Jump to content

Conversation about Sprint Moving to UMTS


Recommended Posts

I wouldn't call it "myopic", considering their consumers could get pretty good UMTS coverage out of the gate at 21 Mbps while EV-DO was limited to 3.1 Mbps.

 

Citing max speeds is a bit misleading. No user ever gets 21 Mbps on a single HSPA+ enhanced W-CDMA carrier. And in that same amount of spectrum, an operator could deploy four EV-DO carriers.

 

Then you consider that AWS was their only deployment option for LTE more or less in Canada due to the much longer period of time it took to clear broadcast out of 700 MHz up there, then it doesn't look like such a bad move, all things considered.

 

Had Bell and Telus stuck to their guns, they could have used their Cellular 850 MHz spectrum to deploy LTE. And that would have given them a big leg up on Rogers, which has never been a CDMA2000 operator.

 

AJ

  • Like 1
Link to comment
Share on other sites

 

I wouldn't call it "myopic", considering their consumers could get pretty good UMTS coverage out of the gate at 21 Mbps while EV-DO was limited to 3.1 Mbps. Then you consider that AWS was their only deployment option for LTE more or less in Canada due to the much longer period of time it took to clear broadcast out of 700 MHz up there, then it doesn't look like such a bad move, all things considered. Sure LTE may supplant UMTS in the long run, but as many people have accurately pointed out, VoLTE is a ways away, even now. When the triopoly up there gets LTE frequencies, they'll be in a really good position to launch LTE over the lower frequencies quickly.

 

LTE is supplanting UMTS now. Also, VoLTE in and of itself is not a long way out. VoLTE as an independent voice network is what will be a ways out. VoLTE will be backed up by 2G /3G networks for quite some time though. Just because of the device ecosystem alone.

 

Robert via Nexus 7 with Tapatalk HD

Link to comment
Share on other sites

Citing max speeds is a bit misleading. No user ever gets 21 Mbps on a single HSPA+ enhanced W-CDMA carrier. And in that same amount of spectrum, an operator could deploy four EV-DO carriers.

 

The actual throughput on their networks was in the 6 Mbps range at launch. Still better than than the 1.5 Mbps range EV-DO had. I'm glad you made that argument so I could clarify the actual speed ranges.

 

 

Had Bell and Telus stuck to their guns, they could have used their Cellular 850 MHz spectrum to deploy LTE. And that would have given them a big leg up on Rogers, which has never been a CDMA2000 operator.

 

AJ

 

That would also have stopped them from gathering roaming in the Olympics. I'm not defending the Canadian wireless industry in any shape or form. I'm merely pointing out they had a completely different set of circumstances than the CDMA2000 operators to the south.

Link to comment
Share on other sites

LTE is supplanting UMTS now. Also, VoLTE in and of itself is not a long way out. VoLTE as an independent voice network is what will be a ways out. VoLTE will be backed up by 2G /3G networks for quite some time though. Just because of the device ecosystem alone.

 

Robert via Nexus 7 with Tapatalk HD

 

It would have been a while before Bell or Telus could have used it for voice. That's the point.

 

I meant to say that VoLTE will be a while as an independent network, I needed to clarify that.

 

Also, to be fair, neither the CDMA development group the 3GPP2, nor the 3GPP, the body to move the GSM standard forward after ETSI gave up control of the GSM standards, won in this country. Consumers lost. Look at our mobile rates compared to the rest of the world.

 

I believe having a truly universal system built on a single standard would have been optimal, but protectionism on both sides of the pond meant that couldn't happen.

  • Like 1
Link to comment
Share on other sites

The actual throughput on their networks was in the 6 Mbps range at launch. Still better than than the 1.5 Mbps range EV-DO had. I'm glad you made that argument so I could clarify the actual speed ranges.

