Olympic race for broadband

GuyClapperton 0 Tallied Votes 272 Views Share

British Telecom has announced that it is ahead of schedule for superfast broadband* and will deliver it in time for the Olympic Games in London. It's less good that only 40 per cent of customers will be able to get at the service but then the Internet was ever thus.

What's really interesting to me is that people still talk about the speed rather than what it can do when they're announcing these services. It's possibly the single least useful thing network providers actually do.

According to my ISM I have an 8 meg service. No laughing at the back. The thing is, it does me perfectly well. I can watch videos on YouTube, listen to the radio live, I get email coming in at real time speed thanks to SMPT. You know the routine. I could, apparently, get 50 megs by going to another ISP and apparently by 2013 I'll have 100 megs. But do you see the way the latter examples are just numbers? No marketer ever tells you the big secret: they're going to improve the speed because they can (and undoubtedly a load of bottlenecking will go away) but actually the broadband people have already got is likely to be fine for most purposes.

I have no doubt other purposes will emerge. If you'd told me in 1999 that music would eventually account for a load of domestic broadband I'd have been puzzled, but that's common knowledge by now. But just for once, I wish someone would tell me why I need the next big thing rather than how fast it's going to be.

* Which, I take it, means we'll all think it deadly slow within three years

Dani AI

Generated

is right to ask "why" instead of being dazzled by headline Mbps, and is right that TV-over-IP is already driving higher home demand. The practical question is what experience you actually want — uninterrupted 4K on several TVs, reliable HD video calls, fast cloud backups, or low-latency gaming. Raw download numbers only matter insofar as they support those outcomes.

Headline download speed is just one axis. Upload speed, latency (ping), jitter and packet loss determine real-time performance. Consistency at peak times matters too: "up to" speeds and ISP oversubscription can make a fast plan feel slow when everyone gets home. Also remember local factors: a clogged Wi‑Fi, an old router CPU, or many simultaneous clients will limit real throughput long before the ISP line does.

Rough rules of thumb (approximate): SD 1-3 Mbps per stream; HD 1080p ~5-8 Mbps; 4K ~15-25 Mbps; HD video calls about 1.5-3 Mbps upstream. Estimate need by summing active streams and add 20-30% headroom. Example: 3 HD streams (36=18) + 1 HD call (3) + phones (20.5=1) = 22 Mbps total — choose a 25-30 Mbps plan to avoid contention and background traffic impact.

Simple test-and-troubleshoot workflow: 1) run wired speed tests at different times; 2) measure latency/packet loss (ping/traceroute) to spot real-time issues; 3) repeat tests over Wi‑Fi to isolate wireless problems; 4) check modem/router firmware, client counts and CPU; 5) if the ISP link is the limiter, ask for line stats or a plan with higher upload/symmetry. Often the best upgrade is better home networking (5 GHz, wired backhaul, newer router) rather than just buying the next headline Mbps.

blud 82 Linux Reject Moderator

The next big thing is going to be TV providers that are going to send video over your internet connection.... Oh wait, we're already doing that. The ability to load a High-def blu ray movie over the internet at the push of a button for on-demand watching is the next wave of media crossing those big tubes they call the internets.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.