Hawaiki Cable

Forums › New Zealand Broadband › Hawaiki Cable

1 | 2 | 3 | 4 | 5 | 6 | 7

943 posts

Ultimate Geek
+1 received by user: 133

UberGroup

Reply # 685913 14-Sep-2012 10:07

So where's this price model for an ISP up here? I know I've chucked some numbers around.

No, Dont land a cable in Whangarei, Not even a branch - It's quite stupid. Northland has ~50k of homes or around 150k of people. It's tiny and building a branch here is costly along with offering no diversity since we have no power generation up here and need Auckland for power/comms/everything.

As for landing anywhere else if you cant make the case for a 2nd cable coming into NZ at all why is there a case for landing it in the south island or lower north island? There isn't.

National transit costs's aren't too bad either - Certainly not MORE than INT unless you're talking about delivery to bluff. I can pick up INT @ $35/mbit in AKL and transit to Whangarei at $4/mbit or NAT Transit WhangareiAucklandWelligton @$13 - And thats at 1gbit buys, You wanna haul 10gbit around and it gets cheaper again.

Also Don, Cables normally dont have 50-100 year lifespans, More like 20-30 so that $75mill spend now? Gonna have to go it again before you even kick the bucket and you dont even factor in O&M cost's like having a crew ready to repair it should some farmer dig it up at 5am. O&M over the lifespan of a cable can be upwards of 25% of the cable build cost

Most problems are the result of previous solutions...

All comment's I make are my own personal opinion and do not in any way, shape or form reflect the views of current or former employers unless specifically stated

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 685920 14-Sep-2012 10:38

Regs: Some probably don't even know what it is.....

Please help Regs everyone... he's trapped in a time warp in 1995.

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 685921 14-Sep-2012 10:41

SaltyNZ: The Internet is everywhere. It is all around us. Even now, in this very room. You can see it when you look out your window or when you turn on your television. You can feel it when you go to work... when you go to church... when you pay your taxes. It is the world that has been pulled over your eyes to blind you from the truth.

Now that's the attitude there!!! Nice.

Mobile companies sure have made sure we're immersed in 1's, that's for sure.

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 685995 14-Sep-2012 13:21

Beccara: So where's this price model for an ISP up here? I know I've chucked some numbers around.

No, Dont land a cable in Whangarei, Not even a branch - It's quite stupid. Northland has ~50k of homes or around 150k of people. It's tiny and building a branch here is costly along with offering no diversity since we have no power generation up here and need Auckland for power/comms/everything.

Power diversity for landing stations generally comes from UPS and a local generator. You jut need to hold a stock of fuel on site.

While I agree Northland had ~50k homes now, what will it have in 20 years?

Beccara: As for landing anywhere else if you cant make the case for a 2nd cable coming into NZ at all why is there a case for landing it in the south island or lower north island? There isn't.

Ok, so we're getting a bit back on topic here looking at the business case for the Hawaiki cable.

I haven't looked at a detailed business case for a cable at all. PF said they had a business case to just over build SCCN and failed. Hawaiki seem to think they have a business case to over build most of SCCN and bring in some islands with less population than you have in Northland.

So it strikes me that it would be worth having a much closer look at the business cases for these projects.

Beccara: National transit costs's aren't too bad either - Certainly not MORE than INT unless you're talking about delivery to bluff. I can pick up INT @ $35/mbit in AKL and transit to Whangarei at $4/mbit or NAT Transit WhangareiAucklandWelligton @$13 - And thats at 1gbit buys, You wanna haul 10gbit around and it gets cheaper again.

Ok, so some rough coffee cup math... $13/mbit to Wellington per month and a video stream needs at least 4 and we can assume 2 TV's running in most homes at peek time in the evenings these days. So that's 8 * $13 of transit per home per month before you add the UFB tail at $37 per month. That's $141 per month before you even get started.

National transit needs to get down to a fraction of that, say 50 cents per mbit per month before this stuff starts to make sense in my head.

I haven't done the math, but I can't believe that in a UFB world there won't be savings to be made by cutting out that unrequired transist if you just land a cable in the south.