 

Because EV-DO Rev A caps out at 16-QAM, it is a more robust (albeit less efficient) airlink. And it never carries voice traffic that drags down its max speeds. Thus, the 1.5 Mbps range is overly conservative. That could be accurate for a backhaul starved Sprint EV-DO carrier, but if Bell and Telus have the backhaul to run HSPA+ 21, then they would have had the backhaul to run EV-DO Rev A full bore. At least 2 Mbps is a more reasonable estimate, compared to the cited 6 Mbps for HSPA+. And at that point, especially considering the time period, it would have been an e-penis measuring contest.

 

The advantage of EV-DO is that it would have given Bell and Telus greater flexibility to refarm for LTE within existing spectrum, all the while maintaining voice on CDMA1X, as has worked well for both VZW and Sprint.

 

That would also have stopped them from gathering roaming in the Olympics. I'm not defending the Canadian wireless industry in any shape or form. I'm merely pointing out they had a completely different set of circumstances than the CDMA2000 operators to the south.

 

I would love to see estimates of the roaming revenue that Bell and Telus expected to garner from the Olympics versus what they actually did. I suspect that Rogers (and Fido) still grabbed the lion's share. And I do not think too highly of operators that focus on international roaming. Worry about your subs at home, not the people or the operators overseas. In that regard, I am a bit of a wireless isolationist. And I am sick and tired of the Eurasians trying to dictate wireless standards to the US -- still the largest economy on the planet.

 

But I think that we are overlooking perhaps the biggest factor in Bell and Telus electing to overlay W-CDMA. The iPhone. Both started offering the iPhone 3GS in 2009 -- long before VZW and Sprint got a CDMA2000 iPhone. Damn you, Apple, damn you. Apple did so much damage to the North American mobile industry with its iPhone policies that I am almost ashamed to be typing this post on a MacBook Air. And Apple's asshattery will likely influence my buying decision when I follow my two year laptop upgrade cycle later this year.

 

AJ

  • Like 3
Link to comment
Share on other sites

But I think that we are overlooking perhaps the biggest factor in Bell and Telus electing to overlay W-CDMA. The iPhone. Both started offering the iPhone 3GS in 2009 -- long before VZW and Sprint got a CDMA2000 iPhone. Damn you, Apple, damn you. Apple did so much damage to the North American mobile industry with its iPhone policies that I am almost ashamed to be typing this post on a MacBook Air. And Apple's asshattery will likely influence my buying decision when I follow my two year laptop upgrade cycle later this year.

 

AJ

 

Verizon turning down the iPhone meant that the CDMA model was going to have to wait a while. So, if we're going to damn Apple, let's also throw Verizon in the mix. Not to mention that out of everyone who used CDMA, it was pretty clear Verizon was the biggest winner who got the best amount of benefit out of it. Look where that's got us.

 

I'm going Android myself at the next phone upgrade I make, but I don't think Apple itself is to blame for the industry being the way it is. If Apple had never built a phone we'd just see a meaner Verizon kicking AT&T around, while AT&T would have had to go look for new leadership at some point as their strategy broke down.

Link to comment
Share on other sites

I'm shocked this caused this much of a flame war. I didn't want it to get that way... I merely thought the dismissal of the idea through Sprint's spectrum scarcity in PCS bands would have calmed the fire.

 

As far as why the GSM track won, ask why VHS won over Betamax. It was all economies of scale. Then you had Qualcomm abandon UMB after they realized Sprint was going WiMax and Verizon was going LTE. So that killed CDMADG activities with Verizon joining the 3GPP.

Link to comment
Share on other sites

I was going to say that Bell and Telus switched to W-CDMA because they wanted to offer the iPhone 3GS. Plus there is the element of "Keeping up with the Jones'".

 

 

Sent from Josh's iPhone 5 using Tapatalk 2

Link to comment
Share on other sites

I'm shocked this caused this much of a flame war.

 

I would not call this a "flame war," just a fierce debate. And I, for one, am enjoying the discourse.

 

AJ

  • Like 2
Link to comment
Share on other sites

That's a fair point. CDMA2000 still has spectral efficiency going for it. That's big for Sprint to keep. That's the original reason I would favor CDMA2000 in certain uses like voice.

