BT Infinity

[apb27]

http://img2.pict.com/fb/fa/9e/3360320/0/sshot1.png


Why doesn't Virgin increase its upload speed, i do love 50mb
but even 40mb with 10mb upload would be amazing.

The BT offer above looks very appealing, cheaper, better upload
and free install, but i dont think i would get the same speeds as
i get on Virgin. Maybe in the future it could be an option

Any news on future products or even current trials?

[Ignitionnet]

A fair bit, and all just a forum search away.

http://www.cableforum.co.uk/board/search.php?searchid=3611468

Is a search for 'upload trial'

[broadbandking]

Just wondering how close to the 40Mb down and 10Mb would you get on BT's infinity service.

[Ignitionnet]

Just wondering how close to the 40Mb down and 10Mb would you get on BT's infinity service.

The speed tests I've seen have been 36-38 down and 9+ up.

[|Kippa|]

Could VM afford not to give us 10mbit now we have some competition with BT?

[Tony.]

Upload is still not a Headline grabber. Perhaps when it is VM will play catch-up

[Ignitionnet]

Could VM afford not to give us 10mbit now we have some competition with BT?

I would speculate they won't supply 10Mbit up until 100Mbit release. They can afford to do this if they stick to their schedule as BT's FTTC isn't going to be very widespread until January - March 2011, by which time VM's 100Mbit rollout will be in progress.

[Sephiroth]

I would speculate they won't supply 10Mbit up until 100Mbit release. They can afford to do this if they stick to their schedule as BT's FTTC isn't going to be very widespread until January - March 2011, by which time VM's 100Mbit rollout will be in progress.
How would VM deal with the upstream channel issue?

[telfordcable]

virgin will never get more than 1.6mbps upload if customer want 10mbps upload, then vm will put the price up

[sollp]

virgin will never get more than 1.6mbps upload if customer want 10mbps upload, then vm will put the price up

"virgin will never get more than 1.6mbs upload"!! Well,they can and will.

If VM was to increase the price as you say, do you think it would be a bad thing? Surley any business has to look at it's cost's and can't always be giving away a service?

Then again they might not increase the subscription charges, who knows?

[Ignitionnet]

How would VM deal with the upstream channel issue?

Could take another channel from the legacy network and either bond them or run them as a single 6.4MHz channel. Could also clean up network some more and/or use better upstream PHY migrating everything to DOCSIS 2 and use previously unusable frequencies for SCDMA / 16QAM TDMA to add additional capacity.

Reducing node sizes or upstream service groups further will have the effect of cleaning up networks, at present a number of overlay upstreams span more than one node.

These would produce considerably more capacity in the same RF space.

[renty]

hi, my brother has bt infininty installed and he gets 9mb up and 39mb down and its stable ...he used to get 3 mb on old connection....this may answer your question on speeds...

[Sephiroth]

Could take another channel from the legacy network and either bond them or run them as a single 6.4MHz channel. Could also clean up network some more and/or use better upstream PHY migrating everything to DOCSIS 2 and use previously unusable frequencies for SCDMA / 16QAM TDMA to add additional capacity.

Reducing node sizes or upstream service groups further will have the effect of cleaning up networks, at present a number of overlay upstreams span more than one node.

These would produce considerably more capacity in the same RF space.

You see, what's being demanded is a 6 fold increase in upstream speed as compared with the 1.5 Mbps users now. At best (fag packet calculation), you could double upstream, maybe treble it by the methods you've mooted.

I see it as an infrastucture thing once they've got the vacated RF bandwidth to add to the plan. They need more cables in the street providing upstream and more fibre to the cabinets to ensure that the cabinet doesn't become the bottleneck (or again more cable to the MUX).

[Ignitionnet]

They need more cables in the street providing upstream and more fibre to the cabinets to ensure that the cabinet doesn't become the bottleneck (or again more cable to the MUX).

What MUX? What cables in the street providing upstream?

I mentioned more fibre to cabinets - reducing size of service groups / node splitting.

I'm very unsure what you're trying to say here, it makes very little sense.

Take a single service group with a single upstream covering 2 nodes (pretty common), split the nodes by removing the RP combining = doubling of upstream capacity, then reclaim another 3.2MHz of bandwidth and bond those 2 channels, another doubling, and take advantage of the improved cleanliness of the nodes through the split by using 64QAM = another 50% increase.

9Mbit * 2 for 3.2MHz reclaimation = 18Mbit, * 1.5 for 16->64QAM = 27Mbit.

