Sprint TD-LTE 2500/2600mhz Discussion

[digiblur]

Eh. I am not particularly overjoyed with this development. I am not sure how deploying 2500 where it is not needed is enhancing the customer experience. It does however increase costs. I would rather see a targeted strategic use of 2500 similar to what was originally planned.

It is cool to see additional progress but for me personally I feel the same. I have a hard enough time finding some 1900 LTE to use indoors. So it will probably be a couple years before I see it and get to use it.

[S4GRU]

with the amount of money being poured into 2500 development and with the requirement that all 2014 devices will support it , i think 2500 will become the main LTE band for Sprint and 1900 and 800 will be fallback when 2500 signal cannot reach customer. 

 

Definitely.  And this is the ideal network design.  Without purchasing Clearwire, it wasn't possible.  They were going to do a 1900>2600>800 design when they were going to have to pay per usage fees to Clearwire.

 

Robert

[S4GRU]

This is just my opinion but I really don't believe that every device that Sprint has will eventually be tri band. I expect to see it on the high level and perhaps some of the mid level devices. Additionally, you don't have to have a triband device to passivley benefit from 800 and 2500(offloading).

 

They said today that all devices in 2014 will be triband and that a few smartphones in Fall 2013 will have it.

 

Robert

[S4GRU]

You may be correct on this and it may be the exact best way to do things.  As long as they can pull off a smooth hand-off to 1900 or 800 without upsetting the data stream, we may really have something great. I question if they can make it work when driving at 75 MPH on the interstates since constant handing off would be required.

 

Since data doesn't have to be a smooth handoff, I don't think this will be a problem.

 

Robert

[centermedic]

It will provide ample capacity.  It will also provide superior speeds over the rest of the Sprint LTE network.  It will also provide equal or superior speeds to all competitors.  And if a TD-LTE carrier becomes overburdened, adding additional LTE capacity will be a snap with just an additional carrier card needing to be added.  It's not wasted money, but it is investing more money and capacity into the network than Sprint has ever done before.

 

And now it allows all those international travelers who may only have one U.S. LTE band on their phone to have a somewhat cohesive nationwide network.  If you can use TD-LTE 2600 everywhere, and only use LTE 1900 and 800 to fall back on, then that is a very good thing.  This is a 180 degree turn about from what the previous Sprint would ever have done.  This decision is courtesy directly from our new Japanese Overlords.  It's about time the worst decisions the company makes are pro-network experience!!!

 

The Japanese are not the type to think in the present, or near future.  They look much farther out then we tend to do.  It's a refreshing change, even if slightly uncomfortable to what we are used to.

 

Robert 

Hmm, I forgot about international travellers. That reasoning makes sense.

[dmaxx67]

I wonder if they will release a timeline for this part of the network to be completed? Or leave that unknown and just build out as they can.

[S4GRU]

It is cool to see additional progress but for me personally I feel the same. I have a hard enough time finding some 1900 LTE to use indoors. So it will probably be a couple years before I see it and get to use it.

 

I don't think Softbank will run things anywhere like Sprint.  I don't think Masayoshi Son takes no for an answer as easily as Dan Hesse.  Just my thoughts.  But the future should be nothing like the past.

 

Robert

[chamb]

Since data doesn't have to be a smooth handoff, I don't think this will be a problem.

 

Robert

Sure, a slight blip can be overcome by error correction and resending a block or two of data.    But any significant drop in the stream will surely bother video.

[dmaxx67]

I like how there was no comment on the Iphone.

[WiWavelength]

But any significant drop in the stream will surely bother video.

 

Streaming media is well buffered.

 

AJ

[S4GRU]

I wonder if they will release a timeline for this part of the network to be completed? Or leave that unknown and just build out as they can.

 

I think we will see the LTE overlay over the existing Clearwire network get expedited and may even complete in six months.  As for adding TD-LTE to the entire Sprint Network Vision network, I expect that to be more of a 12-14 month program.  They may get to a point in a few months that all new NV LTE conversions go live with LTE 800/1900/2600 all at once.

 

Robert

[S4GRU]

Streaming media is well buffered.

 

AJ

 

I was just about to respond with that point.  

 

EDIT:  I guess video chat could be an issue.  But if you are using video chat while being a mobile passenger, there are probably issues on all networks.

 

Robert

[chamb]

I was just about to respond with that point.  

 

EDIT:  I guess video chat could be an issue.  But if you are using video chat while being a mobile passenger, there are probably issues on all networks.

 

Robert

Yes, there is buffering, but how much???   Video does not wait around for any software to fix an issue. When a stream stops, normally the video burps and you can see it.   If  any significant amount of the stream is lost, the video is going to show it.

