iansltx

CricKet is not using its spectrum efficiently at this point, and has elected not to pay Sprint for roaming access fees last I checked. They may be able to be acquired on the cheap, and customers can be very easily migrated to Sprint. Shut down CricKet's own network, add the spectrum to Sprint, then swap T-Mobile for even more PCS spectrum and you've got a good business case for the acquisition. After all, where are customers going to go if they want to defect from Leap after a Sprint acquisition?

By the way, there are a few markets where the scale of this transaction is very visible: NYC - T-Mobile will end up owning AWS C-F, 25x25 MHz contiguous (VZW now owns the other 20x20 MHz) Dallas - T-Mobile will end up with AWS A + D-F, 10x10 + 20x20 contiguous (VZW wons 10x10 here, AT&T owns 5x5) In Dallas, T-Mobile has 15x15 in PCS of its own. MetroPCS has 5x5 additional, which is adjacent to Sprint but not T-Mobile. TMo could eventually drop an HSPA+ carrier on that channel, or it could sell the slice to Sprint for a healthy sum (or swap it for AWS if Sprint buys CricKet). EDIT: LA will have 25x25 contiguous AWS as well post-MetroPCS.

Pretty sure they're using RRUs. And as long as the spectrum is harmonized (it is), there's no huge cost to add to an existing deployment.

T-Mobile, like Sprint, has been able to do incredibly well without using sub-GHz spectrum (of which Sprint has 14MHz). With RRUs (which T-Mobile has been deploying), their PCS site spacing should work well enough for AWS LTE, particularly since modulation complexity of HSPA+ is not far off from LTE (if I remember correctly, both are 64QAM at full rate). As for their WCDMA network, T-Mobile only needs to heep a single 5x5 channel in AWS. The rest can go in PCS, along with GSM and (temporarily) CDMA. Bets on which 2G tech gets phased out first? The real question is how much spectrum each carrier has for LTE, I suppose, accounting for their legacy networks. Sprint has effectively 40MHz in most areas, spread between PCS and SMR, since they'll probably need to keep 5x5 in PCS earmarked for CDMA for the foreseeable future (enough for three CDMA carriers). Out of the remaining 30MHz you have 5x5 in SMR and 5x5 in the PCS G block, which Sprint has to keep as-is because a) There's no more space in SMR and Many Print phones only support 5x5 LTE. That leaves at best a single 10x10 channel for LTE on Sprint, discounting Clearwire spectrum. The formula for T-Mobile is more straightforward: take AWS holdings and subtract 5x5. The company can cherry-pick phones that support 20x20 LTE-FD, relying on HSPA+ (including DC-HSPA) for phones that can't do that bandwidth. But back to 20x20...T-Mobile could already do that in a few markets (such as Seattle) and now they can do that in more. This is in AWS spectrum too, so you're looking at a capacity/coverage ratio 4x that of Sprint's with a single 5x5 carrier. O fcourse, the math doesn't quite work out this way...Sprint can add more 5x5 LTE channels...but normal issues with wide channels (reduced transmit power) are mitigated by the fact that LTE is made up of tons of tiny subcarriers, so it appears as though you can transmit 20MHz with just as much RSRP as 5MHz. The scary thing for Verizon, AT&T and Sprint is that T-Mobile's phones will be capable of 20x20 FD-LTE out of the gate if the company so chooses. No one else has the right phones with the right bands to compete with this. And TMo can deploy their network at a good clip since they already have RRUs on some H+ sites and enhanced backhaul to practically every H+ site. The elephant in the room, of course, is Newco's debt. If DT gets greedy, TMo will not be able to move quickly enough. If DT is relaxed about collecitng the loan, be prepared for a company that, once they get CDMA dealt with, will give the other three carriers a run for their money in and around any medium-sized or larger city. Related: I expect Sprint to buy CricKet and swap its AWS for T-Mobile PCS within a year. Now especially since CricKet has a few key pieces of AWS spectrum that T-Mobile would love to have, and between T-Mobile and MetroPCS there's a fair amount of PCS spectrum that Sprint would like.