Beccara: Also Don, Cables normally dont have 50-100 year lifespans, More like 20-30 so that $75mill spend now? Gonna have to go it again before you even kick the bucket and you dont even factor in O&M cost's like having a crew ready to repair it should some farmer dig it up at 5am. O&M over the lifespan of a cable can be upwards of 25% of the cable build cost

Yes, the 25% was way I was talking in terms of $100m, that and because I know landing in Hobart would be more expensive than Sydney.

However, a lot of the work you do for this run would hold value for 100 years or more. You wouldn't need to completely redo everything in 30 years. A new cable, or cable parts could be laid, but on the same path.

A lot of the cost is in survey runs, planning and permissions.

I'm no expert in this, just have a very keen interests, but it strikes me that we should be looking at funding a business case to actually look at all the numbers properly.

I would like to see New Zealanders at least get the same Internet value in New Zealand that the Australians get over there.

We don't and our biggest providers don't actually seem as committed to bringing us equavlialant value as just delivering shareholder profit.

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

SaltyNZ

3907 posts

Uber Geek
+1 received by user: 1499

Trusted

Subscriber

Reply # 685999 14-Sep-2012 13:26

DonGould:

While I agree Northland had ~50k homes now, what will it have in 20 years?

25k homes?

iPad Air + iPhone SE + 2degrees 4tw!

These comments are my own and do not represent the opinions of 2degrees.

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 686015 14-Sep-2012 13:41

SaltyNZ:
DonGould:

While I agree Northland had ~50k homes now, what will it have in 20 years?

25k homes?

Why? Because the rates are so high after putting in a new poo plant?

Are you suggesting that putting in a new internet feed would drive even more people away?

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

SaltyNZ

3907 posts

Uber Geek
+1 received by user: 1499

Trusted

Subscriber

Reply # 686020 14-Sep-2012 13:50

DonGould:

Why? Because the rates are so high after putting in a new poo plant?

Bah! Don't talk to me about ridiculous rates bills until you've lived in Rodney Hyde's glorious SuperCity.

Full disclosure: I live in the part of Rodney that campaigned REALLY HARD to NOT be in the SuperCity.

Also I'm a little bit bitter and twisted because I live 500m outside the cabinet zone, and my exchange is not scheduled to ever be upgraded past ADSL 1 (even though I am less than 2000m from it and get 8M/960k) and although the main Telecom trunk fibre runs past the end of my driveway I will never get UFB...

Are you suggesting that putting in a new internet feed would drive even more people away?

Well, it might drive away those who choose to live in Northland because they hate all those evil modern conveniences, but mostly I think population will further consolidate in the cities.

iPad Air + iPhone SE + 2degrees 4tw!

These comments are my own and do not represent the opinions of 2degrees.

oxnsox

1923 posts

Uber Geek
+1 received by user: 139

Reply # 686054 14-Sep-2012 15:11

Beccara:As for landing anywhere else if you cant make the case for a 2nd cable coming into NZ at all why is there a case for landing it in the south island or lower north island? There isn't.

One musn't let the corporate accountants and politicaids dictate what works best..... we'll all end up living in Auckland, stuck in traffic, watching new volcanoes erupt.

It would be more sensical to run a secondary transtasman cable from Melbourne to Nelson (with a split to Kapiti Coast/Wgtn) where there is a reliable power source and an ability to offer improved services to the other half of the countries population as well as a viable alternate (diversity) path for folk at each end. Might help grow economies in both regions too.

But. If we waited for everything to be commercially valid think about the infrastructure we have now that we never would have had. (S.I. electricity in the N.I. is one thing that springs to mind.)

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 686062 14-Sep-2012 15:37

SaltyNZ: Also I'm a little bit bitter and twisted because I live 500m outside the cabinet zone, and my exchange is not scheduled to ever be upgraded past ADSL 1 (even though I am less than 2000m from it and get 8M/960k) and although the main Telecom trunk fibre runs past the end of my driveway I will never get UFB...

See this is even more argument as far as I'm concerned for landing a cable in Christchurch.

If we can take 2/3rd's of the .au market from SCCN then won't that cause more drive in Auckland to actually work harder to get more capacity delivered to every Aucklanders and not just the ones that are easy to hit from the Sky Tower?

D

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 686072 14-Sep-2012 15:53

oxnsox: It would be more sensical to run a secondary transtasman cable from Melbourne to Nelson (with a split to Kapiti Coast/Wgtn)

What is the capacity out of Nelson south like though?