[WiWavelength]

My prediction for idle state tri band LTE UEs is that network priority will go LTE 1900 > TD-LTE 2600 > LTE 800.  EARFCN 8665/26665 LTE 1900 in the PCS G block will be the same center frequency and carrier bandwidth everywhere across the country.  TD-LTE may always be the same carrier bandwidth, but it will likely vary in center frequency.  And LTE 800 will be a crapshoot -- different center frequency, different carrier bandwidth, or even not available -- per market.

 

AJ

[dmaxx67]

I think we will see the LTE overlay over the existing Clearwire network get expedited and may even complete in six months.  As for adding TD-LTE to the entire Sprint Network Vision network, I expect that to be more of a 12-14 month program.  They may get to a point in a few months that all new NV LTE conversions go live with LTE 800/1900/2600 all at once.

 

Robert

Completing all at once does sound good especially keeping contractors from visiting same site over and over again. So are we thinking maybe by end of 2015 we should have a damn near complete network? That ought to give them some time to account for all the stuff they cant control to finish a site

[WiWavelength]

Yes, there is buffering, but how much???

 

Plenty.  Usually several seconds worth of video.

 

AJ

[S4GRU]

Yes, there is buffering, but how much???   Video does not wait around for any software to fix an issue. When a stream stops, normally the video burps and you can see it.   If  any significant amount of the stream is lost, the video is going to show it.

 

A hard handoff is typically a fraction of a second to a full second.  Most videos buffer seconds, if not even minutes ahead.  It depends on the length of what your watching and the speed of the connection.  This is not a concern of mine, really.  It really would only be an issue if you handed off while it just started and there is not much cached ahead of your current reference point.

 

Robert

[Dkoellerwx]

Completing all at once does sound good especially keeping contractors from visiting same site over and over again. So are we thinking maybe by end of 2015 we should have a damn near complete network? That ought to give them some time to account for all the stuff they cant control to finish a site

 

The way it sounds, it seems like they're aiming for end of 2014, not 15.

[S4GRU]

My prediction for idle state tri band LTE UEs is that network priority will go LTE 1900 > TD-LTE 2600 > LTE 800.  EARFCN 8665/26665 LTE 1900 in the PCS G block will be the same center frequency and carrier bandwidth everywhere across the country.  TD-LTE may always be the same carrier bandwidth, but it will likely vary in center frequency.  And LTE 800 will be a crapshoot -- different center frequency, different carrier bandwidth, or even not available -- per market.

 

AJ

 

Idle state, it seems reasonable to park on 1900 over 2600 because of the common nationwide frequency.  Although, do you think they may park on 800 where possible, so they get the benefit of increased bars showing?

 

Robert

[dmaxx67]

Yea i always give extra time. I am a mechanic and I have learned to give yourself the extra time so your not finishing up when the customer needs their equipment. Also helps when you run into issues not seen. Instead of having a customer looking at you asking why isn't my truck ready.

[Boosted20V]

Well if BRS goes nationwide at least we know GMOs will be taken care of.

[dmaxx67]

My prediction for idle state tri band LTE UEs is that network priority will go LTE 1900 > TD-LTE 2600 > LTE 800.  EARFCN 8665/26665 LTE 1900 in the PCS G block will be the same center frequency and carrier bandwidth everywhere across the country.  TD-LTE may always be the same carrier bandwidth, but it will likely vary in center frequency.  And LTE 800 will be a crapshoot -- different center frequency, different carrier bandwidth, or even not available -- per market.

 

AJ

I would like it to be 800>1900>2600. Off load to next band as each one gets loaded to ALMOST full Capacity. But ya are they experts and are probably right in how they will handle this 

[chamb]

A hard handoff is typically a fraction of a second to a full second.  Most videos buffer seconds, if not even minutes ahead.  It depends on the length of what your watching and the speed of the connection.  This is not a concern of mine, really.  It really would only be an issue if you handed off while it just started and there is not much cached ahead of your current reference point.

 

Robert

If you really can buffer several seconds, then you might be ok. But video conferencing may not work well with it having excessive delay. One way video may be fine but two way may not work properly if we get into heavy buffering.  I am asking about this more than trying to say I have the answer.  Airaves(remember them) had a delay just on voice and it made them hard to use.

[chamb]

Idle state, it seems reasonable to park on 1900 over 2600 because of the common nationwide frequency.  Although, do you think they may park on 800 where possible, so they get the benefit of increased bars showing?

 

Robert

Never underestimate the value of having a whole pile of bars at the top of the screen.  For years, Sanyo had an extra bar over the other brands not because of being a better phone.  They simply tweaked the setting to make it appear that you had more signal.

People actually believed they had better service when they actually only had a false extra bar.

[WiWavelength]

Never underestimate the value of having a whole pile of bars at the top of the screen.

 

CDMA1X 800 will take care of that.

 

AJ