Discussion in 'Networking, Telephony & Internet' started by Schnooper, Sep 11, 2013.
Can you be more precise?
Is it still bugger all when multiplied by 70,000?
in the grand scheme of things, yes.
I work on the technology every day, including to about 1 o'clock this morning fighting with a site not doing what I wanted it to. precise values come under my heading of company confidential, but they are a lot less than is probably though.
power concerns are a tiny, infinitesimal, minute drop in the bucket. they're not even on my radar as an operational cost compared to other concerns. lasers cost money to run too, and what about all the NTDs that *won't* be needed to be powered? or, for that matter, the capital saving of not having to provide 16 million NTDs and power supplies?
FTTN would very probably cost more to run in terms of raw power consumption than FTTP, but it's not an argument that will get any traction. if nothing else, it's an operational cost as opposed to a capital one.
I hope you're not one of the 1100 Telstra people facing the axe today
There are such things as quick connect terminations already - i use them quite regularly. They are intended as a temporary measure mainly( but like all things meant to be temporary....) and work quite well.
scotch locks are not a fix for anything.
they're a bandaid measure that usually gets used as permanent solution in lieu of a real one.
much like these bad boys.
Theres still the labor costs to get to the break, sure, but once you have both pieces of cable in your hand, you can fix the copper one quickly and easily by yourself. (hence the 'free')
Got a link? We are talking about something I can use to join 2 pieces of unprepared fiber right? (much like I would have if someone cut a cable).
I know you can terminate fiber into quick-release type terminators that you then patch around the place, but for rejoining 2 pieces, I always though you needed a splicing machine?
I do not work for Telstra, so no, but thank you anyway. TBH I suspect a lot of the job shedding may be contingent on what's agreed between the LNP and Telstra re the use of their copper for FTTN, although that will in turn depend on the outcome of the company review that Malcolm Turnbull has ordered. it will be most interesting to see how many buried bodies are discovered, along with how many contracts there are with rather nasty exit clauses. it would not surprise me at all if the ALP ordered NBN management to very much nail their colours to the mast on that one - it wasn't ever going to bite them either way.
every NBN SDU connection to a customer premises has a field-fitted SC/APC termination at the box where the optical lead-in terminates. the lead-in needs to be cut and fitted to length, thus the method used.
they work fine in the context in which they are used - NBN connections have more than adequate optical budgets on the GPON link to allow for the minor inefficiency of using a field-installable terminator. I wouldn't use them for transmission links, but nor does the NBN.
the alternative is the method Telstra uses/used for their FTTP product, which was preterminated lead-in cables that came in lengths of 25m. they are a pain in the arse. Murphy's law states that a 50m cable will be 1m too short, so you have to use a 75m one. you now have 24m of excess cable coiled up in a pit, along with anything up to 7 other cables with the same challenge. getting the lid on or off a pit is like wrestling with the Kraken, and that's assuming you haven't put a pressure point on a cable in the process.
NBNCo also has to get a leadin through an existing LIC, which will have an existing cable (or two) in it, and that's assuming it was actually done using compliant conduit in the first case, isn't full of silt, or cracked by ground subsidence/reactive soils or tree roots. trying to get a preterminated SC/APC through that would be next to impossible. the bare ADSS self-rodding lead-in is quite small and can generally be muscled through OK unless the LIC is a total basket case.
the current Fujikura connectors seem to be working quite well, when combined with a simple optical power level test to check for correct installation.
I can't see such a fix ever really being used for assurance reasons. anything with a single fibre, you'd just replace end to end. anything with multiple fibres is ribbon, so deribbonising to kludge is a horrible idea for all sorts of reasons. if the cable segment is short, replace it. (my idea of "short" is <1km.) if it's longer than that, consider dropping in a couple of pits with inline butt splices and closures, with a bridging segment - this depends as much on civil considerations, it doesn't work in a concrete jungle of footpaths. you leave plenty of maintenance loop length to allow for just this sort of flexibility, because backhoe fade is a fact of life if you have plant underground.
Visiting Sydney this weekend I was reminded of this thread and the sage advice given by some people suggesting if you want faster internet move to a capital city.
I moved my sister to a new house in an inner city suburb of Sydney over the weekend and their adsl connection was hooked up yesterday afternoon. Expecting something reasonably decent imagine my surprise when the best they could muster was 1600kbs downspeed and 400kbs upspeed. In the boondocks twice the distance to the exchange I can manage more than triple the down and double the upspeeds.
Move to a capital city for better internet. Yeah, I'll believe that when I actually see it.
NTU's would cost what, $50 each on a grand scale....? Under $1B.
Power supplies, sure, cut them out and save some cash, power goes out, too bad, only those would need 24*7 phone should get it (mainly elderly). Most people have mobile phones anyway.
One could have the same argument about building 70,000+ node units, only to be pointless unwanted boxed once they finally decided to remove the last 400m of copper. One could also argue that the cost to "test each and every single copper line" would be too expensive and wasted time.
I just don't get FTTN at all. Install new Fiber cable to like 90% of your property, and lets leave the last 10% / 400m on old copper and bottle neck it to get speeds that in 5 years time won't be anything to brag about.
I'd rather spend $0 and leave it as is, than spend $29B on a FTTN network. That's a far better cost saving argument imo. I'd also mention that FTTH should generate a ROI, a FTTN does not, so whilst you might justify on saving money on capital items, one must look at the bigger picture.
If we are too lasy to build something right from the start, don't bother at all.
Btw, don't forget my 14.4Kbps pair gain is still in the ground to this day and age, pretty sad affair...
Naturally they are. Wouldn't want to give any weight to a potentially massive ongoing OPEX that the alternative simply doesn't have.
They sure do. Pity they consume so much less power than Cu kit.
NTDs or VDSL modems, that particular aspect of power is pretty much equal between the schemes.
It sure is an hefty capital expense when considered all in one shot. Much better to sneak in and hide a vastly greater running cost that accumulates day by day, week by week and year after year.
As an aside, why is it 16,000,000? There's barely 8,000,000 households in AU. Been attending the Turnbull school of making-scary-numbers-up?
Save $100M on not providing NTDs but spend $1B on electricity.
Much better to sneak in and hide a vastly greater running cost that accumulates day by day, week by week and year after year.
go up by a factor of at least 6, and that's buying by the boatload.
got any figures on that? hint: I do.
not that it matters, the decision is political.
Only what Alcatel publish in their product documentation
Excellent. What is the power consumption of a VDSL head unit?
According to Turnbull, they're technology agnostic.
I very much doubt the liberal party is agnostic anything
Hey hey hey, don't you go around bringing facts into this discussion !
Caspian, if you really have the figures, why don't you tell us what they are?
• The nodes themselves
• The exchanges the nodes are connected to
• The VDSL modems
Basically, the total of all the required devices for FTTN as opposed to FTTP (and there are more devices required for FTTN).
Anyone know how much a Top Hat install cost?
OMG, seriously, think about what your typing before you do it. You don't think for a moment that such information is:
1) Commercially in confidence
2) Cabinet in Confidence
3) he will be covered by an NDA, or Code of Conduct to prohibit sharing of internal corporate information without approval (just like most businesses?)
I don't understand...
it's not as if anyone is going to ring up Telstra and quote his C number
even if he had one - cmon...