Dish Network/Boost Mobile cell/5G buildout thread

[bigsnake49]

13 hours ago, iansltx said:

Thing is, building a coverage focused network to cover 70% of the US population using 600 isn't *that* expensive. It's the capacity play that's expensive. You need capacity if you have a ton of customers, which 10MM isn't.

Alltel had 12 million customers when Verizon bought them, and covered a larger area than Dish will need to. If you spend $5 billion building a network for, say, 15 million customers, and don't have to subsidize those customers, that's not a huge outlay in the scheme of things.

I am sure that they can cover the 70% of American using their low frequency spectrum no problem. The only problem would be the relatively low customer count. They will have to partner with somebody to spread out the deployment costs. 

[iansltx]

Dish bought Ting today, will continue to contract with Tucows for MVNO billing purposes, and will switch Boost over to Tucows' billing system in the second half of next year.

So Tucows has a bit under a year to get their system set up to handle 30x the customer count that it does now...more if Dish gets super aggressive on customer acquisition.

As an aside, there's every indication that Dish is building their network out as cheaply as they possibly can, using cloud providers for their core network rather than their own data centers, buying base stations from Fujitsu rather than Nokia/Ericsson/Samsung, and now farming out their billing software to a company that they fully expect will sell that software to other customers.

Which...fair enough. This feels a bit like Clearwire's strategy, among others', and if that's what it takes to build out a nationwide network in three years without getting bought by Comcast/Spectrum/AT&T/VZW in the process I can live with it.

This'll also make it way easier for me to test out Dish's network, since I'll just be able to pick up a Ting SIM for it when the network becomes a thing.

[Tengen31]

I am also interested in what T-Mobile or Dish will do with the b26 spectrum. While Dish has an option to buy it after 3 years and a penalty if they don't, I am not sure why they would want to. I would like for T-Mobile to sell it to the Cellular Carriers (band 5) and have them redo band 5 to they can each have 15x15 allocations.

It's only contiguous with the cellular A block not the B block.. TMO does have the A block B5 themselves. I think Dish will buy it as it will help with coverage even tho B71 will be more useful. I also think I heard the equipment that dish has ordered support N26

Sent from my SM-G975U1 using Tapatalk

[Tengen31]

...which would be B26. Just like "reconfiguring B4 to include extended frequencies" resulted in B66.
FWIW the slice of B26 Sprint uses for LTE is right next to the bottom of B5, with a center frequency of 821.3 MHz on the uplink. That leaves 200 KHz between the top of the "advertised" 5 MHz LTE band (actually a bit more than that because as I recall a 5 MHz LTE carrier has 500 KHz of guard band built in) and where CLR starts. Then the 1xA carrier Sprint's using runs below that, rounding out their use of the 7x7 of SMR they have.
Catch here is, you'd have to reband the entirety of CLR + SMR to get what amounts to 5 MHz of extra spectrum over what's available in CLR + SMR right now...and that would have to happen across everyone in the band. 'cuz right now you have 10x10 + 10x10 + 5x5...potentially even better spectrum usage for folks who are shimming a 1x channel or some GSM in there (Verizon or rural carriers). You're not just dealing with all three major carriers.
At which point you only benefit spectrum-wise if you drop from three carriers to two in the band...and you can get to two carriers in the band with one of them having 15x15 without rebanding. Hard sell when the two carriers you'd see in that band already have B12 and B13 networks.
I'd much rather see Dish hold onto that spectrum, shunt all LTE users to B12, and replace the B26 carrier with 5x5 NR. T-Mobile can rent the remaining 2 MHz back from Dish for 1xA until they shut down CDMA entirely, allowing T-Mo to drop PCS CDMA (RIP) more quickly. Phones have had SMR CDMA for 8+ years at this point so switching PCS CDMA off shouldn't be a big deal at all. T-Mobile can use the existing ex-Sprint equipment to run the network on the cell sites they're keeping since you don't need nearly the density for 1x on SMR that you need for anything at PCS or higher.
Yes, this assumes Dish will actually launch a network. Holding out hope that that does indeed happen.

People don't get that stuff just like the ones that attacked sprint for using B41 2.5 instead of B7.. Can't do B7 which is why TMO now it's using n41

Sent from my SM-G975U1 using Tapatalk

[Tengen31]

I personally don't think Dish will buy the SMR spectrum from Sprint. 

The equipment they are ordering supports N26

Sent from my SM-G975U1 using Tapatalk

[Tengen31]

Right now you have a total of 12.5x12.5 of cellular spectrum for each cellular carrier. You have 7 MHz of ESMR. You could have 30 Mhz of band 26 owned by the cellular carriers and couple of Mhz for guard bands. T-Mobile does not want the band 26 and neither does Dish. Otherwise Dish would not have an option, they would have a firm agreement. 

TMobile did want B26 FCC and the DOJ said no

Sent from my SM-G975U1 using Tapatalk

[bigsnake49]

12 hours ago, Tengen31 said:

TMobile did want B26 FCC and the DOJ said no

Sent from my SM-G975U1 using Tapatalk
 

I did note that they are buying NR 26 capable equipment. I am surprised and not what I would do but it is what it is.

