[thunder79]

Yay...got Chicago (didn't much doubt that though)

Got a way to afford a Nexus 4 later this month too

[icecoolfsu]

This list is old. I got an email from T-Mo confirming 42mbps service in Little Rock, AR and I am indeed getting it. Got over 20 down on an Amaze 4G.

[mrmoe110]

I wouldn't believe anything T-Mobile says concerning coverage until I actually see it in action.

[CBMC]

This list is old. I got an email from T-Mo confirming 42mbps service in Little Rock, AR and I am indeed getting it. Got over 20 down on an Amaze 4G.

What email did you use top contact them? Can't find any in their page.



Sent from my Galaxy Nexus using Tapatalk 2

[ChromeJob]

Seeing a lot of folks asking, so here's the list in alphabetical order by city name.

Thanks.

Of course. "Subject to change."

I spend time in Durham, Raleigh, Sunnyvale, Mountain View, Santa Clara,... I can tell you that while I do get 4 bars of 4G in some spots, and some nice download speeds, I HAVE NOT as yet field tested anything better than 17MBps (and that was in my bedroom, probably line of site a 1/4 mile from a tower). HSPA+ 42 my eye. [This is with a 01/2013 made HTC Amaze which should be on all the necessary freqs.) And I see the phone switching 4G, 2G, UMTS/HSDPA a lot, causing a lot of battery drain I'd guess.

FYI I'm not a cellular wireless technician, don't even play one on TV. So I don't know much about how this sheot works (my last professional accreditation was on VOR, TACAN, RADALT, GPS, etc, not even close).

[sixdust]

This is the list according to T-Mobile's press page. I certainly can't go check them all :-P

Sent from my Nexus 10 using Android Central Forums

Why not? ROAD TRIP!

[laststop311]

Can anyone confirm for me if youngstown OH market has 42mbps? i know it has atleast 21 but not sure if its 42.

Also is there a list somewhere of all the areas with refarmed 1900mhz ?

Baltimore,MD
Houston,TX
Kansas City,KA
Las Vegas,NV
Washington DC

Thats all I've seen announced in press releases so far is there more?

[CBMC]

Can anyone confirm for me if youngstown OH market has 42mbps? i know it has atleast 21 but not sure if its 42.

Also is there a list somewhere of all the areas with refarmed 1900mhz ?

Baltimore,MD
Houston,TX
Kansas City,KA
Las Vegas,NV
Washington DC

Thats all I've seen announced in press releases so far is there more?

Did kc obtain more spectrum? I know they used the be a single 10 mhz channel. Not enough to run hspa+42.

Sent from my Nexus 7 using Tapatalk 2

[valeuche]

One thing that's very important to note is that when a carrier refers to HSPA+42, and that 42 represents 42 megabits / second, that is referring to actual symbol bits being transmitted. It does not account for error correction bits, IP framing, TCP or UDP packetization or TCP retransmission. You will never, on any carrier, see a speedtest which is making use of UDP or TCP to transmit its data get anywhere near the actual line speed. There's no way even to measure it. All you can do is measure delivered payload, which AT BEST is going to be maybe 75% of the optimal line speed.

You can see the same thing on a 10 MHz 2x2 MIMO LTE carrier like Verizon, which if measured the same way as "HSPA+42" would bill itself as 108 mbps (9 million symbols per second at 10 MHz * 6 bits per symbol * 2 spatial layers minus whatever error correction rate there is - though +42 counts error correction bits) . There has never been a test I've seen anywhere near that.

You are one of the lucky ones because t mobile is a joke in Atlanta!

They say i should get great coverage here on the map, but I'm almost always having dropped calls, can't get a signal in buildings, distortion, and i really wish i didn't leave Sprint.

This hspa+ 42 never goes ANYWHERE NEAR 42mbps, most i get is 22 and that's in the middle of the night when no one else is in. 21mbps doesn't even go that high nor close to it.