It could be that, by shuffling spectrum around, T-Mobile can isolate MetroPCS CDMA on 1900MHz, rather than just shutting it down ASAP. I think this might take away from T-Mo HSPA spectrum in Las Vegas, but otherwise it's doable within MetroPCS's own holdings (in Dallas they have a 5x5 LTE carrier, which means that all of their voice/non-LTE data is on PCS). And for AWS-only MetroPCS markets, T-Mobile could still just shut down CDMA entirely and push customers over to Sprint CDMA. The result: T-Mobile gets between 5x5 and 10x10 of AWS spectrum to add to their network (10x10 in NYC, adjacent to their current holdings). The company will be able to deploy 20x20 LTE and 5x5 HSPA in a number of markets that they weren't able to before (NYC being a huge one), with the added bonus of more PCS to deploy H+ on. Can't wait for AJ to get a spectrum chart up for AWS and PCS...because I'm too lazy to do it myself

Guess I've got some firmware flashing to do...I'll grab the firmware now so I'm not waiting around for it EDIT: On second thought, I'll wait for LIH to be available fully for download.

Not the foggiest.

EvDO roaming on former Alltel's Cellular-A license? Channel 28 would be in the CLR-A band, but I'm too lazy to look up who wons it in Boston. EDIT: Apparently PCS has that channel number as well. It would have to be the lowest possible CDMA channel in PCS (so the A block), with a center frequency 1.4 MHz above the bottom of the band.

If I was T-Mobile and if this deal went through, I would immediately ink a deal with Sprint for voice and data roaming on CDMA (but not LTE), effectively making MetroPCS's subscribers MVNO customers of Sprint for the time being. Sprint is selling airtime cheaply enough that this is a reasonable deal. Why not LTE? Because MetroPCS has LTE and T-Mobile doesn't quite yet...and you can have phones using LTE on one network and 1x on another, particularly with Sprint eHRPD deployed. Then push a custom PRL to all MetroPCS handsets that re-marks Sprint SIDs in MetroPCS areas as preferred, shifting all CDMA traffic off of MetroPCS's network. Then turn off CDMA entirely on MetroPCS, leaving subscribers roaming on Sprint for the time being for the cheap stuff (voice and texting). Use the extra spectrum to up LTE channels on MetroPCS sites to 5MHz for the time being, using only MetroPCS spectrum (a data network is easier to admin than a voice network, and TMo will have LTE soon enough anyway). At some point during all of this, you've stopped selling CDMA phones on MetroPCS, converting the company on a going-forward basis to a T-Mobile MVNO running on HSPA+. Maybe rename Monthly4G to MetroPCS and call it a day. Subscribers who want a new phone and are on CDMA MetroPCS (which tends to have inferior coverage to T-Mobile anyway) can get a T-Mobile phone at a discount, incrementally reducing TMo's roaming bill to Sprint. The next phase of the plan happens when T-Mobile rolls out LTE in a given area. All MetroPCS phones that have LTE onboard support LTE in AWS, so when a T-Mobile LTE site comes online, the redundant MetroPCS site(s) in its shadow can be shuttered. In fact, by the time this deal goes through, T-Mobile may already have big markets like Dallas upgraded to LTE, allowing them to take MetroPCS entirely offline in the city rather than just turning CDMA off on MetroPCS cell sites. Once MetroPCS sites are offline, T-Mobile has that much more spectrum to use for its LTE-A network, built on NSN and Ericsson equipment. But what about MetroPCS's current LTE network? Well, its vendors are Samsung and Alcatel-Lucent if I remember correctly. T-Mobile's vendors are NSN and Ericsson. My bet is that T-Mobile sells the MetroPCS equipment to Sprint at a discount, since Sprint can then just ship the gear to the correct parts of the country, leaving both companies with vendor-coordinated LTE rollouts nationwide. As an added LTE bonus, if T-Mobile can get VoLTE working on the NSN/Ericsson equipment, they have even less of an MVNO bill to pay to Sprint. Though there aren't very many VoLTE capable MetroPCS handsets out there at this point, and there probably still won't be if the merger goes through. Hey, T-Mobile, wanna hire me for biz dev? EDIT: Ericsson and Samsung are MetroPCS's LTE vendors, not AlcaLu and Samsung. Now the big question is whether MetroPCS's equipment is too old to support LTE-A via a software upgrade and LTEr9 now. If it's new enough, T-Mobile can either reuse the equipment internally (Ericsson) or sell it to Sprint (Ericsson or Samsung). If it's r8-only, maybe sell to CricKet? lol

LTE-only won't allow you to make calls. Maybe that was your issue?

