T-Mobile LTE & Network Discussion

[maxsilver]

Let me reiterate my disbelief that T-Mobile had 51,000 macro sites before the merger. They were always 1-2,000 sites behind Sprint. I am not convinced. I will find the truth even if it kills me or I have to use the wayback machine.

 

That's ok. Don't let facts get in the way of your opinion.   

 

The way that the confusion may arise is that T-Mobile might have 37,000 HSPA+ sites and about 14,000 GSM sites, but they are colocated.

 

The 'colocation' argument has already been debunked, by the T-Mobile press release. T-Mobile doesn't double count HSPA + GSM sites (just as Sprint doesn't double count 1X + EVDO sites).

 

 

It's not just me that has doubts. AJ, our local guru, is also incredulous and we have been around for awhile.

 

I'm sure you guys have been around a while, that's all well and good, but even if the number turns out to be artificially high, this argument holds absolutely no value, and is a logical fallacy.

 

 

- -

 

In all honesty, my guess is that the T-Mobile 51,000 number includes DAS's as a single site. (For instance, if they throw a DAS all throughout a university, that counts as a "site"). T-Mobile is on a *lot* of DAS's, far more than Sprint is (in my market, anyway). 

 

T-Mobile's number isn't unusually high in any way. Sprint's number seems oddly low. Sites are, by far, the most important metric in network quality. This is where I would start investigating.

[WiWavelength]

 

That's ok. Don't let facts get in the way of your opinion. 

 

Listen, pal, you need to get off your high horse.  Repeating press releases is not the same as verifying facts.  The latter is what we are actually trying to do.  That T-Mobile has 51,000 discrete cell sites is not yet a proven fact.

 

AJ

[maxsilver]

http://www.tmonews.com/2013/06/t-mobile-sending-out-event-invites-for-july-10th-rolling-out-their-boldest-moves-yet

 

Perhaps this banging' announcement of "boldest moves yet" includes some kind of swan song about upgrading those 14,000 odd GSM only sites.  Nice to dream, isn't it?

I'd love to see that happen, but I'm not holding my breath. Seems *highly* unlikely to me.

 

 

Keeping 14k 2G sites in operation with no ongoing plans to upgrade doesn't seem like the strategy of a carrier that is in it for the "long haul"... it sounds like a carrier looking for consolidation.  I've always thought that about tmo's rural coverage

I don't think those towers are there to attract customers. They are there to prevent roaming. It's cheaper for T-Mobile to keep those towers running, rather than pay AT&T crazy-high roaming rates for all their subscribers usage in those areas.

[bigsnake49]

 

That's ok. Don't let facts get in the way of your opinion.   

 

 

The 'colocation' argument has already been debunked, by the T-Mobile press release. T-Mobile doesn't double count HSPA + GSM sites (just as Sprint doesn't double count 1X + EVDO sites).

 

 

 

I'm sure you guys have been around a while, that's all well and good, but even if the number turns out to be artificially high, this argument holds absolutely no value, and is a logical fallacy.

 

 

- -

 

In all honesty, my guess is that the T-Mobile 51,000 number includes DAS's as a single site. (For instance, if they throw a DAS all throughout a university, that counts as a "site"). T-Mobile is on a *lot* of DAS's, far more than Sprint is (in my market, anyway). 

 

T-Mobile's number isn't unusually high in any way. Sprint's number seems oddly low. Sites are, by far, the most important metric in network quality. This is where I would start investigating.

 

Neither Sprint nor T-Mobile have been historically known as DAS rich carriers. Metro on the other hand...

 

My disbelief is based on the fact that T-Mobile was historical known a 35-37,000 site carrier. All of a sudden this 51,000 site number appears without an announced large scale new site buildout program. So color me a sceptic.

[maxsilver]

Listen, pal, you need to get off your high horse.  Repeating press releases is not the same as verifying facts.  The latter is what we are actually trying to do.  That T-Mobile has 51,000 discrete cell sites is not yet a proven fact.

Woah buddy, no need to get all defensive. (Did you see the smiley? It's an international symbol for a statement made in jest).