I wish t mobile would stop lying about that

My wife has Verizon, she gets a slightly better signal than Sprint (which was excellent btw) and she is able to talk indoors! I can't with t mobile unless i use Wi-Fi calling. That's gay.

She gets a signal everywhere i go.

Verizon is definitely worth the extra 10 dollar a month

sent from the best smart phone (not phablet) available - the Galaxy S3!

[valeuche]

Did kc obtain more spectrum? I know they used the be a single 10 mhz channel. Not enough to run hspa+42.

Sent from my Nexus 7 using Tapatalk 2

They presently have (in Kansas City, KA):
1870-1885 MHZ, 1950-1965 MHz (15 MHz FDD PCS)
1895-1910 MHz, 1975-1990 MHz (15 MHz FDD PCS)
1735-1740 MHz, 2135-2140 MHz (5 MHz FDD AWS)
1730-1735 MHz, 2130-2135 MHz (5 MHz FDD AWS)

So that 40 MHz, and three blocks with at least 10 MHz contiguous that they could do DC-HSPA on.

[bobsacomano]

What exactly does refarming of 1900 spectrum do anyway?

Sent from my Galaxy Nexus using Tapatalk 2

[fabulas]

Tmobile and AT&T stores out here use the store Wi-Fi so I can never do a speed test. Please post a screenshot when you can.

Turn Wi-Fi off in the handset, then it will use the networks data. Most T-Mobile stores in Memphis do the same thing.

Sent from my Galaxy Nexus using Tapatalk 2

[Darrkman]

Was just in San Juan, Puerto Rico and did a speed test while on the beach. Was very pleased:



Beach was beautiful.



Sent from my HTC One S

[erwaso]

Turn Wi-Fi off in the handset, then it will use the networks data. Most T-Mobile stores in Memphis do the same thing.

Sent from my Galaxy Nexus using Tapatalk 2

Lol I have. There aren't any SIM cards activated in them.

[erwaso]

Any US speed tests screenshots out there?

[valeuche]

What exactly does refarming of 1900 spectrum do anyway?

Sent from my Galaxy Nexus using Tapatalk 2

1900 mhz pcs is one of the very common north and south American umts bands, unlocked AT&T phones will be able to use umts / hspa on tmobile (up until now they were limited to edge).

Additionally, you have to look at it from the point of view of tmobile, where once they might have had 330 people being served by 10 mhz, now those same 330 are being served by 40 MHz in some places.

[reign831]

Thanks.

Of course. "Subject to change."

I spend time in Durham, Raleigh, Sunnyvale, Mountain View, Santa Clara,... I can tell you that while I do get 4 bars of 4G in some spots, and some nice download speeds, I HAVE NOT as yet field tested anything better than 17MBps (and that was in my bedroom, probably line of site a 1/4 mile from a tower). HSPA+ 42 my eye. [This is with a 01/2013 made HTC Amaze which should be on all the necessary freqs.) And I see the phone switching 4G, 2G, UMTS/HSDPA a lot, causing a lot of battery drain I'd guess.

FYI I'm not a cellular wireless technician, don't even play one on TV. So I don't know much about how this sheot works (my last professional accreditation was on VOR, TACAN, RADALT, GPS, etc, not even close).

Yeah, I'm in an HSPA 42 area (Salinas, CA) and with full bars the highest speeds I've seen are 17.97MBp/s

Certainly good speeds, but I haven't seen anything near the 42 area.

That said, when I had my Galaxy Nexus on Verizon LTE, I think I hit like ~43MBp/s in the same area. But in reality I don't think many people would ever know the difference. Once speeds are fast enough, it's hard to tell.

[valeuche]

Yeah, I'm in an HSPA 42 area (Salinas, CA) and with full bars the highest speeds I've seen are 17.97MBp/s

Certainly good speeds, but I haven't seen anything near the 42 area.

