Jump to content

Marcelo Claure, Town Hall Meetings, New Family Share Pack Plan, Unlimited Individual Plan, Discussion Thread


joshuam

Recommended Posts

It actually working is something new.

 

Sent from my Nexus 5X

So how is wifi to CDMA smooth handoff any different from LTE to CDMA smooth handoff? Aren't they both switching over from the IMS to the CDMA core network? Also is there a good way for me to attach a 30 second video here?

  • Like 1
Link to comment
Share on other sites

So how is wifi to CDMA smooth handoff any different from LTE to CDMA smooth handoff? Aren't they both switching over from the IMS to the CDMA core network? Also is there a good way for me to attach a 30 second video here?

If you have a YouTube account just use that.

  • Like 2
Link to comment
Share on other sites

Ok so I've confirmed that there is some form of SRVCC running on Sprint's network as of today. Today, as I left a friends house, a call I was on handed off from a wifi call to the macro network without dropping. It didn't occur to me at the time but I messed with it a bit later and found that the call handed off successfully from wifi to the macro network about 90% of the time. Also btw I have an iPhone 6S. I took a video of this happening with my ipad although I am not entirely sure how I should upload it. If anyone has any suggestions or comments I'm all ears.

This is honestly the best news and I'm very excited to try it out. Is this on iOS 9 or iOS 10 beta?

  • Like 1
Link to comment
Share on other sites

So how is wifi to CDMA smooth handoff any different from LTE to CDMA smooth handoff?

 

Very different.  The first combination long has supported simultaneous transmission.  The second combination has not.

 

AJ

Link to comment
Share on other sites

Ok so I've confirmed that there is some form of SRVCC running on Sprint's network as of today. Today, as I left a friends house, a call I was on handed off from a wifi call to the macro network without dropping. It didn't occur to me at the time but I messed with it a bit later and found that the call handed off successfully from wifi to the macro network about 90% of the time. Also btw I have an iPhone 6S. I took a video of this happening with my ipad although I am not entirely sure how I should upload it. If anyone has any suggestions or comments I'm all ears.

 

 

The mythical beast of CDMA SRVCC is huge news if true. 

Link to comment
Share on other sites

If that's the case AJ, can Sprint do bonded calling like Republic Wireless offers?

 

That is a proprietary solution, requiring custom firmware or operator bloatware.  People who care about VoLTE -- not I -- seem to want 3GPP/3GPP2 standard solutions compatible with BYOD, etc.

 

AJ

Link to comment
Share on other sites

That is a proprietary solution, requiring custom firmware or operator bloatware.  People who care about VoLTE -- not I -- seem to want 3GPP/3GPP2 standard solutions compatible with BYOD, etc.

 

AJ

 

 

VoLTE isn't really standardized to be uniform at this stage anyway. What we've had pop up has been a number of different VoLTE implementation profiles that are different than each other. I realize this will get worked out eventually, but the 3GPP and GSMA were asleep at the switch on this standard as far as making it a standard. That's why VoLTE support has lacked on a bunch of unlocked devices. 

 

3GPP compliant solutions that work on all carriers are really hard. People have to realize that. Unfortunately, I don't think a lot of people do realize that. 

 

Even on T-Mobile, which people will sometimes cite as the best carrier for unlocked devices, VoLTE support for unlocked devices is fairly limited. 

 

http://www.spectrumgateway.com/compatible-phones

  • Like 1
Link to comment
Share on other sites

I don't mind unlimited, so long as its done not to break the networks and it begins at an affordable price. My starting point reference for that is $45, but $60 is still within the range of being good. I just don't like this move to charging people $70, or telling them to go prepaid. That would be my main talking point at these meetings if I were to be invited. Although, I think Verizon would hate me. :P

 

The problem is that point of reference has no real world basis to it. Picking an arbitrary number doesn't work because there is significant additional cost considerations in picking MRCs.

  • Like 1
Link to comment
Share on other sites

I want the ability to turn on network side compression/bandwidth limitation for my Brother In Law who uses 18-20GB per month of our 80GB share plan.  Maybe someday.

 

Change the plan ;) lol

  • Like 1
Link to comment
Share on other sites

Change the plan ;) lol

Yeah, I'd love to kick his line to the curb and make him pay for it, especially since he is not going to be part of the family much longer.  If I switch plans, I lose my Netgear Zing and I don't care to pay $$$ for a separate plan.  $20 is already too steep when I typically use less than 3GB per month.  I keep it for when the family goes on a road trip and we use 30-40GB...damn YouTube Kids.  I still don't know why Google doesn't put a quality setting on that app so I can limit bandwidth usage.

  • Like 1
Link to comment
Share on other sites

Sprint's Fair & Flexible plan was one of the best rate plans offered by any of the carriers, in my opinion. Plus, they had great commercials to go with it, and I usually hate advertising. I would have liked, and maybe still would like to see a carrier use a Fair &  Flexible plan-like rate scheme for data buckets, at least for customers who'd prefer that.

