using Verizon VOLTE for a few days now. I can say that from my current location Verizon is superior to AT&T (so is T-Mobile). Small data point but VOLTE at home works great.
    That's not really relevant to the topic..

  5. #50
    Join Date
    Jul 2004
    Location
    San Jose, CA
    Posts
    7,337
    Device(s)
    iPhone 7 Plus 256GB Apple Black
    Carrier(s)
    Verizon, T-Mobile
    Feedback Score
    0
    If you've never read it, here is one man's historical account about how the wireless industry went to GSM instead of CDMA. Basically it was mandated by law that all carriers in Europe use GSM and be interoperable so they wouldn't have incompatible networks (like we ended up with in the US.) Anyway it's a fascinating and educational read.
    It's an old piece written long ago but has historical value.
    http://denbeste.nu/cd_log_entries/2002/10/GSM3G.shtml
    iPhone 11 Pro 256GB on Verizon eSIM (Get More Unlimited plan) + Mint 2nd SIM.

  6. #51
    Join Date
    Apr 2002
    Location
    d.c.
    Posts
    1,955
    Device(s)
    pixel xl
    Carrier(s)
    google fi
    Feedback Score
    0
    Quote Originally Posted by RF9 View Post
    If you've never read it, here is one man's historical account about how the wireless industry went to GSM instead of CDMA. Basically it was mandated by law that all carriers in Europe use GSM and be interoperable so they wouldn't have incompatible networks (like we ended up with in the US.) Anyway it's a fascinating and educational read.
    It's an old piece written long ago but has historical value.
    http://denbeste.nu/cd_log_entries/2002/10/GSM3G.shtml

    old school..


    excellent.
    >:-,

    “There's no benchmark for how life's "supposed" to happen. There is no ideal world for you to wait around for. The world is always just what it is now, it's up to you how you respond to it.”
    ― Isaac Marion, Warm Bodies

  7. #52
    Join Date
    Nov 2007
    Posts
    1,102
    Device(s)
    Nokia Lumia 920
    Carrier(s)
    Fido
    Feedback Score
    0
    Quote Originally Posted by RF9 View Post
    If you've never read it, here is one man's historical account about how the wireless industry went to GSM instead of CDMA. Basically it was mandated by law that all carriers in Europe use GSM and be interoperable so they wouldn't have incompatible networks (like we ended up with in the US.) Anyway it's a fascinating and educational read.
    It's an old piece written long ago but has historical value.
    http://denbeste.nu/cd_log_entries/2002/10/GSM3G.shtml
    Thanks for sharing this link. It was quite interesting! Even if some of it is wrong IMO (other than nitpicking details like him using "baud" instead of "bps", we had big details like him suggesting Americans enjoyed interoperability as good as Europeans which is *ridiculous*). Also, all his predictions failed to come true. He predicted the GSM coalition wouldn't get WCDMA right, but they did. The author criticises Europe's decision to go with GSM, but doesn't have similar criticism for South Korea's government's decision to force CDMA on people. The author says that Europe failed because it chose co-operation over competition, but then heralds the CDMA coalition's decision to cooperate instead of compete. He says that no one other than Qualcomm could ever implement CDMA because Qualcomm decided to not document or patent a lot of what's necessary for a successful CDMA implementation, and yet celebrates how other (non-Qualcomm) members of the CDMA alliance are being successful with it. He complains about the taunting from GSM purists in the '90s, but then admits that it was mostly just Ericsson, and mostly just their head office staff as he compliments the engineers and un-sung heroes there. And then goes on a self-indulgent schadenfreude about CDMA2000's success, which is kind of amusing with the benefit of hindsight. He's lording GSM's failures and his anticipated ensuing death of the GSM Alliance, but they goes on to make WCDMA work, and they marshalled market forces to beat CDMA2000. For every carrier that adopted CDMA2000, the GSM Alliance convinced even more carriers to switch to HSPA (including Canada, where at one point only one of the big three was on HSPA, and then suddenly *every* carrier switched to HSPA).

    The author also makes a ridiculous assertion. He argues that IS-95's inherent superiority is why it beat GSM in the U.S. market. Except, I rather think that the fact that GSM didn't play nice with AMPS and the fact that there was no 850MHz implementation of GSM until years and years after the 800MHz carriers had made other choices, were significant. Had GSM 850 been set up at a much earlier stage, GSM likely would have been chosen by far more carriers.

    The author also seems to have failed to imagine what the IS-136 carriers would do. Almost every single one of them went GSM, and did so after this article was posted. Key IS-136 players like AT&T and Rogers went GSM, overnight turning GSM from a rump (perhaps 10% marketshare) to a dominant force vying with IS-95 for first place in North America.

