Rewiring Costs

[Taff in Glaws]

So while I'm still on the first fix and need to decide in next few weeks anyone like to summarise and conclude?
Slap in the some old 'Cat 5 (6,7 or whatever) whilst the boards are up and we're back to the bones or say ++ugge% it and use wireless! :?

 

[Flyfisher]

Well, if it was me, I'd definitely install data wiring, but it depends on your needs really. This article might help choose between cat5e and cat6. In general, it's usual to install the latest spec cabling but in practice it might not make much difference, depending on what you want to do with it. For a fully cat6 compliant system you need to use all the right connectors and patch leads as well. Fortunately, most practical applications are very tolerant of less-than-perfect data wiring.

I can stream video over my 100Mbps/cat5 home network with no problems, and I could upgrade to Gigabit Ethernet by changing the LAN equipment if necessary, so my cabling is still going strong more than 10 years after installation, well before video streaming was the norm.

Of course, if you're really keen, you could install ducting that would support 'blown fibre' technology - then you really would be future-proof :wink:

 

[Penners]

future-proof

Is that "future-proof" as in "takes 7 years to be superceded, rather than 5 years"?

 

[Flyfisher]

I was thinking of optical fibre as being 'future-proof' in the sense of providing sufficient bandwidth for anything in someone's lifetime, which is all the future-proofing most of us really need - but I take your underlying point about how quickly technologies become superceded.

However, superceded is not necessarily the same as obsolete. I'd argue that my cat5 data wiring is nowhere near obsolete, even though it has certainly been superceded.

 

[Penners]

I just think back to my first computer, the Sinclair ZX81, 26 years ago, and compare that with the machine that I'm typing on now. The development in that time has been absolutely staggering.

So with that rate of progress and change, I find it inconceivable that one could install any IT equipment that would be remain even halfway useable for the remainder of one's life (even at my age, let alone yours!).

 

[Flyfisher]

. . . . but I was only talking about the actual optical fibre (or cat5 cable). I entirely agree that the equipment connected to it is likely to change in one's lifetime.

 

[NT]

Its hard to predict what will last and what wont. Twisted rubber pair 1930s mains wiring can still be used for pc networking today.

All we can do is hedge our bets, pick the most likely and the most flexible contender.

I'd agree that conduit is the most futureproof option, as you can put most types of cable and fibre down it, but whether those will be how future networking is done we dont know for sure.


NT

 

[Flyfisher]

"Conduit" for data cabling has been in commercial use for at least 20 years, but it's not always as basic as metal tubing. Even if a company (or individual) took the long term view to install optical fibre, there are many different types to choose from, which means plenty of scope for making the wrong choice.

In response to this, I think it was BT that developed the concept of 'blown fibre', whereby a hollow tube is installed and then the required optical fibre is literally 'blown' down the tube. Whoever it was, the technology has been around for a long time. During 'Big Bang' in the City of London during the mid-1980's, there were many merchant banks installing such things into their new trading floors. Google for 'blown fiber' and you'll find loads of info.

However, I doubt there is any real justification for such technology in a domestic environment.

 

[NT]

I would not assume that optic fibre is the network of choice for future home networks. Its just as likely to remain too expensive, with other options cheaper and thus used instead. We really dont know. We could just as easily end up with something totally different, such as thin 100 core cable.


NT

 

[Flyfisher]

I agree that no-one can be totally sure of what the future holds and there are always new technologies being developed in research labs, although that alone doesn't mean they'll ever find the commercial light of day.

But such speculation doesn't really help anyone interested in choosing what data cabling to install today.

For my money twisted-pair cabling is the answer. It's a well-established, global standard and is extremely low-cost. Continuous advances in the connecting equipment (also extremely low-cost) also means that it can keep pace with modern bandwidth requirements - after all, the same 'cheap and nasty' telephone cable designed for 3KHz speech can now be used many decades later for 2-8Mbit broadband.

Similarly, when I wired my house with cat5 twisted-pair just over 10 years ago, 10Mbit ethernet was the norm, whereas today it is 100Mbit with Gigabit ethernet now firmly established as well, all of which I can run over the same cat5 cable.

I believe the incredible versatility of twisted-pair cabling is one of the key reasons why optical (or other) wiring has not caught on more widely and for anyone wanting to install domestic data cabling today, twisted-pair (i.e. cat5e or cat6) is the best choice.

 

[NT]

I totally agree that 5e is currently the best bet, but the main reason I would give would be different. If we look back at historic technologies, the ones that were practical to use the longest were the ones most popular in their day, because the parts and expertise stayed available the longest.

I suppose the ideal would be some kind of supercheap tubing that could be buried in plaster, then future cables or fibres could be put in easily.


NT

 

[Penners]

I suppose the ideal would be some kind of supercheap tubing that could be buried in plaster

With a piece of cord fed right through it, as a future drawstring?

 

[Flyfisher]

I totally agree that 5e is currently the best bet, but the main reason I would give would be different. If we look back at historic technologies, the ones that were practical to use the longest were the ones most popular in their day, because the parts and expertise stayed available the longest.

I'd go along with that; I think it's all more-or-less the same reasoning. I think that "best" is always difficult to define unless you narrow things right down to a specific feature or just accept the most widely used as a qualification in it's own right - which then leads on to all the points you mention which, in turn, helps to perpetuate things.

So, back to Taff in Glaws's question that resurrected this thread - yes, 'slap in some Cat5e' while it's an easy job.

I would, though, recommend that you think about installing two cables to each outlet though, or at least in those locations where you're likely to have a PC or other networked device because you might find it useful/essential to have a telephone in the same position. I'm speaking here as someone who only installed one cable per outlet (except in my study) and have regretted it ever since. C'est la vie :roll:

 

[Pford75]

I work for Cisco Systems (global market leaders in wired and wifi networking) and certainly the company view is that the future is ultimately wireless.