Edited August 4, 2020 by bigsnake49

[red_dog007]

14 hours ago, iansltx said:

Dish bought Ting today, will continue to contract with Tucows for MVNO billing purposes, and will switch Boost over to Tucows' billing system in the second half of next year.

So Tucows has a bit under a year to get their system set up to handle 30x the customer count that it does now...more if Dish gets super aggressive on customer acquisition.

As an aside, there's every indication that Dish is building their network out as cheaply as they possibly can, using cloud providers for their core network rather than their own data centers, buying base stations from Fujitsu rather than Nokia/Ericsson/Samsung, and now farming out their billing software to a company that they fully expect will sell that software to other customers.

Which...fair enough. This feels a bit like Clearwire's strategy, among others', and if that's what it takes to build out a nationwide network in three years without getting bought by Comcast/Spectrum/AT&T/VZW in the process I can live with it.

This'll also make it way easier for me to test out Dish's network, since I'll just be able to pick up a Ting SIM for it when the network becomes a thing.

That's cool.  Ting has 154k subscribers with 272k devices.  A $37 APRU and 3.25% churn.  I might look at Ting when it's time to renew my MVNO plan. I wonder if Dish is gonna plan on boosting their subs via additional acquisitions or if they acquired Ting subs so they could get access to Tucows software.

I wouldn't equate cheap = bad either. Not that you specifically are but usually cheap implies junk or not the best.

Edited August 4, 2020 by red_dog007

[iansltx]

7 hours ago, red_dog007 said:

That's cool.  Ting has 154k subscribers with 272k devices.  A $37 APRU and 3.25% churn.  I might look at Ting when it's time to renew my MVNO plan. I wonder if Dish is gonna plan on boosting their subs via additional acquisitions or if they acquired Ting subs so they could get access to Tucows software.

I wouldn't equate cheap = bad either. Not that you specifically are but usually cheap implies junk or not the best.

Dish is making unconventional decisions to save money. It will succeed in doing so. As someone who uses Ting, their billing systems are fine. Nothing super special, but fine. As someone who runs plenty of stuff on cloud services, they're fine, but I'm skeptical about putting them at the core of your mobile network. Using a smaller company for NR gear may be fine, but Dish will be the first one Fujitsu builds low band NR equipment for.

If one of these links in the chain goes askew, Dish will wind up competing against CricKet/Metro/Visible rather than the postpaid equivalents. They may have enough of a cost advantage to make a go of it, just like CricKet/MetroPCS did back in the day. But it's definitely a tougher road when you can't get anyone to pay more than $45/mo for your service.

[bigsnake49]

On 8/4/2020 at 7:10 PM, iansltx said:

Dish is making unconventional decisions to save money. It will succeed in doing so. As someone who uses Ting, their billing systems are fine. Nothing super special, but fine. As someone who runs plenty of stuff on cloud services, they're fine, but I'm skeptical about putting them at the core of your mobile network. Using a smaller company for NR gear may be fine, but Dish will be the first one Fujitsu builds low band NR equipment for.

If one of these links in the chain goes askew, Dish will wind up competing against CricKet/Metro/Visible rather than the postpaid equivalents. They may have enough of a cost advantage to make a go of it, just like CricKet/MetroPCS did back in the day. But it's definitely a tougher road when you can't get anyone to pay more than $45/mo for your service.

Rakuten was the pioneer in cloudified RAN or C-RAN.They are selling their expertise in putting together their network. I can see even the RRH's converted into software run on general-purpose hardware and just the antennas being at the site. Of course you need dependable and inexpensive fronthaul for all this to work. Rakuten could find that in Japan but is the US fiber situation similar?

[ingenium]

As an aside, there's every indication that Dish is building their network out as cheaply as they possibly can, using cloud providers for their core network rather than their own data centers, buying base stations from Fujitsu rather than Nokia/Ericsson/Samsung, and now farming out their billing software to a company that they fully expect will sell that software to other customers.

I'm pretty sure Fujitsu equipment is actually fairly advanced. The Japanese carriers tend to use domestic suppliers for their equipment and we don't see them much outside of Japan. One reason likely is the unique bands in Japan and support for proprietary and Japanese only legacy protocols. From what I've seen and heard, the equipment is surprisingly good and solid, and is definitely competitive with the big vendors we're used to seeing.

Sent from my Pixel 4 XL using Tapatalk

[red_dog007]

https://www.fiercewireless.com/wireless/dish-marks-progress-5g-network-build

States that Dish is looking to launch one market by end of year.  Wonder which city this will be.

[greenbastard]

38 minutes ago, red_dog007 said:

https://www.fiercewireless.com/wireless/dish-marks-progress-5g-network-build

States that Dish is looking to launch one market by end of year.  Wonder which city this will be.

My guess is Knoxville, TN. 