    The author's not wrong that CDMA is a better air interface, and that was broadly accepted by the time the article was posted because the GSM Alliance had licenced it. There's no disputing it. But at the time that I adopted GSM (1998), you couldn't even send SMSs on any of the IS-136 or IS-95 carriers yet! Yes CDMA offered greater capacity than any flavour of TDMA, but IS-95 didn't offer the call quality at the time that GSM did. So despite his CDMA fanboyism, there were legitimate reasons to favour GSM. And, as history tells, the writing was one the wall as HSPA beat the pants off of CDMA2000 in the author's beloved all-knowing and infallible open market.

    As I say, the article was interesting, and I enjoyed reading it. But those in glass houses shouldn't be the first to throw stones, and all that.
    Asus ROG Phone -- SIM 1: Rogers (10GB + unlimited) -- SIM 2: Public Mobile (talk/text/250 MB)

  8. #53
    Join Date
    Apr 2012
    Posts
    1,888
    Feedback Score
    0
    Quote Originally Posted by Steven Hurdle View Post
    Thanks for sharing this link. It was quite interesting! Even if some of it is wrong IMO (other than nitpicking details like him using "baud" instead of "bps", we had big details like him suggesting Americans enjoyed interoperability as good as Europeans which is *ridiculous*). Also, all his predictions failed to come true. He predicted the GSM coalition wouldn't get WCDMA right, but they did. The author criticises Europe's decision to go with GSM, but doesn't have similar criticism for South Korea's government's decision to force CDMA on people. The author says that Europe failed because it chose co-operation over competition, but then heralds the CDMA coalition's decision to cooperate instead of compete. He says that no one other than Qualcomm could ever implement CDMA because Qualcomm decided to not document or patent a lot of what's necessary for a successful CDMA implementation, and yet celebrates how other (non-Qualcomm) members of the CDMA alliance are being successful with it. He complains about the taunting from GSM purists in the '90s, but then admits that it was mostly just Ericsson, and mostly just their head office staff as he compliments the engineers and un-sung heroes there. And then goes on a self-indulgent schadenfreude about CDMA2000's success, which is kind of amusing with the benefit of hindsight. He's lording GSM's failures and his anticipated ensuing death of the GSM Alliance, but they goes on to make WCDMA work, and they marshalled market forces to beat CDMA2000. For every carrier that adopted CDMA2000, the GSM Alliance convinced even more carriers to switch to HSPA (including Canada, where at one point only one of the big three was on HSPA, and then suddenly *every* carrier switched to HSPA).

    The author also makes a ridiculous assertion. He argues that IS-95's inherent superiority is why it beat GSM in the U.S. market. Except, I rather think that the fact that GSM didn't play nice with AMPS and the fact that there was no 850MHz implementation of GSM until years and years after the 800MHz carriers had made other choices, were significant. Had GSM 850 been set up at a much earlier stage, GSM likely would have been chosen by far more carriers.

    The author also seems to have failed to imagine what the IS-136 carriers would do. Almost every single one of them went GSM, and did so after this article was posted. Key IS-136 players like AT&T and Rogers went GSM, overnight turning GSM from a rump (perhaps 10% marketshare) to a dominant force vying with IS-95 for first place in North America.

    The author's not wrong that CDMA is a better air interface, and that was broadly accepted by the time the article was posted because the GSM Alliance had licenced it. There's no disputing it. But at the time that I adopted GSM (1998), you couldn't even send SMSs on any of the IS-136 or IS-95 carriers yet! Yes CDMA offered greater capacity than any flavour of TDMA, but IS-95 didn't offer the call quality at the time that GSM did. So despite his CDMA fanboyism, there were legitimate reasons to favour GSM. And, as history tells, the writing was one the wall as HSPA beat the pants off of CDMA2000 in the author's beloved all-knowing and infallible open market.

    As I say, the article was interesting, and I enjoyed reading it. But those in glass houses shouldn't be the first to throw stones, and all that.
    HSPA didn't beat the pants as you say over CDMA here as CDMA carriers immediately started deploying LTE when HSPA started getting faster.. During the slow slow time when HSPA+ was being deployed in the US, it was plagued with dropped calls and slow/failed data connections vs. 1X/EVDO.... And actually at the time Verizon was deploying LTE, AT&T's HSPA+ wasn't performing any better than EVDO was in many markets..

