New IrelandOffline blog post

IrelandOffline_

http://irelandoffline.org/2010/05/the-national-broadband-scheme-again/

The National Broadband Scheme. A Ghost Broadband System for our Ghost Estates.
3 Ireland have silently announced that they are unilaterally changing the terms of
the NBS contract. They will be dropping the data allowance for customers from
15GB a month of data down to 10GB of data.
This clearly demonstrates that the 3 Ireland network is not capable of handling
the data throughput required for the NBS and highlights in general the unsuitability
of any 3G network to handle the large volumes of data required for such a
network to be considered as real broadband.
In our opinion this is clearly a breach of contract and the NBS should be scrapped
with the €80m of public and EU spent on something more suitable to the goal of
delivering broadband to those that are currently not served by broadband.
Other key contracted road map events such as an increase in MINIMUM data
speeds for NBS users to 1.6mbits on the 1st of July 2010 and completion of
the entire roll out to the intended 388 masts (or their replacements)
by the end of September 2010 will not be met.
While there have been issues with planning authorities with regard to the
proposed 160 entirely new masts planned for this scheme (for which one
feels some sympathy for three) the fact is that fewer than half of the
160 new masts have been installed to date. In Kerry only 2 masts have
eceived permission out of the 10 applied for. This scheme is to be
installed and fully commissioned in less than 5 months. 3 are showing
no sense of urgency.
Contract changes are flagged here, please note that NBS customers are
considered to be “pay monthly”

http://www.three.ie/terms/customer_notification.htm
For the record here are the original terms as originally announced:
http://www.dcenr.gov.ie/Communications/Communications+Development/National+Broadband+Scheme/NBS+FAQs.htm#Limits
Will there be download and upload limits and what will they be?
An uncharged monthly data cap of 15GB (12GB download and 3GB upload)
will apply for the wireless product while 11GB (10GB download and 1GB
upload) will be available for satellite users.
Are there plans to upgrade speeds?
Two upgrades of the wireless product are planned by 3 in the coming
years without any increase in the monthly recurring charge:
July 2010 Min (at cell edge) Max (at cell centre)
Download Speed 1.6Mbps 6.8Mbps
Upload Speed 1.2Mbps 4Mbps
Maximum contention ratio of 22:1
Round-trip latency: 100ms
October 2012 Min (at cell edge) Max (at cell centre)
Download Speed 2.3Mbps 10.4Mbps
Upload Speed 1.4Mbps 4.8Mbps
Maximum contention ratio of 18:1
Round-trip latency: 100ms
Furthermore, since the inception of this scheme they have been
unable to maintain the initial minimum contracted download of
1.2Mbps (in common with all Mobile operators), 100ms Latency,
always on or always connected.
With W-CDMA/3G/UMTS/HSPA it’s not even possible to have a
guaranteed minimum connect speed at the cell edge.
This “distance” will vary depending on the number of simultaneous
connected users.
The maximum contention isn’t possible to achieve with mobile
subscribers. The industry standard is that contention
is the sum of package speed per user divided by the system
speed, as if all users connected. It’s not the real time
number of users per mast. The actual reality is that only a
certain percentage of customers connect at once,
thus on a real broadband system you might only see a real
time reduction of package speed from 8Mbps to
7.5Mbps due to real time contention. The 12:1, 24:1 or 48:1
contention rates quoted refer to the bandwidth
needed by total number of customers, not actual connections.
Three Themselves maintain that the commonly accepted distances
for 3g cell sites in varying environments are:
150m – 350m in urban areas
800m – 1000m in suburban areas
2km – 5km in rural areas
That shows clearly the density of cells required for adequate service
levels and also why the NBS can never possibly achieve its stated aims.
If there was a true representative and unbiased audit comparing 3G
system in areas where the customer sign-ups have reached a mature
stage, we suggest that the system would not ever have met either
the NBS specification or met the commonly accepted standards of Broadband.

clohamon

In Kerry only 2 masts have
received permission out of the 10 applied for.