Would that not be just putting more traffic on the North --> South leg?

It also leaves a long national transit route to Dunedin and Invercargill.

My whole thinking what simply to balance out the traffic flows up and down the country or am I not making sense?

Also how many friends do we loose if we drop into Kapiti as well as Nelson?

Who owns the Cook Straight transit?

In suggesting Christchurch I've been trying to think in terms of:

* Opening up new markets
* Using under used back capacity
* Not proposing something that takes to much business off everyone and over builds stuff that's already got free capacity but focus on delivering more capacity to then empower others to focus on service and product delivery that raises APRU.

Nelson and Kapiti appear to fail.

Also landing in Melbourne side steps Tasmania, why did you want to do that? It would put Tasmania 'on the way' to some where else, which has advantages as I see it.

D

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

mercutio

1375 posts

Uber Geek
+1 received by user: 131

Reply # 686076 14-Sep-2012 16:08

i think south america is a more worthwhile target as their economy is closer to ours.

australia has no good reason to give us a good deal. and without competition prices will stay high.

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 686086 14-Sep-2012 16:42

mercutio: i think south america is a more worthwhile target as their economy is closer to ours.

australia has no good reason to give us a good deal. and without competition prices will stay high.

Australians are building data centers like we build dairy farms. They will have CPU cycles to burn and doing nothing while they're asleep.

Cable Cost - $       100,000,000
mbits on cable -             8,000,000
Cost per mbit over cable life -12.5
Cable life - 25
Cost per mbit over cable life per month - $                  0.04

Just simple math here again... but...

Beccara: I can pick up INT @ $35/mbit in AKL and transit to Whangarei at $4/mbit or NAT Transit WhangareiAucklandWelligton @$13 - And thats at 1gbit buys, You wanna haul 10gbit around and it gets cheaper again.

If we assume that a 2 pair cable can be loaded to 8Tb with todays technology (remembering that SCCN started out at 240Gbit on 10Gbit interfaces and now I think it's at 40Gbit interfaces in less than a decade?)

If we assume a build cost to Hobart at $100m (let's just put aside inflation and the home loan calculator mentioned earlier).

Then we get a cost per mbit over the life of the cable, assuming no upgrades to quadruple the capacity as happened with SCCN, of $12.50.

If we then accept that the cable life is only 25 years (which we know is just silly but let's just go there)...

You're at 4c per mbit.

If we accept Beccara's comment that operational costs will be 25% then we're at 5c per mbit per month.

Come on...

$4 from Auckland to Whangarei, $13 from Whangarei to Wellington?!

This is why we're not getting the same value that the Australians get and it's why I think we should be building directly to Australia.

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

Beccara

943 posts

Ultimate Geek
+1 received by user: 133

UberGroup

Reply # 686088 14-Sep-2012 16:52

Don, You bring this up in almost every thread - A link from the south island to AU simply doesn't make sense. To land it you either need to run via cook straight to get it to a major population base in which case you might aswell land it in Wellington OR you have to get the cable over the southern alps and run the last 200k's over land. Both are simply stupid idea to try and drive down cost for 25% of the population that seem to be behind a "F#$@ You, Got Mine" attitude

I doubt you'll see transit prices inside NZ @ $0.5/mbit anytime before 100gb per wavelength tech has paid off the R&D and it's price drop's (Around the same time 10gbit ethernet will move into the SOHO market), Alot of firms are putting off migrations to 40gbit/wave tech because of the cost and 100gb/wave so we'll all be stuck with 10gbit/wave for a little longer. Notice how bandwidth and use always move up together? By the time we see 100gbit/wave deployed by the big 3 operators in NZ we'll be complain then that $0.5/mbit is too costly for our 3D holographic movie streaming and it needs to be $0.005/mbit. No doubt you'll still be waiting for this CHCH cable then aswell.

You then want to make this case for investing into this cable by taking away traffic from SCCN to AU, The cheapest section of the cable for SCCN where clients can pick up transit from under $6/mbit TODAY (SCCN normally drops ~20% per year too), A cable that best estimates is around $100mill USD to BUILD more if it was to be protected. These are some of the many reasons why PF failed

This cable MIGHT work if someone's willing to drop some cash on it for the good PR of fast access to remote pacific islands.