 

I get that you don't like the press release. I understand that it challenges some closely-held opinions. I'm not trying to persuade you of anything. (I'm not in here claiming T-Mobile's coverage or service is better, or anything like that).

 

But it does insult your own experience, to challenge an official statement with your own personal ancetodes. Surely, someone with your knowledge and experience will appreciate this sentiment.

 

If someone came in here and said "I used a Sprint phone and it sucked, therefore all of Sprint sucks," everyone here would (completely understandably) challenge that statement with facts. But for some reason, if someone says the same of T-Mobile, it passes as a valid argument.

 

All I'm trying to do is promote consistency. We used Sprint documents (internal and external) as reference. I see no reason why T-Mobile's documents can't hold equal weight about their network.

[maximus1987/lou99]

Guys, end of 2014 we'll have final confirmation: if TMO has any native 2G left, then 37,000 is just 3G/4g but if it has no 2G left, then 37k is entire footprint.

 

This is because TMO has said they're gonna finish upgrading 37k by end of 2014.

 

" T-Mobile will be improving approximately 37,000 cell sites over the next 18 months. These upgrades include:

 

Replacing copper lines with fiber optic lines

Adding new radios

Moving ground equipment to the top of towers

Adding new antennas"

 

https://t-mobile.jive-mobile.com/#jive-document?content=%2Fapi%2Fcore%2Fv2%2Fdocuments%2F5736

[maxsilver]

Neither Sprint nor T-Mobile have been historically known as DAS rich carriers. Metro on the other hand...

Your right. Neither are on as many DAS's as they should be. (I personally believe they should be on *every* DAS, but I understand there are financial limitations involved).

 

 

All of a sudden this 51,000 site number appears without an announced large scale new site buildout program. So color me a sceptic.

I don't think that's true. I don't think the 51,000 number "appeared out of nowhere", I think T-Mobile just doesn't like to mention it because those towers are "unloved". (Not maintained, not improved, not upgraded.)

 

I mean, if I worked in T-Mobile HR, I'd be pretty embarrassed about the EDGE portions of the network. I'd never mention them.

 

 

If it helps, here's yet another source : Dave Mayo (Senior VP at T-Mobile), who's quoted in Light Reading, from just last week, saying "We're upgrading about 37,000 of our 52,000 cell sites". http://www.lightreading.com/wireless-backhaul/tmobile-to-debut-ltea-features-in-2013/240157256

[maximus1987/lou99]

 

 

 

 

 

That's ok. Don't let facts get in the way of your opinion.

 

 

 

 

The 'colocation' argument has already been debunked, by the T-Mobile press release. T-Mobile doesn't double count HSPA + GSM sites (just as Sprint doesn't double count 1X + EVDO sites).

 

 

 

 

 

I'm sure you guys have been around a while, that's all well and good, but even if the number turns out to be artificially high, this argument holds absolutely no value, and is a logical fallacy.

 

 

 

 

 

- -

 

 

 

In all honesty, my guess is that the T-Mobile 51,000 number includes DAS's as a single site. (For instance, if they throw a DAS all throughout a university, that counts as a "site"). T-Mobile is on a *lot* of DAS's, far more than Sprint is (in my market, anyway).

 

 

 

T-Mobile's number isn't unusually high in any way. Sprint's number seems oddly low. Sites are, by far, the most important metric in network quality. This is where I would start investigating.

 

 

Neither Sprint nor T-Mobile have been historically known as DAS rich carriers. Metro on the other hand...

 

My disbelief is based on the fact that T-Mobile was historical known a 35-37,000 site carrier. All of a sudden this 51,000 site number appears without an announced large scale new site buildout program. So color me a sceptic.

We found a Tmobile press release with that 51,000 number from 2012:

 

http://newsroom.t-mobile.com/phoenix.zhtml?c=251624&p=irol-newsArticle&ID=1804179&highlight=

[WiWavelength]

I get that you don't like the press release. I understand that it challenges some closely-held opinions. I'm not trying to persuade you of anything. (I'm not in here claiming T-Mobile's coverage or service is better, or anything like that).

 

But it does insult your own experience, to challenge an official statement with your own personal ancetodes. Surely, someone with your knowledge and experience will appreciate this sentiment.

