T-Mobile LTE & Network Discussion

[iansltx]

Why not just do microwaves? Can at least bring pretty decent speeds...

 

'cuz that's not how Poser Mo...T-Mobile...rolls.

 

Also, you have to haul microwave from somewhere that's fiber-connected. If you don't have line of sight to a fiber-powered tower a few miles away then you're back at square one.

 

Sprint is willing to do microwave backhaul. But MW BH is one reason why some Sprint TD-LTE (can barely contain my excitement) sites in DEN are running at 8 Mbps rather than 40.

[iansltx]

Well I would have to guess that's likely why they waited to deploy LTE due to it not being that much faster and not increasing voice capacity.  

 

Yep. But 10+ Mbps of upload capacity on 5 MHz, which can scale linearly to 10MHz and 20MHz, is nearly double what you can get on H+. And 35-37 Mbps real-world capacity for small cell sizes is a big boost from 21 Mbps...which ends up being around 14-16 Mbps in real life...in the same spectrum allotment.

 

The difference between T-Mobile and any other large carrier in the US is that, to deploy LTE, they have to refarm existing, not-all-that-much-less-efficient, spectrum from the get-go. Sprint is starting on the (vacant) G block, Verizon on the (vacant) 700 band, followed by the (vacant for them) AWS band. AT&T? Same thing, but with a different 700 band. But for T-Mobile, adding 37 Mbps of LTE capacity means, in many cases, removing 21 Mbps of H+. And equipment on both sides of the link is less expensive on H+ than LTE, though now the gap has closed enough that TMo doesn't mind making the jump. Particularly if it means that their early adopter users, with an ARPU of $80+ including EIP, have a network to themselves that they can rant and rave about.

[reedacus25]

Yep. But 10+ Mbps of upload capacity on 5 MHz, which can scale linearly to 10MHz and 20MHz, is nearly double what you can get on H+. And 35-37 Mbps real-world capacity for small cell sizes is a big boost from 21 Mbps...which ends up being around 14-16 Mbps in real life...in the same spectrum allotment.

 

The difference between T-Mobile and any other large carrier in the US is that, to deploy LTE, they have to refarm existing, not-all-that-much-less-efficient, spectrum from the get-go. Sprint is starting on the (vacant) G block, Verizon on the (vacant) 700 band, followed by the (vacant for them) AWS band. AT&T? Same thing, but with a different 700 band. But for T-Mobile, adding 37 Mbps of LTE capacity means, in many cases, removing 21 Mbps of H+. And equipment on both sides of the link is less expensive on H+ than LTE, though now the gap has closed enough that TMo doesn't mind making the jump. Particularly if it means that their early adopter users, with an ARPU of $80+ including EIP, have a network to themselves that they can rant and rave about.

An unintentional compliment, but very true. T-Mobile is making the most of its spectrum constraint and I feel they will do very well despite it. Leaps spectrum would have helped, but that's another story. T-Mobile had a fast network already, but LTE is obviously faster for the spectrum.

 

The main issue is with keeping voice saturation at bay, and this is being combatted with PCS HSPA+ carriers (of which they could do Dual Carrier with if they chose and had the spectrum to do so).

 

The biggest "speed" factor that is improved with LTE over HSPA is the latency, which is improved over the already low latency over HSPA provided by the modern equipment. Sheer numbers are a useless metric for 98% of real world usage, but the latency is critical 99% of the time.

 

So until VoLTE matures. T-Mobile must make a difficult decision on how much spectrum to allocate for HSPA in the AWS band for voice+data, while maximizing usable spectrum for LTE.

[bucdenny]

John talks mad crap! Hurry it up already Sprint, show them some 100Mbps TD-LTE so this guy can stfu! http://www.tmonews.com/2013/07/t-mobile-ceo-john-legere-taunts-the-competition-with-lte-speed-tests/

[Rawvega]

The guy has no discretion or class. That said, the way to shut the clown up is to take away his ammunition and make the network perform as it should.

[lilotimz]

Funny thing is...

