AT&T dangles gigabit broadband plans over 100 US cities Keeping pace with rival Google Fiber, AT&T has announced a possible expansion to its GigaPower broadband internet service. The company said that it would be looking at plans for bringing its gigabit service to 100 new cities in 21 metropolitan areas around the US. The proposal would include major population hubs such as Los …
I know of about a dozen people in Austin that have been waiting at least 6 months for AT&T to deliver their gigabit fiber. OTOH I also know of a couple still waiting on Google, so it goes both ways.
It's also interesting to see the same old game that any particular house in Austin has a choice of one and ONLY one gigabit fiber provider. You sign up for AT&T, Google, or whoever serves your block or you do without. There's no true competition.
It's also interesting to see the same old game that any particular house in Austin has a choice of one and ONLY one gigabit fiber provider. You sign up for AT&T, Google, or whoever serves your block or you do without. There's no true competition.
Given how horribly expensive these roll-outs are it probably doesn't make sense to have more than one local loop provider. Better for them to cover different areas rather than duplicating effort.
But be careful what you wish for. Competition drives down prices which in turn damages RoI making mass roll-out of fibre unlikely. This is the situation the UK is in. Most areas only have a single local loop operator (though nearly 50% has cable running alongside). The competition is in the retail market though with the major local loop owner (British Telecom) being forced to give wholesale access to its network and to allow other operators to install their own equipment in exchanges.
As a result we have some of the cheapest broadband in the world but our speeds are second-best. On the plus side we have broadband of at least some kind available on pretty much 99% of lines in the country amongst the highest internet take up per head of population in the world. So for headline speeds competition may not be the way to go. But for cheap, widely available and popular it seems to work well.
I have symmetric gigabit fibre, and it does me plenty good:
1ms ping to google, my company's DC, 5 ms to the office.
Insanely fast downloads.
Lots and lots of HD teleconferencing, no cut-outs, stuttering, buffering or quality drops, even when someone jumps on a download.
Uploads as fast as downloads. BT wanted to sell me 300Mb down/20Mb up for £60/month.
Did I mention the 1ms ping? It's pretty useful after hours too. Boom HS.
My ISP is hyperoptic, they only do certain areas. I'd post a speed test, but a) I'm at work, and b) most speed test servers don't have the capacity to fill my pipe. Usenet does it pretty well, the highest I've hit is around 858 Megabit/s (80MB/s) when I'm downloading my linux isos.
This post has been deleted by its author
This post has been deleted by its author
This is a company that is currently selling customers U-verse which it claims is fiber in all locations and fiber only and then turns out to be installed as DSL in the 1-2mbps range. They won't provide contract information to their customers when its requested. But require you to basically sign a blank check to them for service.
Yeah maybe they will install gigabit fiber in a few places, but everywhere else they are currently getting paid to upgrade or bring up to standard as usual will suffer.
AT&T has a monopolist mentality. A monopolist staff, a lobbyist staff, they won't invest a penny they don't have to. In 1972 I was paying the equivalent of $78 a month for a single line, no long-distance, and a telephone that was from around World War II.
Home users don't need "Gigabit" speed. Only businesses do. Ten to fifteen MEGABITS is plenty to deliver multiple HD video streams and surf the web with abandon. What we cannot afford is to allow them to implement metered billing. This is their absolute objective: Unbelievably profitable, and it will help them facilitate putting Netflix and every other streaming service out of business. Imagine you get meter billed for Netflix, but not for "ComcastFlix" or "UverseFlix." It will look just like cable TV does today.
Only if "per user" is watching multiple HD streams at once. Unlikely. In my home (I have six Rokus) there is never more than 2 simultaneous "watchers." Two people can each watch an HD video stream and others can be doing e-mail, surfing, etc. easily with 15 Mbps. That's why I say fifteen megabits is plenty. NOT "hundreds." Never "hundreds."
Only an SMB or larger, with POS terminals, dozens or hundreds of employees using cloud apps, etc. would need "Gigabit" speed. A home user gets benefit only if s/he is downloading big files and wants them in seconds. Video streaming and surfing is plenty fine with 10's of megabits per second.
Again I say, Gigabit speed means nothing if you can't afford to pay for the "Gigabytes." In the absence of real ISP competition, I see pricing (like the old days with "long distance") taking people off the web early in the month, or forced to turn to "non-metered" cable company services like "ComcastFlix" or "UverseFlix." That's what they want. And will get if the FCC doesn't step in.
Only if "per user" is watching multiple HD streams at once.
Two people can each watch an HD video stream and others can be doing e-mail, surfing, etc. easily with 15 Mbps.
And yet HD is usually broadcast at a minimum of 10Mb/s with most pundits saying that's barely enough. In the UK the BBC broadcasts at 10Mb/s average using stat multiplexing across multiple streams. Last I checked most commercial HD is broadcast at 12Mb/s to 16Mb/s. Oh and that's just bog standard 1080i. If 4K ever catches on (hah!) bandwidth is going to quadruple.