You seemingly consider anything that fits your world view a "fact." Anything that runs counter to that view is an "opinion." This needs to stop. Everything is open to question.

 

FYI, we have openly questioned Sprint's PR department about its launch of LTE markets and depiction of LTE coverage. We have more solid, accurate info than Sprint PR does. So, you might want to take note of that when you put so much credence in the info from T-Mobile PR.

 

All I'm trying to do is promote consistency. We used Sprint documents (internal and external) as reference. I see no reason why T-Mobile's documents can't hold equal weight about their network.

 

No, we used internal Sprint engineering documents to construct a full accounting of native Sprint cell sites.  Your T-Mobile press releases do not present the same level of info.  Do not equate the two.

 

Look, reportedly, you are quite the T-Mobile shill at other sites.  If you are here largely to troll for T-Mobile, you will not last.  Consider yourself warned.

 

AJ

[bigsnake49]

OK, here's a blurb from the 2008 fourth quarter and year end results:

 

 

•  

  T-Mobile USA continued to improve network coverage in the fourth quarter of 2008, adding approximately 1,100 new GSM/GPRS/EDGE cell sites, bringing the total number of cell sites at the end of the quarter to 44,000.

   

  T-Mobile USA continues to invest in the UMTS / HSDPA (3G) network, which now reaches 107 million people in 130 cities. 

   

  http://cache.t-mobile.com/Cms/Files/Published/0000BDF20016F5DD010312E2BDE4AE9B/0000BDF20016F5DE011FB943D848D0CB/file/TMUSQ4PressReleaseFINAL.pdf

   

  But that brings up more questions, as in why would they be adding 1,100 GSM/GPRS/Edge only sites at the same time they're expanding their 3G network to cover 100 Million people during 2008?

[maxsilver]

So, you might want to take note of that when you put so much credence in the info from T-Mobile PR.

Right. And if it was just T-Mobile PR, you'd have an argument here.

 

But it's not just T-Mobile PR. It's also numerous T-Mobile executives (Neville Ray, Dave Mayo), and they are all saying the same thing.

 

You seemingly consider anything that fits your world view a "fact." Anything that runs counter to that view is an "opinion." This needs to stop.

You already know this isn't true.

 

I'm quoting documents and statements from staff. I don't work for T-Mobile. I don't pretend to have inside information. I'm just repeating the information available to me (and directly linking to it for others to examine as well).

 

Look, reportedly, you are quite the T-Mobile shill at other sites.  If you are here largely to troll for T-Mobile, you will not last.  Consider yourself warned.

This isn't an argument. This is an attack on my character. Your just calling me names now.

[maximus1987/lou99]

I just don't see how TMO can wait until 600 MHz can be flipped on to cover rural America. It's sad and all that TMO doesn't even some "sow's ear" spectrum to run even 3x3 LTE but that's not the customers' problem to deal with.

 

ATT will destroy them with AIO.

It can even debut a service that's capped at 1mbps - like TMO's GoSmart $45/month plan that has "3G" speeds - and use its scale to CRUSH them; that's what I'd do if I were ATT.

 

Everyone here I think is understanding that TMO can't afford to put up MORE PCS towers but it should at least upgrade its current ones.

 

I think a mod here said late 2016/early 2017 was the earliest it could be "flipped on" even if the towers are ready.

 

AJ, can you comment?

[maxsilver]

But that brings up more questions, as in why would they be adding 1,100 GSM/GPRS/Edge only sites at the same time they're expanding their 3G network to cover 100 Million people during 2008?

They've added EDGE (or "edge-like" 3G) sites here over the last year. Not a lot, but about 12 over the last 18 months.

 

I'm guessing they do it to reduce AT&T roaming costs. (I don't have evidence of this, but I have observed that they block AT&T roaming for a few months before the new EDGE sites near here came online).