 

I'm not a fundamentalist for either format. I've used GSM, WCDMA, and CDMA2000 phones. Where I live, CDMA2000 is way better. That's just the reality of Verizon being the largest carrier in my town as far as coverage goes.

Link to comment
Share on other sites

 

I thought VHS won because the Porn Industry wasn't allowed to use Betamax? I wonder had Verizon not been bought by Vodaphone, would they have considered UMB, or was LTE always in the cards? I also presume the 2.5/2.6 Ghz bands cannot be used for UMB, which is why they chose WiMax, that an they had to deploy something for the licenses.

 

I tend to think the porn aspect of VHS vs. Beta was overplayed. The real problem there was Sony vs. everyone else, and everyone else won.

 

As far as VZW goes, Dick Lynch, their CTO at the time, angled toward LTE to get global roaming potential. Verizon's LTE dominance here is why they have the lowest churn, along with their rock solid CDMA2000 voice network. I abhor VZW's practices toward consumers and their Bell duopoly status, but I must admit they're the gold standard as far as networks go in this country for now.

Link to comment
Share on other sites

There's no denying that swappable SIMs is nice. Nor is there denying that H+/DC-H+ are fast networks. But remember that T-Mobile had to obtain AWS spectrum before it could roll out WCDMA, because their PCS was full of GSM. In areas where Sprint only has 10MHz of PCS A-F, it would have been impossible to run WCDMA and CDMA side by side, necessitating purchases of new spectrum, band class changes, and the like. Not good.

 

Fast forward to now. LTE has a theoretical capacity of around 37 Mbps per 5 MHz (I've seen 35 Mbps in real world environments) on the downlink and 14 Mbps per 5 Mhz (I've seen 13.5) on uploads. By contrast, HSPA+ has 21/5.76 Mbps of capacity in the same spectrum...and I've only ever seen about 14.4 Mbps on a single H+ carrier (28.8 Mbps on DC-H+). Granted, EvDO rA's theoretical capacity in the same 5 MHz is 12.4/7.2 (realistically you're getting 2.7 Mbps down, 1.2 Mbps up per channel, for 10.8 / 4.8). But at least there you don't have to wall off an entire 5x5 swath of spectrum for a single technology...a 5x5 that could be pushing a LOT more bits with LTE.

 

Would I like to use my Nexus 4 on Sprint? Sure. Would swappable SIMs be great on Sprint? Yes they would...and that's coming. But adding WCDMA at this point would mean that Sprint would have one less "slot" for 5x5 LTE in PCS, and that's just not worth it.

 

As for why T-Mobile went for WCDMA over EvDO, that's just what you do if you're a 3GPP based operator and the tech is out. Remember that HSPA only becomes more spectrally efficient than EvDO at 14.4 and above, and EvDO rA was for the longest time competing with HSPA 3.6. By the time T-Mobile had H+ deployed, Sprint had its ill-fated WiMAX network, which in ideal conditions could give HSPA+ a run for its money. Was the WiMAX venture a misstep? You could say that. But if you're going to razz on someone for not going to 3GPP standards it would be Verizon; they were releasing EvDO-only phones right up until their LTE network launched...and got away with it because it's very, very hard to find an area that has their network without 3G. With AT&T it's still possible, and with T-Mobile it's dead easy.

  • Like 3
Link to comment
Share on other sites

Well I am gonna swing by the Sprint store and see how things are coming, the last time I used them for my phone was around the time things really hit the skids (around the iPhone 4S launch), but most of my friends use Sprint and one just got a LTE phone and seems to be a lot happier with Sprint now. It will be a tough call on handsets, I haven't used a non-Nexus for my last 3 smartphones...

 

Software-wise, you'll probably miss the cleanliness of base Android when switching from a Nexus 4 to something else. But hardware-wise, there's a reason Google is coming out with a new phone at I/O in a week or two (so I've gathered): the current crop of Samsung, HTC and LG handsets beat the Nexus 4 hands-down. And I have the Nexus 4.