 

I read your other response regarding Unlimited Data, so I'll respond here. I agree with you with regard to Unlimited. I like the idea T-Mobile is presenting, just not the cost. Hence why I haven't criticized Sprint's new plan, as the cost reductions after the initial line are more fair, plus they still offer data bucket plans for those not wanting unlimited, with overage protection.

 

If data buckets are to be eliminated, then there ought to be a starting cost at $60, not $70, unless there is some sort of speed cap option to give a lower cost option to differentiate from the $70. I read online today T-Mobile is going to keep the 2gb and the 6gb option, for now. Although they may eliminate it in the future, which I expect they'll do. Hopefully when they do, they'll reduce the starting rate down to $60, though I expect they'd just rather use that pricing point to get people over to MetroPCS instead. 

 

This is the problem, all the stuff you mentioned makes the world more complex. The industry shifting to a simple plan rate card, and the carrier introducing traffic shaping to keep their networks up and running.

Link to comment
Share on other sites

The honest truth is that SERO has had to go for a long time and the simplification of plans is way, way overdue in that regard. 

 

Either we're OK with Sprint's network being good enough, or we pay more to be competitive with Verizon, AT&T, and T-Mobile in that front. No shortcuts. 

 

To be honest I'd also say no more duplicate, out of standard networks. 

  • Like 3
Link to comment
Share on other sites

The honest truth is that SERO has had to go for a long time and the simplification of plans is way, way overdue in that regard. 

 

Either we're OK with Sprint's network being good enough, or we pay more to be competitive with Verizon, AT&T, and T-Mobile in that front. No shortcuts. 

 

To be honest I'd also say no more duplicate, out of standard networks. 

 

The problem appears Sprint can't keep the rates up high enough to build their cash that can be used to improve their network. Sprint has tried, but then it comes up with new plans that bring the rates back down, and as long as Sprnt does that, they may never get the amount of cash they need to get from customers to do so.

 

I think this is why the merger thing keeps being mentioned with Masa clinging to some hope he can get T-Mobile. With T-Mobile's already well-built network, combined with Sprint's massive spectrum holdings, Sprint would be resolved of its network issues pretty soon after working on the necessary conversions.

 

The thing left to wonder is what happens to Sprint if it doesn't get its way with T-Mobile. We'd all like to see Sprint succeed on its own, but from how it appears to be going, Sprint must really need T-Mobile, at least from what I've been getting from the articles I've read online.

  • Like 1
Link to comment
Share on other sites

I think this is why the merger thing keeps being mentioned with Masa clinging to some hope he can get T-Mobile. With T-Mobile's already well-built network, combined with Sprint's massive spectrum holdings, Sprint would be resolved of its network issues pretty soon after working on the necessary conversions.

A merger would NEVER help Sprint for many reasons, and here's why:

 

1) The cost of adding up debts - if Sprint was to buy T-Mobile, or T-Mobile buy Sprint, they would have to combine their total debts. Think about it. They would have to go further in debt just to buy the other one, and Sprint has large debts to pay for soon.

2) The cost of new equipment - not only would they have to remove redundant cell sites, but they would need to get all new panels and radios and cabinets to support band 4 and 12 at Sprint sites, band 41 25 and 26 at T-Mobile sites.

3) Conflicting tech - this new carrier would have to find a way to bring VoLTE everywhere because of T-Mobile's existing VoLTE plans, but CDMA network subs wouldn't have access to VoLTE because no handoff unlike the UMTS subs.

4) Regulatory hell - you think FCC would let them get 600MHz or 5G frequencies with their total spectrum holdings?

5) This wouldn't fix a damn thing - what Sprint needs is more capex going to an efficiently fast install of new towers and small cells. Why would buying a second, more redundant network with totally different tech help?

 

We have a greater chance of seeing Sprint go bankrupt than a merger happen because a merger of the two would be complete hell.

  • Like 3
Link to comment
Share on other sites

A merger would NEVER help Sprint for many reasons, and here's why:

 

1) The cost of adding up debts - if Sprint was to buy T-Mobile, or T-Mobile buy Sprint, they would have to combine their total debts. Think about it. They would have to go further in debt just to buy the other one, and Sprint has large debts to pay for soon.

2) The cost of new equipment - not only would they have to remove redundant cell sites, but they would need to get all new panels and radios and cabinets to support band 4 and 12 at Sprint sites, band 41 25 and 26 at T-Mobile sites.

3) Conflicting tech - this new carrier would have to find a way to bring VoLTE everywhere because of T-Mobile's existing VoLTE plans, but CDMA network subs wouldn't have access to VoLTE because no handoff unlike the UMTS subs.