  9. #54
    Join Date
    Oct 2005
    Posts
    5,346
    Device(s)
    N/A
    Carrier(s)
    Verizon Wireless;
    Feedback Score
    0
    One thing to remember too, AT&T and Cingular had gone from analog to TDMA initially. I think everyone knew TDMA was an interim solution though; it interoperated with exisiting analog switches, simply replacing 1 analog channel at a time with 6 digital timeslots. By the time VZW and Sprint were well underway with EVDO, AT&T decided to migrate one market (Seattle if I recall correctly) to UMTS to meet an agreement with NTT Docomo (who was doing UMTS in Japan.) They later ended up scrapping that UMTS network and started over, the equipment they bought was immature and apparently had no upgrade path to HSPA and so on.

    I would say even a year or two later, it would have made more sense to use UMTS when building out from scratch. But Sprint and VZW already had live EVDO networks before UMTS came out. Also keep in mind, with 1x and analog interoperability, and 1x and evdo interoperability, and current LTE interoperability, Sprint and T-Mo haven't had to go through a single instance of having to run duplicate networks or replace people's phones in bulk as AT&T, the CDMA->HSPA Canadian carriers, etc. have had to do.

  10. #55
    Join Date
    Apr 2002
    Location
    Olympia, WA
    Posts
    2,963
    Device(s)
    HTC One M8
    Carrier(s)
    Sprint
    Feedback Score
    0
    This is a long post... but I think this has the answers (at least as I see it) -- would be curious if everyone else agrees:

    Something that everyone is forgetting: CDMA probably would have been the de-facto "US standard" while GSM would have ended up with just Voicestream (Western Wireless, who started Voicestream was a CDMA provider, but chose GSM for their PCS licenses and spun off the company -- for a reason I'm still trying to dig up) -- and a Nextel-like situation of no other roaming partners -- except that one company stepped in: NTT DoCoMo.

    AT&T Blue, the largest TDMA provider in the US was going to take the CDMA route -- but then they got a sizable investment by NTT DoCoMo around 2001, which specified GSM, and a launch of UMTS see:
    http://connectedplanetonline.com/bro...ss_launches_7/

    I remember when this network launched -- and I played with a Motorola A845 at the time and thought it was slick -- I almost would have jumped, but the UMTS coverage stopped about 15 miles north of me.

    Had AT&T Blue went CDMA, the US would have been standardized around Qualcomm's CDMA... Qualcomm probably would have pushed ahead with UMB (their stillborn LTE competitor) -- and we'd probably have a mess of standards. So in the end, it's worked out for the best -- now we just have a mess of frequencies that LTE is on, but that is being worked out fairly fast through multi-band radios.

    A couple reasons why a number of US carriers picked CDMA over GSM:

    1. It was designed here in the US, by a US company (Qualcomm) -- and said US company had a base station business (that they later sold to Lucent), and a phone business (that they later sold to Kyocera) -- they offered extremely competitive pricing and financing for network builds -- that's the big reason Sprint cancelled plans for GSM and went CDMA.

    2. Being a US standard, CDMA was backwards compatible with AMPS switching and billing -- it's the reason it doesn't support network codes and "advanced" GSM calling features -- it works the exact same was as your US based landline, with * codes and flashing between calls -- and it still does. Plus the carriers had more CONTROL with no SIM cards and no "universal requirements". The HUGE difference of why GSM has always offered better calling features is that GSM's core (MAP) is "authenticate then connect", while the AMPS/CDMA/TDMA core (ANSI) is "connect then authenticate" -- just like a landline.

    3. Due to #2, GSM didn't play well with AMPS -- the authentication was totally opposite, the back ends were different, the chipsets were completely different. I remember for a VERY short while, Voicestream and Nokia released an "Analog Sleeve" for the 51xx series phone -- it basically fit between the phone and the battery, added a good bit of additional width, and was quite a "kludge" that you had to have if you wanted any coverage outside their native area. If you didn't have a Nokia 51xx you couldn't get it, and were stuck with the coverage you got on a developing PCS network... There was another "kludge" that came out around 2001: GAIT phones. A Siemens and a couple Nokias (I actually had a Siemens S46 -- it was a nice little phone) that combined AMPS, TDMA and GSM -- basically to take advantage of a carrier's TDMA on 850 as well as their GSM on 1900.

    4. It offered better call quality and capacity than IS-136 TDMA -- TDMA was originally called D-AMPS (digital AMPS) -- and those carriers that wanted to upgrade to digital before CDMA was available (1996) could really only pick TDMA due to:

    5. There was no GSM band class for 850 until around 2005 -- which meant carriers that only had cellular licenses only had one option until CDMA, which was TDMA. Had the GSM association created a band class for 850 around 1994 they would have had a MUCH easier time. It would have been a harder upgrade than a D-AMPs upgrade (which used the same AMPs channel scheme and just added digital carriers, as well as the same ANSI backend that CDMA later used) -- but the voice quality and capacity would have been a huge selling point. I have NO IDEA why it took as long as it did. This means that AT&T Blue and any other carrier that upgraded to GSM before 2005 stuck it on 1900, even if they had plenty of 850 available. It's also why some carriers like US Cellular, that started having capacity issues with TDMA before 2005 moved to CDMA: They didn't have any PCS, and there were no GSM 850 devices or equipment.