There are some IP based services that deliver good quality HD at lower than that. Sky in the UK manages decent HD at only 800kb/s but that's a catch-up service so presumably has the advantage of more time in which to encode the stream. The BBC service which is designed to handle generic decoders and which carries live data requires 3.5Mb/s for HD and suffers from artifacting (not hugely so but it's clearly not really enough b/w for some content).
So in the absence of a sea-change in encoding/decoding technology (one that applies fairly consistently across all platforms that might be in use) I think it fair to say you need 10Mb/s for good quality video if you're going to be replacing TV broadcasting. Add in a couple of Mb/s for browsing and sundries and you're at 15Mb/s per user.
Huh? I don't know anything about the UK, but ALL streaming services in the US, including the live streaming services like MLB.tv, say 3-5 Mbps is enough for full HD video. I pay for 20 Mbps and I never have -any- problem with multiple HD streams in my household. I watch Vudu movies, Netflix stuff, live TV streams--all on multiple Rokus and I still have bandwidth to spare. No pixilating or any other artifact.
Only very big downloaders and businesses with dozens of on-line workers or POS terminals benefit from "Gigabit" speed.
Huh? I don't know anything about the UK, but ALL streaming services in the US, including the live streaming services like MLB.tv, say 3-5 Mbps is enough for full HD video
I think you should read more carefully. I wrote that broadcast was over 10Mb/s and that catch-up (which would include streaming services) could be lower because they had more opportunity to encode.
Search for 'Channel 4 HD' or 'Discovery HD' . 12Mb/s and 13Mb/s respectively. There are some HD channels that appear to broadcasting at around 9Mb/s (eg; search for 'AXN SPAIN HD') but if you bring up the details you see that the bitrate varies and on some days that can go as high as 18Mb/s so probably stat-muxed.
I would think this technically savvy audience would understand that "Gigabit" speed means nothing to a home user. NOTHING. Gigabit speed only helps businesses. It helps businesses who need big transactions-per-second speed. Ten to fifteen MEGABIT per second speed is PLENTY to deliver multiple HD video streams.
What consumers need to guard against is METERED BILLING. A total rip-off. Multiple credible studies have shown that incremental billing for internet usage is price gouging pure and simple. It's what a monopoly does: Create artificial scarcity and invest not one penny more than necessary. Internet service is already the most profitable product the monopoly cable and telephone companies sell--by far. The only reason we get "buffering" and "freezing" is because they haven't invested. Like monopolies NEVER do.
It doesn't do you any good to have "Gigabit" speed if you have to turn off the internet on the 12th of every month because you can't afford the "Gigabits."
If everybody had Gigabit Internet and IPv6 we could do away with all of these clumsy personal "cloud" solutions. You think Amazon/Box/Google/Apple/Microsoft storage is too expensive for remote backups? Tired of swapping files around on sharing services to get your vacation pictures to Mom? Your NAS, if recent, already has remote backup syncing to another NAS. You probably already have at least 1 low power device that can securely share any number of files over the Internet at 600+Mbps. Nobody has it turned on because it's slow (Basic AT&T is 768Kbps down/384Kbps up), complicated (need to buy IPv4 addresses from idiot tech support), expensive (must buy "business" upgrades), and likely prohibited (servers and caps) with ISPs like AT&T.
AT&T's U-verse is the most underwhelming hardware upgrade that I've ever seen. It's designed to be nothing more than a viable downgrade for people who've been screwed by Comcast. There's no way that AT&T will produce a viable Gbps product to consumers.
Steal all Verizons most valuable customers while Verizon chokes on its own PR. ATT costed out the effect of Google's Checklist upon the cost of fielding fiber to homes that want it. Costs decrease by 60%. Both Google and ATT are now racing to fiber high rises in Manhattan and condo associations, homeowner associations and other fiberhoods where 100% participation is guaranteed by the 24/7 lifestyles of homeowners. Those homeowners are strongly influential in the decisions of their businesses. Hence Verizon stands to lose both home and business contracts if it continues to act upon the otter-snot it is putting out about no-need-for-speed.
Both ATT and Google are racing to sign contracts with entire high rises. Once those 24/7 homeowners are signed up, ATT and Google will take their business-business as well. ATT costed the effect of the Google Checklist; cost of fielding fiber to homes decreases 60%, making it highly profitable and highly probable both will take advantage of Verizon's little nappy-time.
I'm in Raleigh, NC, where Time Warner Cable provides most people with Broadband. AT&T offers its U-verse service here also, but at slower speeds AND with the requirement that you abandon any router you already have to use their all-in-one box. Many of my friends have tried U-verse, and every single one has returned to TWC due to speed and reliability issues. I have an extensive wired and wireless network in place that I've built to serve my needs at home. I will not use any companies high speed solution if that is a requirement. Hopefully, Google Fiber and this new ATT product will arrive here soon and save us all from the evil monster that is TWC. Lower costs always follow when new players arrive in town.
What is the estimated startup time for both of these systems in terms of infrastructure placement? Are both of them fiber-to-your-house?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