4) Regulatory hell - you think FCC would let them get 600MHz or 5G frequencies with their total spectrum holdings?

5) This wouldn't fix a damn thing - what Sprint needs is more capex going to an efficiently fast install of new towers and small cells. Why would buying a second, more redundant network with totally different tech help?

 

We have a greater chance of seeing Sprint go bankrupt than a merger happen because a merger of the two would be complete hell.

A merger would help sprint. First the networks are converging to a single technology in LTE. Second, the larger debt is of set by scale and cost reductions of the new merged company. Wireless is a capital intensive business which implies higher returns to scale.

 

Sent from my LGLS992 using Tapatalk

  • Like 2
Link to comment
Share on other sites

VoLTE isn't really standardized to be uniform at this stage anyway. What we've had pop up has been a number of different VoLTE implementation profiles that are different than each other. I realize this will get worked out eventually, but the 3GPP and GSMA were asleep at the switch on this standard as far as making it a standard. That's why VoLTE support has lacked on a bunch of unlocked devices. 

 

3GPP compliant solutions that work on all carriers are really hard. People have to realize that. Unfortunately, I don't think a lot of people do realize that. 

 

Even on T-Mobile, which people will sometimes cite as the best carrier for unlocked devices, VoLTE support for unlocked devices is fairly limited. 

 

http://www.spectrumgateway.com/compatible-phones

 

Great.  There is the next frontier of complaints.  Sprint or any operator implements VoLTE, and we get to hear Joe Unlocked grousing, "Why won't VoLTE work on my Googorolawei OneMinus Nexus?  This is bullshit!"

 

 

AJ

  • Like 4
Link to comment
Share on other sites

 

 

Great. There is the next frontier of complaints. Sprint or any operator implements VoLTE, and we get to hear Joe Unlocked grousing, "Why won't VoLTE work on my Googorolawei OneMinus Nexus? This is bullshit!"

 

 

 

AJ

Next frontier? This has been happening on T-Mobile for the last year. It wouldn't be an issue if the 3GPP and GSMA weren't asleep at the switch. Unfortunately here we are.

 

Sent from my SM-N920V using Tapatalk

  • Like 1
Link to comment
Share on other sites

Next frontier? This has been happening on T-Mobile for the last year. It wouldn't be an issue if the 3GPP and GSMA weren't asleep at the switch. Unfortunately here we are.

 

I am referring primarily to Sprint.  You know that some here and elsewhere who are pining for Sprint to enable VoLTE for simultaneous voice/data immediately will turn into ingrates if they then cannot use VoLTE on certain unlocked handsets.

 

As for VoLTE, it seems like a solution in search of a problem.  Who needs it?  I have seen no evidence that VoLTE QoS is absolutely necessary.  By its own nature, LTE is a fragile airlink.  VoLTE QoS is unlikely to save a call from dropping if the LTE airlink becomes unstable.  Robustness comes not from QoS but from signal quality and capacity -- both of which are being achieved through densification.

 

Meanwhile, VoIP OTT has been around for years.  Maybe the operators should get out of the voice business altogether.  Various VoIP OTT dialers have almost ubiquitous compatibility and can figure out interoperability -- even if the latter means a lowest common denominator of dumping the call onto the PTSN most of the time.  Google already is doing that with the Hangouts Dialer, while Apple and Microsoft easily are capable of doing likewise.

 

Forget VoLTE.  Let the operators be just the "dumb pipes" that consumers by and large want them to be.

 

AJ

  • Like 4
Link to comment
Share on other sites

A merger would NEVER help Sprint for many reasons, and here's why:

 

1) The cost of adding up debts - if Sprint was to buy T-Mobile, or T-Mobile buy Sprint, they would have to combine their total debts. Think about it. They would have to go further in debt just to buy the other one, and Sprint has large debts to pay for soon.

2) The cost of new equipment - not only would they have to remove redundant cell sites, but they would need to get all new panels and radios and cabinets to support band 4 and 12 at Sprint sites, band 41 25 and 26 at T-Mobile sites.

3) Conflicting tech - this new carrier would have to find a way to bring VoLTE everywhere because of T-Mobile's existing VoLTE plans, but CDMA network subs wouldn't have access to VoLTE because no handoff unlike the UMTS subs.

4) Regulatory hell - you think FCC would let them get 600MHz or 5G frequencies with their total spectrum holdings?

5) This wouldn't fix a damn thing - what Sprint needs is more capex going to an efficiently fast install of new towers and small cells. Why would buying a second, more redundant network with totally different tech help?

 

We have a greater chance of seeing Sprint go bankrupt than a merger happen because a merger of the two would be complete hell.

 

Thing is a merger would assist SoftBank!

 

The new company would have a large user base nearing the big Two! Which equals income for wallstreet.