    --and--

    6: The reason why most TDMA carriers moved to GSM is that they had already moved to a digital technology early -- they were able to wring out every last bit of life out of it and hold out (AT&T having the best example of this: anyone remember CDPD data and AT&T PocketNet? It was a "wedge" type upgrade that let TDMA phones get 19.2k data -- whereas CDMA had data from the beginning -- I also remember reading that AT&T had another trick up their sleeve to extend TDMA's life that ended up not happening due to DoCoMo's involvement: tacking a GPRS data layer to the TDMA network so they wouldn't have to change out the core network -- luckily DoCoMo forced their hand, because TDMA voice quality really truly sucked), and when they were ready to switch they were able to actually weigh and compare GSM and CDMA--since by 2005 UMTS was already available, they already knew where GSM was going (WCDMA), and by picking GSM at this point they got the benefits of CDMA as well as world compatibility and scale of GSM. In particular with carriers that were primarily 850, they either switched to TDMA before 1996 and then clung onto it, or CDMA if they were late switching from AMPS between 1996-2005. If they were a TDMA carrier, and made the move to a better technology before 2005, it was to CDMA (US Cellular being the big one).

    In addition, it was Airtouch who "commissioned" CDMA since they needed something better than TDMA -- I'm not sure if this was before or after Vodafone got involved with them. I'd be surprised if it was after Vodafone, as that was one of their gripes about Verizon -- and you'd think they would have been able to force an 850 band class...

    So in short, it could have went either way:

    Had AT&T not went to GSM (and they did it at only 1900 initially), it would have been VERY much reduced to Voicestream's PCS and some small regional PCS's they bought out (eg. Pacific Bell in California); Cingular would have went CDMA as well. T-Mobile would probably have a wider GSM network than they already have, or DT would have got out and sold it to someone who would have switched to CDMA. There would still be 2 standards since UMB probably would have moved forward...

    Had there been a GSM band class for 850 around 1994, TDMA may not have happened -- or if it did, it would have been the "cheap option", and those carriers would have still upgraded to GSM by now. Initial CDMA was not great -- but it was still better than TDMA. GSM at that point was already proven -- but I'm not sure if CDMA would have even been commissioned by Airtouch... We would have had one standard from a long time ago.

    You also have to remember that GSM was created in Europe in the analog days for the same reason that everyone is finally standardizing on LTE rather than CDMA2000: Europe had tons of different Analog standards (5 IIRC) and there was no roaming between countries... if anything was going to "match up", they would ALL have to rip out their core networks and go to the same thing -- which was GSM. In the US, there was only ONE analog standard, AMPS -- so CDMA was our logical "advanced answer" to match up with our AMPS network (TDMA was like the first try with limitations) -- whereas GSM required more of a rip and replace.

    Also in the case of Qualcomm -- they ended up making out pretty good anyway, even though they no longer control their own standard... they still make the VAST majority of LTE and UMTS chipsets, and because CDMA is part of UMTS, they've made tons of money on their patents. I don't think their hurting...

    So, I think the big answer comes down to why it took the GSM association until 2005 to come up with an 850 band class... they would a lock on the US market a LOT sooner, had they done this. Otherwise, it's mainly a matter of money... in this case NTT DoCoMo investing in AT&T for picking GSM.

    N

  11. #56
    Join Date
    Apr 2007
    Location
    Michigan
    Posts
    1,330
    Device(s)
    iPhone 11 Pro Max
    Carrier(s)
    AT&T
    Feedback Score
    0
    Quote Originally Posted by NGeorge View Post
    This is a long post... but I think this has the answers (at least as I see it) -- would be curious if everyone else agrees:

    Something that everyone is forgetting: CDMA probably would have been the de-facto "US standard" while GSM would have ended up with just Voicestream (Western Wireless, who started Voicestream was a CDMA provider, but chose GSM for their PCS licenses and spun off the company -- for a reason I'm still trying to dig up) -- and a Nextel-like situation of no other roaming partners -- except that one company stepped in: NTT DoCoMo.

    AT&T Blue, the largest TDMA provider in the US was going to take the CDMA route -- but then they got a sizable investment by NTT DoCoMo around 2001, which specified GSM, and a launch of UMTS see:
    http://connectedplanetonline.com/bro...ss_launches_7/

    I remember when this network launched -- and I played with a Motorola A845 at the time and thought it was slick -- I almost would have jumped, but the UMTS coverage stopped about 15 miles north of me.