I don't think that's correct.
Most of the Kerry applications were eventually granted with conditions on appeal to ABP. PM'd you on this.

Have H3GI confirmed that changes to cap includes NBS?

IrelandOffline_

clohamon wrote: »

I don't think that's correct.
Most of the Kerry applications were eventually granted with conditions on appeal to ABP. PM'd you on this.

Have H3GI confirmed that changes to cap includes NBS?

Thanks for clarifying that about the masts...

3 have neither confirmed or denied so we are attempting to get clarification on that point.

One of the most important points (probably missed in the post) was confirmation of the size of the cells required and that cell shrinking also was confirmed by 3.

To cover an area the size of Kerry would require cell sites every 5 km rather than 10 or so cell sites as granted. This is an important point for coverage in that particular area. It shows that the number of cell sites required to deliver any sort of reasonable midband coverage is enormous.

watty

That would really at that be only voice coverage.

All cell density planning is really based on ensuring a certain high pecentage of voice call attempts succeed. To have the same level of Service for 1.2Mbps would require many more masts. Basically one for approx. every 60 customers assuming about 18:1 contention.

clohamon

IrelandOffline_ wrote: »

One of the most important points (probably missed in the post) was confirmation of the size of the cells required and that cell shrinking also was confirmed by 3.

To cover an area the size of Kerry would require cell sites every 5 km rather than 10 or so cell sites as granted. This is an important point for coverage in that particular area. It shows that the number of cell sites required to deliver any sort of reasonable midband coverage is enormous.

That point was well made thanks, but regarding the possible delay.....

It is not clear whether the 92% hspa/ 8% satellite split applies at the ED, County or National level.

DCENR

In recognition of the fact that some areas will be very costly and difficult to reach, in a very limited number of cases, 3 will make available a satellite product of 1Mbps download and 128kbps upload. This will cover up to a maximum of 8% of fixed residences and businesses in the NBS coverage area.

H3GI

In recognition of the fact that some areas will be very difficult to reach using standard infrastructure, in a limited number of cases 3 will make available a satellite product, which may cover up to 8% of the NBS areas.

If its at national level then some EDs or even whole counties might be served predominantly by satellite so long as the correct HSPA/satellite proportions are met at national level.

AFIK the scheme no longer claims to be ubiquitous in a mobile sense but only has to serve those who request it, within the NBS area; ie its not actually an area based scheme, its a premises based scheme. What's more its also a demand based scheme, so the theoretical catchment is not really relevant. What's important is how many people (premises) actually apply for it and where they are.


If the take up was sufficiently low in remote areas it may not be necessary to build masts at all and still meet the hspa/satellite ratio for the NBS as a whole. Even completing the planning applications could prove extremely wasteful if demand ultimately turned out to be low. The point is, that if the stated HSPA/satellite ratio is not enforced at the ED level, there is every incentive (late fines aside) to hold off building expensive masts in the remote areas until the very last moment.

Sponge Bob

clohamon wrote: »

It is not clear whether the 92% hspa/ 8% satellite split applies at the ED, County or National level..

It was quite clear to Damien Gallagher of 3 who is the rollout manager, it was per ED.

No ED can go live where 92% of the population cannot receive a terrestrial signal. No more than 8% in any given ED can be on satellite.

watty

If the revenue from voice calls + mast subsidy in an ED doesn't look like that over 5 years would exceed the cost, why would they build?

I've never thought they would do all ED.

They don't stand to lose much.

In fact if they got kicked out of NBS after they do all the masts they are prepared to do, it would suit them fine.

Sponge Bob

watty wrote: »

If the revenue from voice calls + mast subsidy in an ED doesn't look like that over 5 years would exceed the cost, why would they build?

As they have the only A licence they should make a point in staying ahead in 3g coverage...after all what about the spectrum under 1ghz to be auctioned soon.

I've never thought they would do all ED.

I agree, they will abandon c10% of all EDs on grounds of planning difficulty ...they need but apply in sacs and spas

So they pass on €8m of cash, so what.

watty

€8m is nothing if you are saving €20m