 

Sprint I'd assume would be using very very lucrative financial tools provided via SoftBank which would hide most of the debt from Wallstreet.

 

The larger company would be much more attractive for a future sale to a new owner and less of an issue for SoftBank if DT ever spun of TMO or if DT decided to go all in, in the US market.

 

Lastly the redundant technology issue is moot as I feel it can be used! Or sold!

 

Just my 2 cents.

 

 

 

 

Sent from my iPhone using Tapatalk

  • Like 1
Link to comment
Share on other sites

I'm really shocked Sprint came in 3rd overall. I had the biggest feeling they were going to either tie for 3rd with T-Mobile it fall to 4th place. How wrong I was.

 

 

Sent from my iPhone 6s+ using Tapatalk

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • large.unreadcontent.png.6ef00db54e758d06

  • gallery_1_23_9202.png

  • Posts

    • On Reddit, someone asked (skeptically) if the US Cellular buyout would result in better service.  I'd been pondering this very issue, and decided to cross-post my response here: I've been pondering the question in the title and I've come to the conclusion that the answer is that it's possible. Hear me out. Unlike some of the small carriers that work exclusively with one larger carrier, all three major carriers roam on US Cellular today in at least some areas, so far as I know. If that network ceases to exist, then the carriers would presumably want to recover those areas of lost service by building out natively. Thus, people in those areas who may only have service from US Cellular or from US Cellular and one other may gain competition from other carriers backfilling that loss. How likely is it? I'm not sure. But it's definitely feasible. Most notably, AT&T did their big roaming deal with US Cellular in support of FirstNet in places where they lacked native coverage. They can't just lose a huge chunk of coverage whole still making FirstNet happy; I suspect they'll have to build out and recover at least some of that area, if not most of it. So it'd be indirect, but I could imagine it. - Trip
    • Historically, T-Mobile has been the only carrier contracting with Crown Castle Solutions, at least in Brooklyn. I did a quick count of the ~35 nodes currently marked as "installed" and everything mapped appears to be T-Mobile. However, they have a macro sector pointed directly at this site and seem to continue relying on the older-style DAS nodes. Additionally, there's another Crown Castle Solutions node approved for construction just around the corner, well within range of their macro. I wouldn’t be surprised to see Verizon using a new vendor for their mmWave build, especially since the macro site directly behind this node lacks mmWave/CBRS deployment (limited to LTE plus C-Band). However, opting for a multi-carrier solution here seems unlikely unless another carrier has actually joined the build. This node is equidistant (about five blocks) between two AT&T macro sites, and there are no oDAS nodes deployed nearby. Although I'm not currently mapping AT&T, based on CellMapper, it appears to be right on cell edge for both sites. Regardless, it appears that whoever is deploying is planning for a significant build. There are eight Crown Castle Solutions nodes approved for construction in a 12-block by 2-block area.
    • Starlink (1900mhz) for T-Mobile, AST SpaceMobile (700mhz and 850mhz) for AT&T, GlobalStar (unknown frequency) for Apple, Iridium (unknown frequency) for Samsung, and AST SpaceMobile (850mhz) for Verizon only work on frequency bands the carrier has licensed nationwide.  These systems broadcast and listen on multiple frequencies at the same time in areas much wider than normal cellular market license areas.  They would struggle with only broadcasting certain frequencies only in certain markets so instead they require a nationwide license.  With the antennas that are included on the satellites, they have range of cellular band frequencies they support and can have different frequencies with different providers in each supported country.  The cellular bands in use are typically 5mhz x 5mhz bands (37.5mbps total for the entire cell) or smaller so they do not have a lot of data bandwidth for the satellite band covering a very large plot of land with potentially millions of customers in a single large cellular satellite cell.  I have heard that each of Starlink's cells sharing that bandwidth will cover 75 or more miles. Satellite cellular connectivity will be set to the lowest priority connection just before SOS service on supported mobile devices and is made available nationwide in supported countries.  The mobile device rules pushed by the provider decide when and where the device is allowed to connect to the satellite service and what services can be provided over that connection.  The satellite has a weak receiving antenna and is moving very quickly so any significant obstructions above your mobile device antenna could cause it not to work.  All the cellular satellite services are starting with texting only and some of them like Apple's solution only support a predefined set of text messages.  Eventually it is expected that a limited number of simultaneous voice calls (VoLTE) will run on these per satellite cell.  Any spare data will then be available as an extremely slow LTE data connection as it could potentially be shared by millions of people.  Satellite data from the way these are currently configured will likely never work well enough to use unless you are in a very remote location.
    • T-Mobile owns the PCS G-block across the contiguous U.S. so they can just use that spectrum to broadcast direct to cell. Ideally your phone would only connect to it in areas where there isn't any terrestrial service available.
  • Recently Browsing

    • No registered users viewing this page.
×
×
  • Create New...