4 ms·
Probably because for one, you're not going to get gigabit speeds in practice; the servers only have limited amounts of bandwidth, so you'll probably only be get
by lambda 16y ago
Probably because for one, you're not going to get gigabit speeds in practice; the servers only have limited amounts of bandwidth, so you'll probably only be getting a few megabits to tens of megabits most of the time anyhow.
Also, fiber is generally pretty scalable. Once the fiber is laid, you generally have an awful lot of unused fibers; the limiting factor is the routing hardware, which can be upgraded later. The initial investment, for getting the fiber in the ground, on the poles, and to the homes, is the important part.
Heck, in the US, there are places where they're still working on 10 Mbps FTTH networks, and places where there aren't any FTTH network plans at any point within the near future. 100 Mbps FTTH would be great.
Also, remember that Australia (like the US) is a heck of a lot bigger and more spread out than Hong Kong. That means a lot more miles of fiber for the same number of people, which means more of the initial investment needs to go into just laying the cables.
- andrewstuart 16y agoAll true, but surely the cost of implementing gigabit instead of 100Mbps is trivial relative to the cost of laying all that cable? Why lay a formula one racetrack and only drive pedal cars on it?
- lambda 16y agoEr, no, as I said, the real difference between gigabit and 100 Mbps fiber is in the routers (and repeaters, and switches, and the like), not the cables. And it would cost significantly more to implement gigabit than 100 Mbps, and it wouldn't really make sense anyhow until the backbone infrastructure catches up. But if you build out the 100 Mbps network, you can then just upgrade the hardware later on, having already spent the cost on laying the fiber. 100 Mbps fiber to the home is no "pedal car." 100 Mbps is pretty damn fast. Take a look at the global speedtest.net averages: http://www.speedtest.net/global.php http://www.speedtest.net/global.php . Note that even places with high penetration of FTTH, like South Korea and Hong Kong, only have averages for real-world performance a few times higher than, say, the US, which has very little FTTH penetration (the US is at 9.72 Mbps, Hong Kong at 14.18, South Korea at 32.26, and Australia currently at 7.12). Given these sorts of results, gigabit vs. 100 Mbps FTTH will likely make no difference in real-world performance, while gigabit will cost substantially more. And they both use the same fiber; once the backbone infrastructure and hosts are in place to support that kind of bandwidth, you can upgrade from the 100 Mbps to gigabit hardware at that time, rather than spending the money now when the extra bandwidth will go unused.