[bigsnake49]

SELECTED DATA FOR T-MOBILE USA

YE 07 Q4 07 Q3 07 Q2 07 Q1 07 YE 06 Q4 06

Covered population8

284,000

284,000

283,000

282,000

280,000

277,000

277,000

Customers, end of period2

28,685

28,685

27,734

26,877

26,020

25,041

25,041

Thereof contract customers

23,914

23,914

23,181

22,624

21,937

21,211

21,211

Thereof prepaid customers

4,771

4,771

4,553

4,253

4,083

3,829

3,829

Net customer additions

3,644

951

857

857

980

3,351

901

Minutes of use/contract customer/month

1,125

1,123

1,130

1,150

1,090

1,030

1,020

Contract churn

1.90%

1.80%

2.00%

1.80%

1.90%

2.20%

2.10%

Blended churn

2.80%

2.80%

2.90%

2.70%

2.60%

2.90%

2.90%

($)

($ million)

ARPU (blended) 1, 9

52

52

53

53

52

52

52

ARPU (contract)

57

56

57

57

56

55

56

ARPU (prepaid)

19

20

18

19

19

22

21

Cost of serving (CCPU)3

25

25

26

25

25

25

25

Cost per gross add (CPGA)4

300

300

280

300

310

300

300

Total revenues

19,288

5,068

4,894

4,780

4,546

17,138

4,523

Service revenues1

16,892

4,371

4,332

4,195

3,994

14,511

3,813

OIBDA5

5,350

1,327

1,412

1,386

1,225

4,712

1,172

OIBDA margin 6

31%

30%

32%

32%

30%

31%

30%

Capital expenditures7

2,677

1,009

500

546

622

2,608

675

Cell sites on-air10

37,900

37,900

37,000

36,400

35,800

35,400

35,400

 

http://www.t-mobile.com/Cms/Files/Published/0000BDF20016F5DD010312E2BDE4AE9B/0000BDF20016F5DE01187B5EBB2BA144/file/TMUS%20Q4%20Press%20Release_FINAL.pdf

 

That's from the year end 2007. As you can see they only had 37,900, and from the 2008 info above, they only added 1,100, so where did the other 5000 sites come from?

 

Do you see why we are sceptical?

 

PS. Sorry about the formatting. It looked OK when I copied it. Go to the reference to see the formatted data.

Edited July 2, 2013 by bigsnake49

[bigsnake49]

They've added EDGE (or "edge-like" 3G) sites here over the last year. Not a lot, but about 12 over the last 18 months.

 

I'm guessing they do it to reduce AT&T roaming costs. (I don't have evidence of this, but I have observed that they block AT&T roaming for a few months before the new EDGE sites near here came online).

 Why would they not add full sites including 3G?

[maxsilver]

That's from the year end 2007. As you can see they only had 37,900, and from the 2008, they only added 1,100, so where did the other 5000 sites come from?

 

Do you see why we are sceptical?

Yes, I see why folks are skeptical.

 

But that chart ends at 2007.

 

Here's that *same exact report* from 2008 :

 

Note that now the count for "Quarter 4, 2008 cell sites on air" is now up to 44,000. In 2008.

 

http://s.tmocache.com/Cms/Files/Published/0000BDF20016F5DD010312E2BDE4AE9B/0000BDF20016F5DE011FB943D848D0CB/file/TMUSQ4PressReleaseFINAL.pdf

[maximus1987/lou99]

 

 

They've added EDGE (or "edge-like" 3G) sites here over the last year. Not a lot, but about 12 over the last 18 months.

 

 

I'm guessing they do it to reduce AT&T roaming costs. (I don't have evidence of this, but I have observed that they block AT&T roaming for a few months before the new EDGE sites near here came online).

 

 

Why would they not add full sites including 3G?

Same reason why they don't just upgrade rural right now: cost.

[maxsilver]

Also worth noting, both the 2007 report bigsnake49 found, and the 2008 report I added, have a footnote #10 that mentions :

"Cell sites are defined as the total number of sites in service at the end of the period, excluding small low power, low gain access sites."

Which I would interpret to mean "this number does not include DAS's".

If that interpretation is true, the 52,000 number does not include DAS's, and is equally comparable to Sprint's site number.

 

EDIT : this interpretation is probably incorrect.

[bigsnake49]

Yes, I see why folks are skeptical.

 

But that chart ends at 2007.