Node split = 27Mbit between half as many modems = 6-fold increase in upstream capacity per modem.

It works just fine so long as the nodes are small enough, Cablevision are running 5Mbit and 15Mbit upstream modems on a single 27Mbit channel, Comcast are running 5Mbit and 10Mbit upstream customers on a single 27Mbit channel. No need to increase bandwidth to an area by 6-fold, splitting that area in half and increasing the bandwidth to each area 3-fold has the same effect.

[Stuart]

virgin will never get more than 1.6mbps upload if customer want 10mbps upload, then vm will put the price up

You seem sure they won't.

I suspect Virgin will match this, when BT have rolled it out to enough areas that Virgin can gauge potential demand.

Remember, BT are currently their ONLY major competitor in this, and BT have only rolled it out to a couple of areas. Upload speeds do not currently grab headlines, and VM don't have any competitors that can beat the headline speed for 50meg (whether the 50 meg speed is maintained in real life is not really relevant as they don't advertised the real world speed).

BT will not beat VM on the headline speed at the moment, but there are several ISPs that have expressed an interest in using the Infinity network (O2 included IIRC), but we don't know what speeds they are planning.

[Ignitionnet]

virgin will never get more than 1.6mbps upload if customer want 10mbps upload, then vm will put the price up

Paying more for more bandwidth, oh my the horror!

I'm ignoring there the obvious contradiction in your waffle, that they'll never offer more than 1.6Mbps but will charge more for 10 if customers want it. Odds on it'll be coming with the 100Mbps.

Shoo.

[jtaylor06]

Upload will be more wanted soon as many people now upload to sites such as YouTube, or even the use of Virgin's V Stuff thingie.

At least trials are happening, and it shouldn't be too long now as the news was some time ago.

As for telford cable, you sure do give my home town a bad name!

[Sephiroth]

What MUX? What cables in the street providing upstream?

.......

The ones (cables) that don't exist. We've been over this before. It's a wish list. If they had active taps, upstream could be filtered to a second cable which would be a trunk serving fewer houses than the downstream. This can be reproduced back to the MUX where cable is remodulated to fibre.

They have passive taps as you've said before so they're not gonna do anything as sensible as what I'm suggesting - but they could do this in new builds. It looks forward, adding to the mechanisms they could deploy as per your description which have limits.

[Ignitionnet]

The ones (cables) that don't exist. We've been over this before. It's a wish list. If they had active taps, upstream could be filtered to a second cable which would be a trunk serving fewer houses than the downstream. This can be reproduced back to the MUX where cable is remodulated to fibre.

They have passive taps as you've said before so they're not gonna do anything as sensible as what I'm suggesting - but they could do this in new builds. It looks forward, adding to the mechanisms they could deploy as per your description which have limits.

I see what you're saying now and think you misunderstand the structure of the network. Why would one employ a fibre solely for upstream and connecting to an expensive tap, which doesn't actually exist for very good reasons, why not just split the node and reduce the service group in both directions?

Pointless having upstreams focussed so closely. There's no need at all to dedicate a logical node, and therefore a piece of fibre or a wavelength on WDM kit, to 12 homes.

Upstreams can serve fewer homes than downstreams anyway by splitting the downstream across multiple nodes while mapping upstream service groups 1:1 to node. If downstream is ok one can simply split the node physically and then split a single downstream or bonded group across both the new nodes while running separate upstream service groups - the two nodes are logically a single node downstream but two nodes upstream.

If you start devoting fibre / wavelengths to 8-12 homes you are running fibre deeper than a PON network which devotes a single fibre / wavelength from the headend to each 32 or 64 homes. There's really no need at all to supply so much bandwidth per home passed and the costs would be immense. It would very simply be pointless and would be more expensive than an FTTP overbuild.

Fibre deep solutions are the endgame as far as HFC goes with solutions such as:

http://www.aurora.com/site/applications.an?li=a-fd
http://www.ciscosystems.com/en/US/netsol/ns903/networking_solutions_solution.html
http://www.motorola.com/Business/US-EN/Business+Solutions/Industry+Solutions/Cable+Operators/Broadband+Cable+Networks/Fiber+Deep_US-EN

The next step from this is RFoG (http://en.wikipedia.org/wiki/Radio_Frequency_over_Glass).

In the interim the solutions I mentioned will work and will require nothing more than return path sweeping, channel bonding and node splitting, either through new fibre or WDM, in 99% of cases. Far more attractive to VM than rebuilding areas and gets people their higher upstream services running viably.