Yep. Been using it since Saturday night. I'd say it's 99% rock-solid, to the point that I'm not missing ICS in the least. Then again, I could be missing something major, since I haven't been paying close attention to my phone's ability to connect to LTE in the past few days (I'm just outside of LTE coverage for most of where I do stuff).

I was able to make and receive calls while connected to 4G on my GSIII a week or two ago.

Someone PM me Wednesday afternoon. I might be able to find out the Tx power comparison between AWS and PCS on TMo's network. I was thinking along those same lines, but if TMo is marketing PCS HSPA+ as a coverage benefit, maybe we're off...

...but siphoning off assets right as the proceeds could be used to do serious network upgrades...not a good idea. OTOH if T-Mobile were to merge with someone, optimal tower arrangements would a a bit easier if everything is more or less leased. The carrier isn't tied to some owned location. This is particularly handy since I guarantee that the T-Mobile owned sites are mainly spaced for PCS coverage, rather than AWS.

+1 for the GSIII. It will have Jelly Bean in a month or two (the leaked Sprint ROM appears to be rock-solid) and at that point you've got a fast, full-featured device that, ahem, can latch onto Sprint's LTE network without a problem. The only thing you'll possibly regret on the GSIII is when the next two LTE bands are added to Sprint phones, but that's not going to happen until the middle of next year or so.

At this point, I'm not sure whether battery life has taken a hit or whether I've just done a lot on my phone in the past 16 hours, beginning with the firmware upgrade. I'm going to go with the latter for now I'll echo Richmc about the experience on JB being smoother than on ICS. There are still times when things will take a second to load, but those times are fewer and farther between. At some point I need to see if GWallet still works on here; it's acting as if Ive rooted the phone when I haven't (though I had to flash CWM to get the update on board). Guess I need to figure out where the nearest PayPass-enabled merchant is first. On the modem side, I'm not getting any better 3G speeds than I did before (not too surprising since I'm 99% sure my sector is carrier-limited rather than signal or backhaul limited). I am, however, getting better latency; I hit 81ms to Dallas last night over eHRPD. Some pieces of the UI are more thoughtfully designed as well. The pull-down shade now has less options (no vibrate/silent, no airplane mode), however there's no more scrolling (which annoyed me a bit) and you can still access those settings via the power button menu (where you can access silent/vibrate/non-silent with one touch vs. cycling through them). The phone app now makes it very visually clear when a call is still connecting versus when the connection has been established, and TouchWiz...somewhat...conforms to JB's swipe-up-to-delete philosophy on the home screens, though the trash can is still a touch target rather than a large area.

RIchmc, did you just go back to stock complete stock?), download the file, put it on the root of your SD card, reboot into Recovery Mode and apply it? I'm getting a signature verification error on my stock SIII. EDIT: Bit the bullet and did a full factory reset, after backing up my apps and photos via ASTRO. It took installing ClockworkMod to get past verification errors (I immediately reflashed the stock recovery after I got the new modem and firmware installed). I guess I didn't need to do the full wipe after all ::rollseyes:: Setting up my phone again now...

Don't underestimate TMo HSPA+...I've been able to routinely pull 15M down, 3M up over their DC-H+ with their Rocket 3.0 modem, which is now, I dunno, a year or two old? That saide, Sprint outright sold its towers to Crown Castle a few years back. It makes sense for T-Mobile to do what it did (something similar) since it is getting an upfront paycheck equal to $1000 per month per tower for the duration of the lease. They may end up paying most of that back to Crown Castle by the end of the lease period, but at its end they still own the towers and that kind of money can do a lot of good right this minute for upgrading their network. From Crown Castle's perspective, it's a win as well; they can attract competitors to T-Mobile onto the T-Mobile tower (and get revenue from them), plus they pretty much have a guaranteed revenue stream from TMo. And they now can offer potential lessees 7200 more places to hang equipment. ...and of course I'm all for something that gets that company to launch their AWS LTE-A network sooner, and with $2.4 billion in extra cash as of right now they've got the funds to roll things out a bit more quickly.

I'm going to go with a bit of both. Wouldn't be surprised if he's way long in AT&T, somewhat long on Verizon, and short on everyone else, except maybe cable companies.