Link to comment
Share on other sites

The only thing I really wish sprint would do is allow the use of SIM cards. Keep cdma 1x/3g. I'm fine with that. I don't talk on my phone much anyways. More text and internet. I just hate having to call sprint to swap phones! Doing it online has never worked for me. That's the only thing I actually miss about tmobile. Phone troubles, swap phones easily. SIM burns out. Go get a new SIM. SIM burns out on a sprint phone, get a new phone. That's one annoyance I think sprint would be wise to change. I know the argument about phones being stolen and sold, but if the phones still had some sort of esn type registration along with the SIM, that arument would be moot. Even if I couldn't import a phone or use a tmobile phone, I'd still be happy. The technology currently being used doesn't really matter to me and the majority of people, as long as it works, that's all that matters. And like everyone has been saying, anything but lte is a dying tech that will eventually not be used anymore. So no sense in sprint spending the money on legacy tech(outside of doing NV that is). I'd rather see them pump that money into lte and spectrum and a wider variety of phones.

  • Like 2
Link to comment
Share on other sites

I was going to say that Bell and Telus switched to W-CDMA because they wanted to offer the iPhone 3GS. Plus there is the element of "Keeping up with the Jones'".

 

The other factor is that Canada has a population of 35 million people; even as the legacy carrier in most of their markets, Bell+Telus have something like 15 million customers between them. By contrast, the US's 4th largest carrier, T-Metro has 43 million customers. I think they recognized they simply didn't have the scale to attract custom handsets from leading manufacturers anymore if they stuck to CDMA2000. And, since they had the bandwidth to burn, they were able to transition to 3GPP, unlike the US regional carriers who find themselves in the same position like C Spire, Alltel (before VZW buyout), US Cellular, nTelos, and Cincy Bell (who are stuck often getting lower-end knockoffs of Sprint and Verizon stuff, months later).

Link to comment
Share on other sites

I thought VHS won because the Porn Industry wasn't allowed to use Betamax? I wonder had Verizon not been bought by Vodaphone, would they have considered UMB, or was LTE always in the cards? I also presume the 2.5/2.6 Ghz bands cannot be used for UMB, which is why they chose WiMax, that an they had to deploy something for the licenses.

 

I thought WiMax in EBS/BRS was the fact that only WiMax supported the TD band pattern. LTE wasn't "approved" for TD usage, so Sprint/Clear HAD to go WiMax to meet the FCC build out requirements.

Link to comment
Share on other sites

I thought WiMax in EBS/BRS was the fact that only WiMax supported the TD band pattern. LTE wasn't "approved" for TD usage, so Sprint/Clear HAD to go WiMax to meet the FCC build out requirements.

 

This. LTE TDD was simply not ready by the time that Sprint was required to have service up on BRS+EBS. WiMAX was a suitable TDD 3G system to brand as 4G for Sprint's purposes.

 

Fast forward to now. LTE has a theoretical capacity of around 37 Mbps per 5 MHz (I've seen 35 Mbps in real world environments) on the downlink and 14 Mbps per 5 Mhz (I've seen 13.5) on uploads. By contrast, HSPA+ has 21/5.76 Mbps of capacity in the same spectrum...and I've only ever seen about 14.4 Mbps on a single H+ carrier (28.8 Mbps on DC-H+). Granted, EvDO rA's theoretical capacity in the same 5 MHz is 12.4/7.2 (realistically you're getting 2.7 Mbps down, 1.2 Mbps up per channel, for 10.8 / 4.8). But at least there you don't have to wall off an entire 5x5 swath of spectrum for a single technology...a 5x5 that could be pushing a LOT more bits with LTE.

 

Technically, the T-Mobile HSPA+42 network is Category 24 DL and Category 7 UL. The HSPA+21 network is is Category 14 DL and Category 7 UL. The Category 7 UL means theoretical uplink data rates of 11Mbps. While most phones currently only support Category 6 UL, Category 7 UL is coming soon to most phones. I've used a data card that supported Cat 24 / Cat 7 on T-Mobile's network. The transfer rates are roughly equivalent of what I normally experience on LTE (I'm in 5MHz FDD LTE market) and the latencies are slightly better on average (excepting this last week where things have gone kind of weird).