 

[Rawvega]

Facts be damned, eh John...

[bucdenny]

'cuz that's not how Poser Mo...T-Mobile...rolls.

 

Also, you have to haul microwave from somewhere that's fiber-connected. If you don't have line of sight to a fiber-powered tower a few miles away then you're back at square one.

 

Sprint is willing to do microwave backhaul. But MW BH is one reason why some Sprint TD-LTE (can barely contain my excitement) sites in DEN are running at 8 Mbps rather than 40.

 

It make sense to get back haul in as possible with MV.  MV should come first just to get the site up, then replace with fiber when ready instead of waiting and waiting.

[bucdenny]

Can someone contact Mr. Son and tell him to do some 100Mbps speed test already so we can shut up Mr. John 

 

Wonder if Mr. Son has a twitter account...

[milan03]

John talks mad crap! Hurry it up already Sprint, show them some 100Mbps TD-LTE so this guy can stfu! http://www.tmonews.com/2013/07/t-mobile-ceo-john-legere-taunts-the-competition-with-lte-speed-tests/

He already got smoked by my speeds so he can stfu now

[Paynefanbro]

That new T-Mobile CEO is a real pain on the backside. I'd like to see if he can hold that connection while he drives out of NYC!

[JeffDTD]

The guy has no discretion or class.

 

Agreed.  Sprint was one of the driving reasons he even has a job at the ole' deutche...   Does he really believe Sprint is afraid to compete? Or doesn't want to compete?  

 

I would LOVE to jump on some Tmo pricing + high speed LTE.  Unfortunately, I would have access to HSPA+ only, about 50% of the time, never when I travel, not at any of my family's homes across south MS.    Its sort of a mystery why they even have Tmo stores in my state.  But hey, bring on that competition.  

[maximus1987/lou99]

Funny thing is...

 

Why is this funny?

[Rawvega]

To be clear, I think it's a good thing that a CEO (or his subordinates) is so engaging with the public. And if you want to highlight why you think your product is better than the competition, that only makes sense. However, I just don't see the need to be so childish and crass about it.

 

Also, I see that he responded to bucdenny earlier this morning.

 

Sent from my SPH-L900 using Tapatalk 4 Beta

 

 

[Paynefanbro]

Why is this funny?

John said Sprint doesn't have LTE at 33rd and 3rd but in reality Sprint has a strong signal in that area.

[bucdenny]

Actually, the NY AW-CMA-001A license was won by T-Mobile in FCC Auction #66, all the way back in 2006.

 

What you don't see is that there is a lot of spectrum swapping going on behind the scenes between T-Mo and Verizon, and the FCC "Spectrum Dashboard" is an inaccurate and misleading mess because of it.

 

The A-block is/was T-Mobile's but they haven't moved off of it yet.

 

T-Mobile and VZ are in the process of a 1:1 spectrum swap, where T-Mo gives VZ the A-Block(20MHz), and VZ is giving the F-Block(20MHz) to T-Mobile.

 

This is to alleviate T-Mobile not having a contiguous block of AWS spectrum, (the T-Mo/Metro A, D,and E blocks had VZ's B-block in the middle).

 

At the end of the swap VZ owns the A and B blocks(40MHz Total contiguous), and T-Mo/MetroPCS own the C, D, E, and F, blocks(50MHz Total contiguous).

 

Plus, the A-Block was only for NY and Northern New Jersey, the F-block that T-Mo is receiving is a Regional Block covering the entire Northeastern US.

 

So, YOU don't know what you're talking about, because you have no clue as to what is going on behind the scenes, because you don't work in this industry.

 

 

Is this true? 

[bucdenny]

Mr. Son, here's the link, shut this guy up!     https://twitter.com/john_legere/status/359912782860267520

[WiWavelength]

Is this true? LOL

 

No, it is rife with inaccuracies.  Where was this comment posted?

 

Yes, T-Mobile and VZW and have swapped many AWS A block and F block licenses.  But those transactions are already complete in most, if not all affected markets.  For example, T-Mobile's lease of its former AWS A block license in New York was canceled back in February.