    Had AT&T Blue went CDMA, the US would have been standardized around Qualcomm's CDMA... Qualcomm probably would have pushed ahead with UMB (their stillborn LTE competitor) -- and we'd probably have a mess of standards. So in the end, it's worked out for the best -- now we just have a mess of frequencies that LTE is on, but that is being worked out fairly fast through multi-band radios.

    A couple reasons why a number of US carriers picked CDMA over GSM:

    1. It was designed here in the US, by a US company (Qualcomm) -- and said US company had a base station business (that they later sold to Lucent), and a phone business (that they later sold to Kyocera) -- they offered extremely competitive pricing and financing for network builds -- that's the big reason Sprint cancelled plans for GSM and went CDMA.

    2. Being a US standard, CDMA was backwards compatible with AMPS switching and billing -- it's the reason it doesn't support network codes and "advanced" GSM calling features -- it works the exact same was as your US based landline, with * codes and flashing between calls -- and it still does. Plus the carriers had more CONTROL with no SIM cards and no "universal requirements". The HUGE difference of why GSM has always offered better calling features is that GSM's core (MAP) is "authenticate then connect", while the AMPS/CDMA/TDMA core (ANSI) is "connect then authenticate" -- just like a landline.

    3. Due to #2, GSM didn't play well with AMPS -- the authentication was totally opposite, the back ends were different, the chipsets were completely different. I remember for a VERY short while, Voicestream and Nokia released an "Analog Sleeve" for the 51xx series phone -- it basically fit between the phone and the battery, added a good bit of additional width, and was quite a "kludge" that you had to have if you wanted any coverage outside their native area. If you didn't have a Nokia 51xx you couldn't get it, and were stuck with the coverage you got on a developing PCS network... There was another "kludge" that came out around 2001: GAIT phones. A Siemens and a couple Nokias (I actually had a Siemens S46 -- it was a nice little phone) that combined AMPS, TDMA and GSM -- basically to take advantage of a carrier's TDMA on 850 as well as their GSM on 1900.

    4. It offered better call quality and capacity than IS-136 TDMA -- TDMA was originally called D-AMPS (digital AMPS) -- and those carriers that wanted to upgrade to digital before CDMA was available (1996) could really only pick TDMA due to:

    5. There was no GSM band class for 850 until around 2005 -- which meant carriers that only had cellular licenses only had one option until CDMA, which was TDMA. Had the GSM association created a band class for 850 around 1994 they would have had a MUCH easier time. It would have been a harder upgrade than a D-AMPs upgrade (which used the same AMPs channel scheme and just added digital carriers, as well as the same ANSI backend that CDMA later used) -- but the voice quality and capacity would have been a huge selling point. I have NO IDEA why it took as long as it did. This means that AT&T Blue and any other carrier that upgraded to GSM before 2005 stuck it on 1900, even if they had plenty of 850 available. It's also why some carriers like US Cellular, that started having capacity issues with TDMA before 2005 moved to CDMA: They didn't have any PCS, and there were no GSM 850 devices or equipment.

    --and--

    6: The reason why most TDMA carriers moved to GSM is that they had already moved to a digital technology early -- they were able to wring out every last bit of life out of it and hold out (AT&T having the best example of this: anyone remember CDPD data and AT&T PocketNet? It was a "wedge" type upgrade that let TDMA phones get 19.2k data -- whereas CDMA had data from the beginning -- I also remember reading that AT&T had another trick up their sleeve to extend TDMA's life that ended up not happening due to DoCoMo's involvement: tacking a GPRS data layer to the TDMA network so they wouldn't have to change out the core network -- luckily DoCoMo forced their hand, because TDMA voice quality really truly sucked), and when they were ready to switch they were able to actually weigh and compare GSM and CDMA--since by 2005 UMTS was already available, they already knew where GSM was going (WCDMA), and by picking GSM at this point they got the benefits of CDMA as well as world compatibility and scale of GSM. In particular with carriers that were primarily 850, they either switched to TDMA before 1996 and then clung onto it, or CDMA if they were late switching from AMPS between 1996-2005. If they were a TDMA carrier, and made the move to a better technology before 2005, it was to CDMA (US Cellular being the big one).

    In addition, it was Airtouch who "commissioned" CDMA since they needed something better than TDMA -- I'm not sure if this was before or after Vodafone got involved with them. I'd be surprised if it was after Vodafone, as that was one of their gripes about Verizon -- and you'd think they would have been able to force an 850 band class...