[iansltx]

Boost's new plans, announced at the beginning of this week, move downmarket from the stuff they already had: monthly fees ranging from $10 to $25, and hard-capped data (1-5 GB) rather than throttled. Oh, and streaming is limited to 480p unless you pay more, which is weird considering it's a limited data bucket.

Their previously "unlimited" plans are now throttled after 35 GB. Their plan page isn't terribly clear about hotspot allotments on those plans...I see 12 GB further down the page, but not on the $50 / $60 plan cards.

Big question for me is how they'll differentiate their native network from T-Mobile's service/pricing-wise, if at all. According to their FAQs, only plans at or above $35/mo get access to 5G if you have the one device that supports it. Dish could certainly build out their network where they feel like it's worthwhile, falling back to T-Mobile everywhere else, and only allowing higher-tier plans onto that network because they're the only customers where a large amount of data usage could put a cramp on Dish's profit margins.

As for market launches, my bet is that Dish launches in a market where they own both band 66 and 71 spectrum. That will allow them to get phones a bit more easily. Question is how they'll handle voice. VoNR would work if they got new device support. Other approach would be a dual-SIM setup with the secondary running on T-Mobile LTE. Chipsets down to the Snapdragon 690 support that operation.

Or Dish could get a phone manufacturer to throw a MediaTek Dimensity 820 in a phone for them, as that chip already supports VoNR. That should also be a cheaper chip because, well, not Qualcomm, and I doubt Dish has any illusions about targeting the high end with the initial launch either. They could get a single phone model, order a bunch of 'em, and have that single phone be their launch model for ~$500. Question is where that phone would come from; if they want a launch by year-end with an actual phone, the phone would need to be in FCC testing now if it's a new model. Right now the Redmi 10X 5G would fit the bill, albeit with different LTE and 5G bands...and the whole "being a big Chinese brand" issue.

One final option would be for Dish to start their native network rollout with hotspots, sidestepping the whole voice issue. But that would require getting someone to build out a 5G SA capable hotspot for a reasonable price. Closest thing right now is Samsung's 5G capable tablets.

[bigsnake49]

A more simple question. Does Boost still use Sprint as their base network or T-Mobile?

[dkyeager]

2 minutes ago, bigsnake49 said:

A more simple question. Does Boost still use Sprint as their base network or T-Mobile?

I believe most users were resimmed and now use T-Mobile.

[bigsnake49]

7 minutes ago, dkyeager said:

I believe most users were resimmed and now use T-Mobile.

Thanks, I did not know if they had done that. Do they get to roam on Sprint if T-Mobile has no service?

Edited August 26, 2020 by bigsnake49

[Yuhfhrh]

On 8/26/2020 at 11:24 AM, bigsnake49 said:

Thanks, I did not know if they had done that. Do they get to roam on Sprint if T-Mobile has no service?

No, the new T-Mobile boost sim cards do not roam onto Sprint. The old Sprint sims will roam on T-Mobile however.

[JimBob]

DOJ says no to Dish-DirecTV merger: https://seekingalpha.com/news/3622621-doj-dampens-hopes-for-directv-dish-combination-report

[red_dog007]

Still get higher prices to whomever might buys DirecTV. Might either be a foreign company or a sat company who doesn't care about the TV service.

What spectrum does DirecTV use/own?

FCC could put in the agreement for no price increases for so many years.

[bigsnake49]

That's a brilliant move by AT&T, buy high, sell low, use the losses to lower your tax bills for years to come 🤣.

[iansltx]

https://www.lightreading.com/open-ran/dish-networks-turns-on-first-5g-test-site-taps-intel-for-silicon/d/d-id/765125

Betting they'll start with tri-band base station equipment but only turn on n71, using the spectrum they haven't leased to TMo.

Outdoor-only coverage with n71 sounds like a super thin network deployment. They're gonna have to do something creative pricing-wise to get anyone to actually use the network, given that basically everyone on Boost Mobile doesn't have a 5G phone.

[JimBob]

Dish won’t have major 5G market up and running until Q3 of 2021

https://www.fiercewireless.com/operators/dish-won-t-have-major-5g-market-up-and-running-until-q3-2021

[PythonFanPA]

https://www.foxbusiness.com/markets/dish-ceo-charlie-ergen-in-hot-water-could-face-billions-in-claw-back-to-fcc

[Jenn_3012]

On 6/2/2020 at 10:44 PM, bigsnake49 said:

It will be very interesting to watch a greenfield 5G network being built from the ground up. From what I understand it will be based on open RAN and C-RAn it that the equivalent of enodeB and RRUS can reside in the cloud virtualized as software and ran on general purpose computers. Of course, there are tradeoffs in that you need low latency/high bandwidth connectivity between your antennas and the rest of the network. So a successful network strategy will probably have to mix and match depending on whether you're talking about urban, suburban and rural. In urban setting you could virtualize both the enodeB and the RRU, for the suburban setting virtualize the enodeB but keep the RRU as hardware and in rural setting keep both as hardware.

Could you explain how would virtualizing both the enodeB and the RRU help in the urban settings?

Thanks in advance!