I'm working with a number of municipalities who will ultimately seek to provide wireless broadband as a utility, via the passive infrastructure (street lighting, sewerage etc.) We have projects underway in Amsterdam, San Francsico and a number of rural areas across Africa.

Korea is regarded as the leading country in networks and the vast majority of domestic broadband access is via wireless Mesh networks. It is perfectly possible to acheive speeds of 1GB via wireless.

Furthermore, wireless is 'futureproof' in the sense that there is no physical network infrastructure to be removed (an AP can be changed without problems).

Even BT themselves predict in their 21CN strategy that "the majority of homes outside of major urban developments will be served by wireless communications"

 

[Flyfisher]

I have little doubt that wireless technologies will play a huge part in future comms (as they already do today), but I have serious doubts that wired (copper or fibre) networks will disappear completely.

For public access over services supplied by municipalities then I would agree that wireless is ideal - and it's interesting how many of the far-eastern countries are far more visionary in this respect that the west. I recall Singapore offering free wireless internet in certain public locations many years ago.

But these are not really 'mission-critical' services. It would be a very brave organisation that did away with its wired networks because you surely can't have absolute control over the quality or performance of a wireless network? And while gigabit speeds may be possible with wireless, I suspect we're a long way from such speeds being available in a domestic environment (unless Cisco have some new standard up their sleeve?).

So, while I'm perfectly happy to surf the interweb using my wireless laptop, I much prefer to use the faster, and guaranteed, interference-free, performance of my wired LAN for transferring my multi-gigabyte video files, for example.

However, it's certainly a fast-changing world, so who really knows for sure? What is remarkable is how quickly the internet has become an almost essential 'utility'. Certainly, if I was looking to buy a new house the availability of a good broadband connection would be high on the list of 'must-have' features.

 

[NT]

Even if youre convinced the future of pc networking will be wireless, and I dont think thats a sure thing, it still makes sense to install plenty of cat5, beucase it has many uses besides computer networking.


NT

 

[sidd]

Wow. Since I was here a few weeks ago, I can barely recognise the discussion! I was nervous about getting too technical before...

I've got a few things to add in, in no particular order:

1. Yes get the Cat5/6 wires in if it is not too difficult. They will be useful.

2. Homeplugs may not replace Cat 5/6, but there is some more cool stuff coming along. This press release may seem unimportant (http://www.itrancomm.com/pr2006_03_06_Homeplug.htm) but to me it spells a completely new home automation paradigm. Imagine if for pennies per 13 amp plug, you could make it IP controllable. Every electrical appliance could have IP control by this new standard. And imagine if all TVs were delivered with a homeplug already attached to the main power cord. Sky wouldn't bother running wires through your house.

3. The big demand on bandwidth to the home is video. Now most people like a balance between synchronous and asynchronous media. Many people come home and switch on the TV to see what is on. They don't want to make any effort in choosing anything. That's the easiest media to send, and puts the least load on the network outside the home, as everyone is watching the same stream at the same time. We could get it by satellite, wires or whatever. The video on demand thing is different. It needs huge bandwidth for everyone in the street to be watching the same HD download. And while that may happen eventually, there are so many better ways of getting the media to the viewer in the near term. But because of this bottleneck, I wonder whether it will be the ISPs who will charge extra to cache all the media closer to the end user, and thus not bring the internet to a halt.

4. This just a rant that my new house is not going to get terrestrial freeview until 2012 if ever, because the BBC seems to be more interested in satellite and issuing press releases complaining that the public are at fault for not going digital.

5. On a slight digression: has anyone got any views about Honeywell CM Zone? It is not quite as flexible as the Viessmann solution, but it is cheaper, has longer range and is actually available in the UK.

6. A friend of mine tells me that there is a new ultra fast burst wireless protocol, which was kept secret by the US military, and that will replace everything. I'm a bit skeptical, but it sounds clever.

 

[NT]

ISTR reading about wireless computer comms without network cards of any kind, done by encoding rf data into the video signal, causing the monitor to transmit it. I forget exactly how the receiving end worked.

cant find the link...


NT

 

[Flyfisher]

2. Homeplugs may not replace Cat 5/6, but there is some more cool stuff coming along. This press release may seem unimportant (http://www.itrancomm.com/pr2006_03_06_Homeplug.htm) but to me it spells a completely new home automation paradigm. Imagine if for pennies per 13 amp plug, you could make it IP controllable. Every electrical appliance could have IP control by this new standard. And imagine if all TVs were delivered with a homeplug already attached to the main power cord. Sky wouldn't bother running wires through your house.

HPCC seems like an interesting initiative, but I did a quick search and couldn't find any actual products yet. From a brief reading, it all sounded a bit like the well-established, but not widely used, X-10 standard. perhaps the key issue is the "pennies per 13 amp plug" - if it really could be pennies then I'd see it catching on because they could be installed as standard even if not everyone actually used them, but I recall X-10 mains outlets costing around £25, so that's perhaps why it hasn't been more widely adopted. It'll be interesting to watch this HPCC initiative this year and see if it turns out any practical products.

 

[sidd]

Quite true. And there won't be anything useable for at least a couple of years. X10 is OK. I've used it quite a lot, but it isn't that cheap, and it is pig ugly unless you wire it in to a CU. Everything else is too expensive or inflexible. Like X10, this would be an overlay on existing wiring with no modification - killer criteria. But X10 has enough flaws, that I'd prefer to try something else this time.

I admit I'm speculating somewhat, but given the size of even the new 200 homeplugs, I'm guessing that the incremental manufacturing cost is pennies. Everything will have been put onto a mass-produceable chip. And with IP addressing, it has few limits, unlike X10. I think it could be huge.