    So in short, it could have went either way:

    Had AT&T not went to GSM (and they did it at only 1900 initially), it would have been VERY much reduced to Voicestream's PCS and some small regional PCS's they bought out (eg. Pacific Bell in California); Cingular would have went CDMA as well. T-Mobile would probably have a wider GSM network than they already have, or DT would have got out and sold it to someone who would have switched to CDMA. There would still be 2 standards since UMB probably would have moved forward...

    Had there been a GSM band class for 850 around 1994, TDMA may not have happened -- or if it did, it would have been the "cheap option", and those carriers would have still upgraded to GSM by now. Initial CDMA was not great -- but it was still better than TDMA. GSM at that point was already proven -- but I'm not sure if CDMA would have even been commissioned by Airtouch... We would have had one standard from a long time ago.

    You also have to remember that GSM was created in Europe in the analog days for the same reason that everyone is finally standardizing on LTE rather than CDMA2000: Europe had tons of different Analog standards (5 IIRC) and there was no roaming between countries... if anything was going to "match up", they would ALL have to rip out their core networks and go to the same thing -- which was GSM. In the US, there was only ONE analog standard, AMPS -- so CDMA was our logical "advanced answer" to match up with our AMPS network (TDMA was like the first try with limitations) -- whereas GSM required more of a rip and replace.

    Also in the case of Qualcomm -- they ended up making out pretty good anyway, even though they no longer control their own standard... they still make the VAST majority of LTE and UMTS chipsets, and because CDMA is part of UMTS, they've made tons of money on their patents. I don't think their hurting...

    So, I think the big answer comes down to why it took the GSM association until 2005 to come up with an 850 band class... they would a lock on the US market a LOT sooner, had they done this. Otherwise, it's mainly a matter of money... in this case NTT DoCoMo investing in AT&T for picking GSM.

    N
    This might be one of the best posts on HoFo I've read all year.
    Your results may vary. Network performance differs per user location.

  12. #57
    Join Date
    Apr 2002
    Location
    Olympia, WA
    Posts
    2,963
    Device(s)
    HTC One M8
    Carrier(s)
    Sprint
    Feedback Score
    0
    Quote Originally Posted by TWX View Post
    This might be one of the best posts on HoFo I've read all year.
    Thanks

    One other thing I realized I forgot to add was why Canada went HSDPA -- this stuff makes sense if you parse the timelines out:

    Canada was an AMPS country as well, and their natural progression was AMPS to TDMA then CDMA to UMB (if UMB would have happened) as well.

    Canada's situation is also very similar to ours -- 4 major players (at the time); one jumped early to TDMA (Rogers, who had a partnership with AT&T at the time), two went digital a year or two later and went CDMA (Bell and Telus), and one that only had PCS licenses and went GSM early (Fido)

    In addition, the reason Bell and Telus went CDMA is the same reason it made sense for a lot of carriers--there was no 850 GSM: Although they have a "nationwide presence" it's through a complex sharing agreement -- Bell primarily has PCS west of Manitoba, Telus primarily is PCS east of Manitoba -- they share their 850 networks... since there was no GSM band class for 850, there was no good way to do this until after 2005 with a GSM technology...

    So Rogers had a partnership with AT&T until 2004 -- so they matched whatever they did; NTT DoCoMo got them to move to GSM -- so Rogers followed... I think we can ultimately blame this one on NTT DoCoMo as well as had this not happened in the US, Rogers probably would have went CDMA.

    So here's why I think they went HSDPA before LTE rather than just jump to LTE like Verizon was primarily because we didn't have multi-band phones as much as we have today:

    Canada didn't auction 700 mhz. spectrum for LTE until last year -- and the Canadian CDMA carriers got Verizon's phone models unless they paid through the nose for custom runs. So Canada has been Band 4 primarily for LTE until now -- the US Carrier who is Band 4 for LTE is T-Mobile in the US -- and they are UMTS... the ONLY phones that were Band 4 LTE + CDMA 850/1900 were those custom built for MetroPCS and later, Cricket... not a big economy of scale or selection. Whereas they followed Verizon for their CDMA devices -- and couldn't for LTE. Rogers already went UMTS -- if they had to do custom CDMA + LTE runs or source with Metro/Cricket, their selection would be pathetic compared to Rogers.

    In addition, although Canada used our band plan for 700, they didn't auction the "Upper C" block in one big chunk; it's split into C1 and C2 -- and there was no promise that the CDMA carriers would have got C1 and C2 -- Rogers could have just as easily picked it up (in fact I think Videotron ended up with chunks of it nationwide) -- so there was no promise of a wide array of phones.

    Also Canada's 2600 is Band 7 FDD while ours is Band 41... again, no economy of scale since I don't know of any CDMA + Band 7 LTE carriers -- and if they are, they probably didn't use 850/1900 for CDMA...