The example of the TV's needing $141/mth of transit is off for many reasons as well but mainly because ISP's dont work like that, It's a residential service so that 8mbit they need will be sold to a business connection that needs it during 9-5 only along with anyone else who the ISP could sell off-peak fast bulk data connections too. It's also completely ignoring where the data comes from, If it's nextflix for the next few years thats Akamai and Limelight, Youtube? Google caching.

There's no silver bullet to the issues that UFB will present, It will be a mix of caching at UFB/RBI/xDSL HO points around the country, Local IX's, Smarter Int transit buy's, geographically staggered caching etc etc

I really dont see this cable getting off the ground considering most of the people behind it are the same ones behind other failed cables, Without some handout's it's hard pressed to get clients since it's likely going to be higher latency between NZAU and NZUSA because of the small islands it's visiting. The demand simply isn't there for another cable - Any demand for it comes from camp's who want it at prices that don't stack up

Beccara

943 posts

Ultimate Geek
+1 received by user: 133

UberGroup

Reply # 686091 14-Sep-2012 16:55

DonGould:
mercutio: i think south america is a more worthwhile target as their economy is closer to ours.

australia has no good reason to give us a good deal. and without competition prices will stay high.

Australians are building data centers like we build dairy farms. They will have CPU cycles to burn and doing nothing while they're asleep.

Cable Cost - $       100,000,000
mbits on cable -             8,000,000
Cost per mbit over cable life -12.5
Cable life - 25
Cost per mbit over cable life per month - $                  0.04

Just simple math here again... but...

Beccara: I can pick up INT @ $35/mbit in AKL and transit to Whangarei at $4/mbit or NAT Transit WhangareiAucklandWelligton @$13 - And thats at 1gbit buys, You wanna haul 10gbit around and it gets cheaper again.

If we assume that a 2 pair cable can be loaded to 8Tb with todays technology (remembering that SCCN started out at 240Gbit on 10Gbit interfaces and now I think it's at 40Gbit interfaces in less than a decade?)

If we assume a build cost to Hobart at $100m (let's just put aside inflation and the home loan calculator mentioned earlier).

Then we get a cost per mbit over the life of the cable, assuming no upgrades to quadruple the capacity as happened with SCCN, of $12.50.

If we then accept that the cable life is only 25 years (which we know is just silly but let's just go there)...

You're at 4c per mbit.

If we accept Beccara's comment that operational costs will be 25% then we're at 5c per mbit per month.

Come on...

$4 from Auckland to Whangarei, $13 from Whangarei to Wellington?!

This is why we're not getting the same value that the Australians get and it's why I think we should be building directly to Australia.

There are so many flaw's and horrible assumptions with this I'm going to need bottle of whiskey before I can respond

DonGould

3867 posts

Uber Geek
+1 received by user: 158

Reply # 686140 14-Sep-2012 20:04

Beccara: Both are simply stupid idea to try and drive down cost for 25% of the population that seem to be behind a "F#$@ You, Got Mine" attitude

If that's what you read from what I wrote then I really need to just give up.

I talked again and again about freeing up capacity so we can make better use of existing networks without having to upgrade them all.

I talked about not over building the major part of SCCN (which frankly I think is just a "F#$@ You, Want Mine" proposal).

I talked about attempting to open up new markets, not just over build existing ones.

I talked about advantages to 100% of the New Zealand population, not just 25%.

I talked about advantage to 25% of the Australian population, perhaps more.

I talked about making better, smarter use of cache and DC farms.

I talked about marking a shorter path to the west, to benefit the growing Asian population in this part of the country.

I talked about attempting to make better use of the northern flowing data paths and ways to reduce the amount of traffic space that's used to Wellington just to flow on south.

Do I need to just draw more pictures to explain this stuff?

As for transit to .au being $6, again and again and again everyone on all the public facing lists talks about that not being the problem, we know that. The problem is again and again and again is said to be the national transit.

If I'm wrong, cool. How do we fix it because clearly we're not managing at present.

Australia has better value than we do. 1Tb data plans. Show me a 1Tb data plan with TelstraClear on their HFC network.

Promote New Zealand - Get yourself a .kiwi.nz domain name!!!

Check out mine - i.am.a.can.do.kiwi.nz - don@i.am.a.can.do.kiwi.nz

1 | 2 | 3 | 4 | 5 | 6 | 7