That said, when I had my Galaxy Nexus on Verizon LTE, I think I hit like ~43MBp/s in the same area. But in reality I don't think many people would ever know the difference. Once speeds are fast enough, it's hard to tell.

18 mbps in delivered payload is quite good. To keep chanting the mantra, 42 mbps is 42 mbps raw (except for the 1:16 code control channel overhead), error correction bits / code rate, IP framing and TCP and UDP packetization and TCP retransmission are not counted against that 42 mbps. Phone companies can't really advertise what they would expect you to receive, because even things like distance from the cell tower, how the application configures and uses the network (from an IP level) and other people using the cell at the same time all impact that.

The average cell sector is supposed to contain about 300 subscribers, it's got to be a fairly unusual instance in which you are the only person using streaming data when you run a speed test.

All T-Mobile can do when they say "42 mbps" is say that in optimal conditions, they are indeed transmitting 42 million actual bits every second. A whole lot of those bits, by necessity, are error correction and all the other stuff I listed. Also in any given cell sector, radio conditions won't allow for 6 bits per symbol, and you'll be stuck at 4 or 2 bits per symbol for the majority of the territory of the cell, which is either 2/3 or 1/3 of the full data rate, i.e., 28 or 14 mbps raw of which you would observe about 50 to 70% of that as delivered payload.

They probably should, though, as the actual deployed LTE does vary and "would be called" LTE+13 (1.4 MHz 2x2), LTE+32 (3 MHz 2x2), LTE+54 (5 MHz 2x2) and LTE+108 (10 MHz 2x2) in the same scheme, depending on carrier and where you are (it'd be lower than that actually, as HSPA+42 does, as I mentioned, not count the 1/16th code for control toward it's bandwidth and I'm not sure what the LTE control channel overhead is). That even varies by country, I remember seeing a video from sweden in which a LTE modem is being tested and it was mentioned that in a 10 MHz channel the capacity was 50 mbps, which implies that it was a 10 MHz SISO deployment.

This is true across all of these radio technologies. With LTE (perhaps fortunately, or with the understanding that the general public is not technically literate) no one is advertising or stating the "line speed" or raw bit rate - which would be something like 108 mbps on 10 MHz 2x2 MIMO LTE - to avoid complaints that no one is actually seeing anywhere near that - thus LTE is just LTE, not LTE+108 or what have you.

Another edit. Based on this comment: "The LTE network has theoretical data throughput speeds on the forward link of up to 22.7Mbps for a 5 MHz paired channel", found here: , it appears that the MetroPCS LTE deployment is SISO, thus it would be LTE+6, LTE+16, and LTE+27. The control channel overhead would appear to be 15.9%.

[laststop311]

they refarmed 1900mhz 2g to 1900mhz hspa+ in those 5 markets I was asking if they did more.

Reason I'm asking is because I want to get a Lumia 920 on t-mobile and i need 1900mhz refarmed to get full usage.

Also wanted to know if youngstown,oh was hspa+42 or not as well as the refarming.

[valeuche]

Looks like only Cleveland and Columbus so far.

Someone is tracking reports of 1900 MHz HSPA on TMobile here:

they refarmed 1900mhz 2g to 1900mhz hspa+ in those 5 markets I was asking if they did more.

Reason I'm asking is because I want to get a Lumia 920 on t-mobile and i need 1900mhz refarmed to get full usage.

Also wanted to know if youngstown,oh was hspa+42 or not as well as the refarming.

[evperry]

Tmobile in NYC is awesome and the prepaid is significantly cheaper than most others.

I will say this, I prefer HSPA+ over LTE because of the battery drain issues with LTE. My HSPA Galaxy can go a whole day on a charge. The Verizon version of the same phone couldn't make it half the day.