Heh. Sprint did sell all its towers awhile back, to Crown Castle I believe. T-Mobile and Verizon still own some of their towers, as does AT&T. Verizon is actually building new towers in some areas, as of a couple years ago.

As an LTE iPad owner (VZW), I can actually tether in the opposite direction (iPad as a modem or WiFI hotspot), which has been handy on trips over the last several months. I can also scale my data plan up or down, or even turn it off entirely, as needed. I can also do this with Sprint's tethering add-on, but that's not very helpful until I have LTE on Sprint everywhere I travel (I have it most places on Verizon).

Count me in as one who really, really wished that Alltel + Sprint would have gone through. It was a natural match...Sprint and Alltel already had reciprocal roaming agreements for voice and data (including 3G)...but all for naught. As for Sprint plus T-Mobile, my guess is that's what we'll see within five years. Quad-band LTE phones aren't so far-fetched (the iPhone 5 A1428 already supports CLR, PCS, AWS and 700-lower-B-and-C, while the A1429 supports CLR, PCS, 700-upper-C and 2100), and a merger would solidify AWS as a "common ground" LTE roaming band since the then-big-3 would all have a lot of spectrum there. There are a few catches of course. First, both companies would want to push for VoLTE ASAP to get off their respective, incompatible 2G/3G technologies. To handle the transition to a merged network, you'd basically have to ignore that the companies were combined for a few years and push quad-band LTE devices with VoLTE out as quickly as possible in the mean time...paying tons of cash to swap everyone's phones out when both companies have just completed their respecitve non-4G modernizations is a bad idea. Also, T-Mobile plus Sprint presents a lot of overlapping footprint issues. You'd end up with a lot of decommissioned cell sites. Many of these would probably be Sprint if the deal was done next year, as T-Mobile has fiber to most of their sites already. You could also end up with a much smaller force on the ground doing network upgrades, due to the lower tower count, but that's okay. Considering how slowly things are going for Sprint at this point, maybe the number of boots on the ground wouldn't decrease after all. In short, a T-Mobile plus Sprint merger would be messy, but since both companies are basing their next-gen strategies on LTE in roughly the same spectrum (AWS vs. PCS...cell spacing isn't much different and T-Mobile's network was built for PCS anyway) it's a lot more doable than Sprint + Nextel (which has effectively no network commonality). Of course, there are several acquisitions that would make a lot more sense before then. Scooping up Leap (who owes Sprint money anyway) or MetroPCS, spinning off their AWS spectrum to T-Mobile and migrating their customers to Boost Mobile (while till using the same phones) would be a relatively straightforward task, and T-Mobile would be happy to pay a good bit of money or the extra AWS to subsidize the deal. Sprint could then cherry-pick any good cell site locations, upgrading those to NV (in MetroPCS's case the equipment is already there, but the backhaul might not be, and the equipment might be the wrong vendor for a given region) and shuttering the others. The bits of PCS spectrum each of the two carriers have could be added to Sprint's holdings to increase the amount of LTE capacity available to every LTE-capable Sprint phone. CricKet and MetroPCS customers wouldn't mind any of this as long as they kept the same or better speeds and coverage, and the same or better plan pricing, both of which Sprint can accomodate. Sounds like a straightforward deal to me. As for US Cellular, they seem to be a bit stagnant, have a more rural footprint (but not nearly as big as Alltel) and are deploying LTE across a plethora of bands (700, 850, 1900, AWS). Divesting their 700MHz holdings would probably be the easiest way of dealing with them, but where would those holdings be sold? Their AWS might be too rural for T-Mobile (though it might not be), and their customer base just isn't that large. Better to have them as a seamless roaming partner than to buy them.

The cheap Nook is also really cheap. If someone can live with 8GB, I'd say go for it. It's nice that the iPad is being pushed hard my competitive tablets with large market exposure. It probably won't decrease the next-next-gen iPad's price point, but it will force Apple to innovate in some way or other to keep folks from going to cheaper, nearly feature-comparable, alternatives.

I'd certainly pay a $50 premium for a high-end smartphone if assembly was done in the US. I wonder how many hours assembly takes though...$50 might not be large enough to cover the cost delta between US and Chinese wages...though for putting together phones that's probably only semi-skilled labor.