 

Here's that *same exact report* from 2008 :

 

Note that now the count for "Quarter 4, 2008 cell sites on air" is now up to 44,000. In 2008.

 

http://s.tmocache.com/Cms/Files/Published/0000BDF20016F5DD010312E2BDE4AE9B/0000BDF20016F5DE011FB943D848D0CB/file/TMUSQ4PressReleaseFINAL.pdf

 

 

Yes, I see why folks are skeptical.

 

But that chart ends at 2007.

 

Here's that *same exact report* from 2008 :

 

Note that now the count for "Quarter 4, 2008 cell sites on air" is now up to 44,000. In 2008.

 

http://s.tmocache.com/Cms/Files/Published/0000BDF20016F5DD010312E2BDE4AE9B/0000BDF20016F5DE011FB943D848D0CB/file/TMUSQ4PressReleaseFINAL.pdf

 

So in one year they constructed 6000 new sites. Brand new sites? OK . All while expanding their 3G?

[bigsnake49]

Also worth noting, both the 2007 report bigsnake49 found, and the 2008 report I added, have a footnote #10 that mentions :

 

"Cell sites are defined as the total number of sites in service at the end of the period, excluding small low power, low gain access sites."

 

Which I would interpret to mean "this number does not include DAS's".

 

If that interpretation is true, the 52,000 number does not include DAS's, and is equally comparable to Sprint's site number.

 

No, Macro DAS are full power. Micro DAS, such as repeaters in office buildings and package stores are low power, low gain.

[maxsilver]

No, Macro DAS are full power. Micro DAS, such as repeaters in office buildings and package stores are low power, low gain.

Oops, I am wrong, and you are absolutely correct.

 

I'm flipping through too many PDF's right now. Gotta slow down before posting. 

[ChadBroChillz]

I feel like having 52000 sites with the same coverage as Sprint is very inefficient. I believe that number is similar to what has been reported for ATT/Verizon. I would think they would want to trim some of the excess, especially if they have plans of any sort of expansion of their LTE/FauxG network once they get some 600Mhz.

[WiWavelength]

Right. And if it was just T-Mobile PR, you'd have an argument here.

 

But it's not just T-Mobile PR. It's also numerous T-Mobile executives (Neville Ray, Dave Mayo), and they are all saying the same thing.

Again, if there is any double counting or "fuzzy math" going on, all the king's horses and all the king's men can be saying the same thing. But that does not make it the "fact" that we are interested in knowing. I honestly cannot understand how you fail to get this. You just want to repeat the same sources over and over again but not question the methodology of those sources.

 

You already know this isn't true.

Actually, I would say that I know it is true. You are fond of calling the 51,000 site figure a "fact," while any counter is just an "opinion."

 

I'm quoting documents and statements from staff. I don't work for T-Mobile. I don't pretend to have inside information. I'm just repeating the information available to me (and directly linking to it for others to examine as well).

See above. Quoting publicly released documents does not give us the verification that we seek, especially when the data in those documents does not exactly jive with circumstantial evidence.

 

This isn't an argument. This is an attack on my character. Your just calling me names now.

No, it is not an argument.  It is an admonition.  Trolling for other carriers will not be tolerated at S4GRU.

 

Regardless, the report does not come from me, but it comes from other S4GRU staff. I have no idea who you are. I have never heard of you before. And you have probably never heard of me. But I have been researching and writing about the wireless industry for the last dozen years. Additionally, I have a responsibility as a staff member to help keep order, and you seem intent on stirring the pot about an entirely non Sprint related issue.

 

AJ

[bigsnake49]

From those reports they went from 35,400 in the beginning of 2007 to 44,000 at the end of 2008. That's a hell of a lot of brand new sites in two years. If they did it, more power to them.

Edited July 2, 2013 by bigsnake49

[WiWavelength]

From those reports they went from 35,400 in the beginning of 2007 to 44,000 at the end of 2008. That's a hell of a lot of brand new sites in two years. 

 

It also coincides with T-Mobile's massive AWS W-CDMA overlay push following the 2006 AWS-1 FCC auction.  Hmm, that is worthy of investigation.

 

AJ