    So I think phone selection and price was why they jumped to UMTS first -- there is any phone you can imagine under the sun if you are UMTS + LTE -- while you get what is on Verizon or Sprint if you are CDMA + LTE -- since they lost their major compatibility with them when they had to launch LTE on AWS before 700, that made it tough...

    The excuse of them doing it for the Olympics was BS -- Rogers CLEANED UP on Olympics roaming -- Bell still had to sign roaming agreements worldwide.

    N

  13. #58
    Join Date
    Nov 2007
    Posts
    1,102
    Device(s)
    Nokia Lumia 920
    Carrier(s)
    Fido
    Feedback Score
    0
    Wow! Your post was informative, and fascinating. Thanks for that! A few thoughts...

    Quote Originally Posted by NGeorge View Post
    AT&T Blue, the largest TDMA provider in the US was going to take the CDMA route -- but then they got a sizable investment by NTT DoCoMo around 2001, which specified GSM, and a launch of UMTS see:
    http://connectedplanetonline.com/bro...ss_launches_7/
    Interesting point! Definitely important. Though this was before Cingular bought AT&T, and Cingular already had plans to adopt UMTS according to your link. So had AT&T gone with CDMA2000 in those four cities, that spectrum might have been switched to UMTS later (think Sprint who started with GSM in a few cities and then switched to IS-95). Hard to say, of course, but it could have still gone UMTS (similar to how Cricket was switched from CDMA2000 to HSPA after AT&T bought them, and ditto with MetroPCS being switched to HSPA after T-Mobile bought them). It seems to me that Cingular's long-range plans were the more important consideration here.

    2. Being a US standard, CDMA was backwards compatible with AMPS switching and billing -- it's the reason it doesn't support network codes and "advanced" GSM calling features -- it works the exact same was as your US based landline, with * codes and flashing between calls -- and it still does. Plus the carriers had more CONTROL with no SIM cards and no "universal requirements". The HUGE difference of why GSM has always offered better calling features is that GSM's core (MAP) is "authenticate then connect", while the AMPS/CDMA/TDMA core (ANSI) is "connect then authenticate" -- just like a landline.
    I learned a lot reading this! And as for the analogue sleeve, I still have one (which was also used by Fido in addition to Voicestream). It was originally released with the Nokia 6190 in mind (though could probably be used with European Nokia 61xx handsets here when roaming, allowing the handset to roam strictly onto AMPS with no digital... a crazy situation!). Yes, the Nokia 5190 was later released and it worked just as well with that, but wasn't designed initially or specifically for 51xx handsets. It didn't allow hand-off to AMPS, BTW, so you'd lose your call when moving from GSM to analogue.

    There was another "kludge" that came out around 2001: GAIT phones. A Siemens and a couple Nokias (I actually had a Siemens S46 -- it was a nice little phone) that combined AMPS, TDMA and GSM -- basically to take advantage of a carrier's TDMA on 850 as well as their GSM on 1900.
    Woah... now that's a crazy kludge! Reminds me of some of the early IS-95 CDMA handsets that also had a GSM radio with European frequency support and a SIM slot. They were designed so that North American IS-95 users could roam in Europe (and parts of Asia and Australia). I'm not sure how many handsets like this actually came out, but I remember them being discussed.

    5. There was no GSM band class for 850 until around 2005 -- which meant carriers that only had cellular licenses only had one option until CDMA, which was TDMA. Had the GSM association created a band class for 850 around 1994 they would have had a MUCH easier time. It would have been a harder upgrade than a D-AMPs upgrade (which used the same AMPs channel scheme and just added digital carriers, as well as the same ANSI backend that CDMA later used) -- but the voice quality and capacity would have been a huge selling point. I have NO IDEA why it took as long as it did.
    This remains the $20K question (for those who remember $20K Pyramid). Why did it take so dang long for an 850 band plan? Especially because there were two small regional 800MHz carriers in the U.S. (none in Canada) who got PCS licences and went AMPS for 800, and GSM for PCS. PacBell and BellSouth both went GSM, despite having to stay AMPS at 800MHz, so there were even carriers who could have deployed it immediately. PacBell was bought by SBC, SBC later bought AT&T and called the new company AT&T, and then AT&T bought BellSouth.

    I also remember reading that AT&T had another trick up their sleeve to extend TDMA's life that ended up not happening due to DoCoMo's involvement: tacking a GPRS data layer to the TDMA network so they wouldn't have to change out the core network --
    I hadn't known this, but you're absolutely correct. In fact, they planned to not only do GPRS, but even EDGE: http://www.mobilecomms-technology.co...ts/tdma_is136/ (Scroll down to near the end for the relevant part.)