 

Additionally, the AWS A block license is CMA based, while the F block license is much larger REA based.  But T-Mobile has received geographically partitioned portions of VZW's AWS F block license, not the whole Northeast REA.

 

So, the commenter in question should be told that HE does not know what he is talking about.  And I do not even work in the industry, but I do know what I am talking about.

 

AJ

[maximus1987/lou99]

Mr. Son, here's the link, shut this guy up! https://twitter.com/john_legere/status/359912782860267520

Oh he will. When Sprint debuts 40 MHz TDD.

In the sponsor section, will the TDD sites be marked?

Edited July 24, 2013 by asdf190

[bucdenny]

No, it is rife with inaccuracies.  Where was this comment posted?

 

Yes, T-Mobile and VZW and have swapped many AWS A block and F block licenses.  But those transactions are already complete in most, if not all affected markets.  For example, T-Mobile's lease of its former AWS A block license in New York was canceled back in February.

 

Additionally, the AWS A block license is CMA based, while the F block license is much larger REA based.  But T-Mobile has received geographically partitioned portions of VZW's AWS F block license, not the whole Northeast REA.

 

So, the commenter in question should be told that HE does not know what he is talking about.  And I do not even work in the industry, but I do know what I am talking about.

 

AJ

 

http://www.tmonews.com/2013/07/t-mobile-ceo-john-legere-taunts-the-competition-with-lte-speed-tests

[bucdenny]

Oh he will. When Sprint debuts 40 MHz TDD. In the sponsor section, will the TDD sites be marked?

 

I hope it debuts during the earnings meeting end of this month.

[iansltx]

I hope it debuts during the earnings meeting end of this month.

 

Doubt it. 20MHz TD is the first rev, which will get real-world speeds up to about what you see in T-Mobile's 10x10 markets (like Dallas is now...25 Mbps up on a phone is impressive, though TD 20 MHz won't make it quite that high).

 

I'd expect Nx20MHz to happen as WiMAX is turned down. So give it 18 months. T-Mo might have 20x20 FD online in some markets by then, but not all.

[JohnHovah]

Mr. Son, here's the link, shut this guy up!     https://twitter.com/john_legere/status/359912782860267520

I think that may be a masayoshi poser or fan.    Son's actual Twitter handle is @masason as seen here https://twitter.com/masason.  (verified account)

 

For those who can't read Japanese (or understand the broken english from a shoddy google page translation or have a step grandmother who is actually japanese and can translate) there are some interesting tweets coming from him

[Mobilesolutions]

Why not just do microwaves? Can at least bring pretty decent speeds...

Multiple hops leading back to one fiber site will be a slow pipe for those users depending on your actual TCP/IP capacity at each microwave site. Regardless; if you have 7 hops down a rural road and a saturated network end users will not see fast speeds even if you have a 2*20.

 

I'll put the test results of my 2x20 tests here someday soon.

 

May even through in a few 2x40's

[maximus1987/lou99]

An unintentional compliment, but very true. T-Mobile is making the most of its spectrum constraint and I feel they will do very well despite it. Leaps spectrum would have helped, but that's another story. T-Mobile had a fast network already, but LTE is obviously faster for the spectrum.

 

The main issue is with keeping voice saturation at bay, and this is being combatted with PCS HSPA+ carriers (of which they could do Dual Carrier with if they chose and had the spectrum to do so).

 

The biggest "speed" factor that is improved with LTE over HSPA is the latency, which is improved over the already low latency over HSPA provided by the modern equipment. Sheer numbers are a useless metric for 98% of real world usage, but the latency is critical 99% of the time.

 

So until VoLTE matures. T-Mobile must make a difficult decision on how much spectrum to allocate for HSPA in the AWS band for voice+data, while maximizing usable spectrum for LTE.

That's what WCDMA+ is for. Same voice capacity in 1/3 the spectrum.

Look at the Qualcomm WCDMA+ presentation on their website.

I know WCDMA+ won't be available for years but there is light at the end of the tunnel.