Link to comment
Share on other sites

This. LTE TDD was simply not ready by the time that Technically, the T-Mobile HSPA+42 network is Category 24 DL and Category 7 UL. The HSPA+21 network is is Category 14 DL and Category 7 UL. The Category 7 UL means that theoretical uplink data rates of 11Mbps. While most phones currently only support Category 6 UL, Category 7 UL is coming soon to most phones. I've used a data card that supported Cat 24 / Cat 7 on T-Mobile's network. The transfer rates are roughly equivalent of what I normally experience on LTE (I'm in 5MHz FDD LTE market) and the latencies are slightly better on average (excepting this last week where things have gone kind of weird).

 

Which data card? I have the Rocket 3 and have never seen more than 3.5 Mbps up.

Link to comment
Share on other sites

Which data card? I have the Rocket 3 and have never seen more than 3.5 Mbps up.

 

It wasn't a T-Mobile branded data card. It was a white label data card (no branding on it, but curiously pentaband, nevertheless) that a friend of mine brought from Eastern Europe to try. I don't have it anymore because he went back for the month.

Link to comment
Share on other sites

So regarding why CDMA & W-CDMA got their respective market shares, is CDMA more expensive equipment wise? It seems if it gives better signal and range and requires a lot less spectrum wouldn't more countries want that? I mean had more gone that route Rev C would have made more sense so its not just a strait speed problem. What was the crippling blow to CDMA that it ended up being used less, I was reading about UMB and it seemed pretty cool. I wonder what a Sprint/Verizon network on that would have been like?

 

GSM became the standard in Europe because it was decided by the legislators over there. GSM was a Nokia technology, versus CDMA which was Qualcomm (American). That's why GSM took off.

Link to comment
Share on other sites

GSM became the standard in Europe because it was decided by the legislators over there. GSM was a Nokia technology, versus CDMA which was Qualcomm (American). That's why GSM took off.

 

GSM was also actually developed slightly before Qualcomm's cdmaOne was. The EU decided to unify on the GSM system for digital because of the massive mess of analog systems deployed throughout Europe. As for 3G onward, UMTS uses W-CDMA, which incorporates quite a bit of Qualcomm's IP. However, unlike the cdmaOne and CDMA2000 systems, Qualcomm was forced to sign FRAND agreements to have the technology incorporated. That's why the ecosystem is more "democratic" with more players and more interesting work going on.

 

So regarding why CDMA & W-CDMA got their respective market shares, is CDMA more expensive equipment wise? It seems if it gives better signal and range and requires a lot less spectrum wouldn't more countries want that? I mean had more gone that route Rev C would have made more sense so its not just a strait speed problem. What was the crippling blow to CDMA that it ended up being used less, I was reading about UMB and it seemed pretty cool. I wonder what a Sprint/Verizon network on that would have been like?

 

CDMA is more expensive because there are fewer players in the market for key inputs. For basebands, there is only Qualcomm and (ugh) VIA Telecom. For many years, only Alcatel-Lucent, ZTE, Huawei, and Nortel Networks offered infrastructure equipment. For the US and European markets, Alcatel-Lucent and Nortel flourished, while ZTE and Huawei mainly developed for Asian markets.

 

Since Europe and Asia used technology-specific licensing, most countries were barred from using CDMA systems on standard frequencies. Europe primarily uses CDMA on 450MHz for Wireless Local Loop or rural broadband services after NMT (Nordic Mobile Telephone, one of many analog cellular systems developed for Europe) systems were decommissioned. The equipment is not cheap because there are very few that make equipment on that band.

 

CDMA on 850MHz (3GPP2 BC0, to be clear) did spread quite a bit throughout the Americas, China, India, Japan and Korea (J-CDMA, later OMA CDMA), and Oceania. However, the economic scale of CDMA2000 was crushed over time as UMTS networks grew in number from 2004 to 2009. Latin America decided to migrate in full force a couple of years ago, and one of the last large networks was just shut off this year. Oceania did the same, and now the major CDMA networks actively sold and utilized are Europe 450MHz networks (which will upgrade to LTE in three years), and US networks.