Different everywhere you might be right. But one thing my Verizon Galaxy Nexus never did was hold a signal and had **** poor reception. I come of the street inti my house and instantly my 4g Lte symbol would drop from LTE to 3g to 1x on Verizon. On Tmobile i get constant signal great reception and can watch 2hr movies on Netflix. I was never able to do that on my Verizon Galaxy Nexus. Tmobile has improved drastically since 2009 and I am here to tell you that their upcoming LTE ADVANCED 10 NETWORK which will be up and running by June 2013 which will be back hauled to it's current HSPA PLUS 42MPS network will be super faster than any current Lte network. Remain blessed

Sent from my Galaxy S3 on Tmobile..

[yukichi]

Tmobile in NYC is awesome and the prepaid is significantly cheaper than most others.

I will say this, I prefer HSPA+ over LTE because of the battery drain issues with LTE. My HSPA Galaxy can go a whole day on a charge. The Verizon version of the same phone couldn't make it half the day.

Im pretty sure the battery drain of the Nexus 4 was more than an Iphone 5.

[evperry]

Im pretty sure the battery drain of the Nexus 4 was more than an Iphone 5.

My wife's iphone 4s battery was absolutely amazing. Samsung and LG need to get on board with whatever they are using.

Sent from my Galaxy Nexus using Android Central Forums

[evperry]

I refuse to believe that I am the only one that doesn't know what you are talking about.... Even in the mildest since.

It is cool to have engineers on a board like this.

18 mbps in delivered payload is quite good. To keep chanting the mantra, 42 mbps is 42 mbps raw (except for the 1:16 code control channel overhead), error correction bits / code rate, IP framing and TCP and UDP packetization and TCP retransmission are not counted against that 42 mbps. Phone companies can't really advertise what they would expect you to receive, because even things like distance from the cell tower, how the application configures and uses the network (from an IP level) and other people using the cell at the same time all impact that.

The average cell sector is supposed to contain about 300 subscribers, it's got to be a fairly unusual instance in which you are the only person using streaming data when you run a speed test.

All T-Mobile can do when they say "42 mbps" is say that in optimal conditions, they are indeed transmitting 42 million actual bits every second. A whole lot of those bits, by necessity, are error correction and all the other stuff I listed. Also in any given cell sector, radio conditions won't allow for 6 bits per symbol, and you'll be stuck at 4 or 2 bits per symbol for the majority of the territory of the cell, which is either 2/3 or 1/3 of the full data rate, i.e., 28 or 14 mbps raw of which you would observe about 50 to 70% of that as delivered payload.

They probably should, though, as the actual deployed LTE does vary and "would be called" LTE+13 (1.4 MHz 2x2), LTE+32 (3 MHz 2x2), LTE+54 (5 MHz 2x2) and LTE+108 (10 MHz 2x2) in the same scheme, depending on carrier and where you are (it'd be lower than that actually, as HSPA+42 does, as I mentioned, not count the 1/16th code for control toward it's bandwidth and I'm not sure what the LTE control channel overhead is). That even varies by country, I remember seeing a video from sweden in which a LTE modem is being tested and it was mentioned that in a 10 MHz channel the capacity was 50 mbps, which implies that it was a 10 MHz SISO deployment.

This is true across all of these radio technologies. With LTE (perhaps fortunately, or with the understanding that the general public is not technically literate) no one is advertising or stating the "line speed" or raw bit rate - which would be something like 108 mbps on 10 MHz 2x2 MIMO LTE - to avoid complaints that no one is actually seeing anywhere near that - thus LTE is just LTE, not LTE+108 or what have you.

Another edit. Based on this comment: "The LTE network has theoretical data throughput speeds on the forward link of up to 22.7Mbps for a 5 MHz paired channel", found here: , it appears that the MetroPCS LTE deployment is SISO, thus it would be LTE+6, LTE+16, and LTE+27. The control channel overhead would appear to be 15.9%.

Sent from my Galaxy Nexus using Android Central Forums

[abtxpress]

LTE on Verizon in St Louis.

Sent from my Galaxy Nexus using Tapatalk 2