    In particular with carriers that were primarily 850, they either switched to TDMA before 1996 and then clung onto it, or CDMA if they were late switching from AMPS between 1996-2005. If they were a TDMA carrier, and made the move to a better technology before 2005, it was to CDMA (US Cellular being the big one).
    I intuitively like your proposal that whether a carrier went GSM/UMTS or CDMA2000 was about the technology they previously had, and when they felt compelled to upgrade, rather than the inherent superiority of one technology over the other. It just makes so much sense when you look at it that way. Market forces often win out over technological forces.

  14. #59
    Join Date
    Nov 2007
    Posts
    1,102
    Device(s)
    Nokia Lumia 920
    Carrier(s)
    Fido
    Feedback Score
    0
    Quote Originally Posted by NGeorge View Post
    Canada's situation is also very similar to ours -- 4 major players (at the time); one jumped early to TDMA (Rogers, who had a partnership with AT&T at the time), two went digital a year or two later and went CDMA (Bell and Telus), and one that only had PCS licenses and went GSM early (Fido)
    Plus one more national PCS licence, Clearnet PCS, that went IS-95 CDMA as well, and several regional local phone companies with 800MHz licences who all went IS-95 as well (SaskTel, MTS, TBayTel, and Aliant come to mind).

    In addition, the reason Bell and Telus went CDMA is the same reason it made sense for a lot of carriers--there was no 850 GSM: Although they have a "nationwide presence" it's through a complex sharing agreement -- Bell primarily has PCS west of Manitoba, Telus primarily is PCS east of Manitoba -- they share their 850 networks... since there was no GSM band class for 850, there was no good way to do this until after 2005 with a GSM technology...
    Telus bought Clearnet PCS, and suddenly was a national PCS provider. That brought them into competition with Bell. They later decided to cooperate, with Bell and Telus signing a comprehensive roaming agreement. Then they decided to create a joint HSPA network and decided to join the GSM upgrade path. Their CDMA2000 networks are separate to this day, as is Bell's BRS spectrum which Bell doesn't let Telus users roam onto. Last I knew that was also true of Bell's subsidiaries (Aliant, Northwesttel, Télébec, and NorthernTel).

    Canada didn't auction 700 mhz. spectrum for LTE until last year -- and the Canadian CDMA carriers got Verizon's phone models unless they paid through the nose for custom runs. So Canada has been Band 4 primarily for LTE until now -- the US Carrier who is Band 4 for LTE is T-Mobile in the US -- and they are UMTS... the ONLY phones that were Band 4 LTE + CDMA 850/1900 were those custom built for MetroPCS and later, Cricket... not a big economy of scale or selection. Whereas they followed Verizon for their CDMA devices -- and couldn't for LTE. Rogers already went UMTS -- if they had to do custom CDMA + LTE runs or source with Metro/Cricket, their selection would be pathetic compared to Rogers.
    I think you're onto something here. The same time as they launched their HSPA network, Bell and Telus (often referred to as "Bellus" up here by people who follow the wireless industry) launched the iPhone 3GS. I think phone selection did weigh very heavily in their decision to go HSPA, and your points about the frequencies they had available to them make sense. I'm not sure why Canada waited so long to auction off MBS.

    The excuse of them doing it for the Olympics was BS -- Rogers CLEANED UP on Olympics roaming -- Bell still had to sign roaming agreements worldwide.
    Agreed.

  15. #60
    Join Date
    Sep 2009
    Posts
    8,926
    Carrier(s)
    TMobile
    Feedback Score
    0
    It initially seemed like Verizon needed to really upgrade to lte with those slow 3g speeds they had to fallback on. Now it seems its really starting to pay off for them. In my area they have lte everywhere and are adding capacity also to fix the performance issues they once had.

    The drawbacks they had with cdma gave them a reason to get lte out there faster I think, and with how things are going they are in good shape

Page 4 of 10 FirstFirst 1 2 3 4 5 6 7 8 9 10 LastLast

Similar Threads

  1. Replies: 46
    Last Post: 03-25-2012, 08:13 AM
  2. Verizon Coverage in LA vs. the Rest of the USA?
    By GTwinkie in forum Verizon Wireless
    Replies: 16
    Last Post: 06-18-2004, 01:41 AM
  3. Replies: 3
    Last Post: 06-29-2003, 07:17 PM
  4. Verizon, where is the rest of the phones!
    By nidua in forum Verizon Wireless
    Replies: 1
    Last Post: 11-07-2002, 12:51 AM
  5. Replies: 49
    Last Post: 10-28-2002, 07:52 PM

Tags for this Thread

Bookmarks