  • Like 1
Link to comment
Share on other sites

There's no denying that swappable SIMs is nice. Nor is there denying that H+/DC-H+ are fast networks. But remember that T-Mobile had to obtain AWS spectrum before it could roll out WCDMA, because their PCS was full of GSM. In areas where Sprint only has 10MHz of PCS A-F, it would have been impossible to run WCDMA and CDMA side by side, necessitating purchases of new spectrum, band class changes, and the like. Not good.

 

Well, that's not really as much of an issue anymore. Sprint has all that ESMR for CDMA 1X-Advanced and LTE. ESMR, PCS A-F, PCS G, and soon PCS H will give Sprint plenty of room.

 

UPDATE: Actually reading a bit into Cingular/McCaw Wireless it seems that AT&T Wireless (and later AT&T Mobility under SBC) was a bit of a hodgepodge network). So I am assuming Verizon/Sprint's singular technology path and more organically grown networks rather than merger/acquisitions are why AT&T is much farther behind?

 

This is pretty much the case for everyone. T-Mobile, Sprint, and Verizon Wireless have largely had simpler M&A based expansions. AT&T's has been a mess over the years, with a multitude of acquisitions across a wide range of technologies and frequencies. AT&T's network is such a mess internally that its own engineers dislike dealing with some of the internal bridging done to make everything work. That's why its quality is lower than others, even other GSM/UMTS operators across the globe. AT&T is massively due for a network rebuild, but ironically it can't afford that. The CapEx for AT&T to rebuild its entire network would be ridiculous.

  • Like 1
Link to comment
Share on other sites

So regarding why CDMA & W-CDMA got their respective market shares, is CDMA more expensive equipment wise? It seems if it gives better signal and range and requires a lot less spectrum wouldn't more countries want that?

 

This is a rather interesting read though, admittedly I cannot vouch for the veracity of it's contents:

 

http://denbeste.nu/cd_log_entries/2002/10/GSM3G.shtml

Link to comment
Share on other sites

 

This is pretty much the case for everyone. T-Mobile, Sprint, and Verizon Wireless have largely had simpler M&A based expansions. AT&T's has been a mess over the years, with a multitude of acquisitions across a wide range of technologies and frequencies. AT&T's network is such a mess internally that its own engineers dislike dealing with some of the internal bridging done to make everything work. That's why its quality is lower than others, even other GSM/UMTS operators across the globe. AT&T is massively due for a network rebuild, but ironically it can't afford that. The CapEx for AT&T to rebuild its entire network would be ridiculous.

 

Ironically, in the networks they rebuilt from scratch, their quality is good, for example the Zodiac markets. They had all new Ericsson gear there. Things actually work pretty well.

 

As far as a complete rebuild...you bet AT&T could do it. They're just too cheap. I'd do the rebuild with Ericsson, NSN, and Samsung, the problem is AT&T's senior management wouldn't approve it. They think of that dividend as the mother of all sacred cows, and anything that could possibly reduce that dividend is bad. They're in prime position to get knocked to 3rd place if there's a SoftBank coming in that owns both New Sprint and TMUS.

  • Like 1
Link to comment
Share on other sites

This is a rather interesting read though, admittedly I cannot vouch for the veracity of it's contents:

 

http://denbeste.nu/c.../10/GSM3G.shtml

 

I actually agree with him, too. Back in the 90s, I was in the CDMA camp (which A.J. may not believe). The cdmaOne and CDMA2000 systems were by far technically superior to the GSM camp. No one likes Ericsson front-line people that much. They're like Cisco in that sense. However, I think that the 3GPP has caught up and surpassed them now.

 

And yes, I'm also aware of the dark dealings behind the cover that led to the GSM domination. And apparently, his vicious wishes have come true about the EU, too.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • large.unreadcontent.png.6ef00db54e758d06

  • gallery_1_23_9202.png

  • Posts

  • Recently Browsing

    • No registered users viewing this page.
×
×
  • Create New...