from the bureaucrats-boost-bumpkin-broadband dept.
El Reg reports
Good news, hayseeds: the FCC has ruled that rural broadband should be held to the same standard as that of connections in built-up cities and towns.
The stateside regulator has issued a new order, which states that in the US countryside, providers must maintain at least a 10Mbit/s downlink speed for subscribers if they want to call it "broadband". The ruling will apply to ISPs that receive taxpayer dollars through the FCC's Connect America Fund.
Previously, the FCC required ISPs to deliver a 4Mbps connection in order to get the rural broadband funding. Companies are also required to deliver 1Mbps upload speeds. The increased bandwidth is supposed to match the speeds of urban connections, and is the first [uptick] for the Connect America Fund program since 2011.
[...]The $4.5bn Connect America Fund was established in 2011 with the aim of expanding and improving phone and broadband service in the American countryside. The fund is slated to run through 2020 with the goal of connecting 100 million US homes countrywide with 100Mbps broadband.
(Score: 3, Insightful) by Snotnose on Sunday December 14 2014, @09:44PM
Personally, I could get by just fine with 5 mbps. Most of my internet is gaming, I want low ping times more than high speed.
Why shouldn't we judge a book by it's cover? It's got the author, title, and a summary of what the book's about.
(Score: 2) by dlb on Sunday December 14 2014, @10:12PM
(Score: 2) by Snotnose on Sunday December 14 2014, @10:29PM
Nope. Think of your car as the speed, and the road the lag. Doesn't matter if your car can do 180 on the open road, on the 805 at 8 AM you top out at 10 MPH. That's actually a crappy analogy, but I can't think of a better one.
There are various speedtest sites to test your speed. To check lag ping various sites like google.com, cnn.com, soylentnews.net, etc.
Why shouldn't we judge a book by it's cover? It's got the author, title, and a summary of what the book's about.
(Score: 2) by dlb on Sunday December 14 2014, @10:49PM
Anyway, I think I see what you're saying: Factors other than internet speed can slow down your game play, hence ping time takes into account those other factors as well, and is a better measure than just internet speed.
(Score: 2) by c0lo on Monday December 15 2014, @03:06AM
Actually, this case seems to be a one of insufficient bandwidth causing delays.
A better analogy would be the sneakernet [xkcd.com] - heaps of bandwidth in the case of a station wagon full of tapes hurtling down the highway, but awful lag time.
Addressing the GO question:
It may, in case of congestion. However, there are cases in the ping time (aka latency) may go down without congestion: see the bufferbloat [wikipedia.org].
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 0) by Anonymous Coward on Monday December 15 2014, @05:06AM
The car analogy I'd go for is acceleration vs. cruising speed.
A large download is like a trip where you leave a residential neighborhood, hop on the highway, and cruise at 60mph(~100km) for 4 hours.
There is a low ping Ferrari that lets you leave your neighborhood very quickly, then begin 4 hours of cruising at 60.
A higher ping diesel truck will still get you there in 4 hours, just takes 10 seconds longer to get to 60.
For those large downloads, who cares if it takes 100ms more time to initiate your download, it's all about the sustained transfer rate you're cruising at.
For gaming there are lots of acceleration changes, so it's more like a Laguna Seca. Even if your were to cap both vehicle's speeds at 60, the Ferrari would lap the diesel truck because it can get back up to speed after a turn much faster.*
* Even ignoring exit speeds, down-force, rollover risk; all factors that make the Ferrari an even better choice.
(Score: 4, Interesting) by compro01 on Monday December 15 2014, @12:14AM
Technically yes, but not to any meaningful degree beyond dial up speeds.
For, say, a 44.6kbps dial up connection, it would take about 100ms for a full 576 byte frame to be transmitted down the line. That's quite a bit.
Up that to a 1mbit connection and a full 1500 byte frame, and you're down to about 12ms. That's 88ms of delay cut, which is useful.
Upping it further to 10mbps gets you down to only 1.2ms, but that's only a 10.8ms savings, which isn't that relevant.
(Score: 0) by Anonymous Coward on Monday December 15 2014, @12:18AM
> Doesn't the internet speed affect your ping time? The more mb per second, the faster the ping packets transmit.
Actually you are correct, but there are also other factors because the advertised bitrate is just for the first hop - the one between you and the ISP. Once your packets get out on the internet there are a variable number of hops that your packets take to get to their destinations and each of those hops may or may not be congested and may run at a variety of bitrates.
(Score: 2) by Tork on Monday December 15 2014, @01:17AM
🏳️🌈 Proud Ally 🏳️🌈
(Score: 0) by Anonymous Coward on Monday December 15 2014, @01:48AM
I'm not a big gamer. I have TWC with up to 100mb/s (though I only get half that on a good day). I used to have AT&T DSL years back and even though the bandwidth was only 1.5 MB/s at the time, for gaming, my ping averaged ten milliseconds less. Not a lot but the lower ping does make the game noticeably smoother.
(Score: 2) by M. Baranczak on Monday December 15 2014, @02:22AM
"Internet speed" is a meaningless term. There are two significant metrics here: bandwidth (how many bits per second you can receive) and latency (how long does it take for a packet to get from the server to the client, or vice versa).
(Score: 2) by meisterister on Monday December 15 2014, @02:46AM
Maybe it should be replaced by the terms "Internet Width" and "Internet Speed". Width would be how wide the connection is, ie. how many bits/sec can be pushed over the line, and Speed would be the raw latency.
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.
(Score: 1) by fritsd on Monday December 15 2014, @04:07PM
"internet width".
I like that.
So it measures the number of "lanes" on the "information superhighway".
The thickness of the intertubes, if you will.
(Score: 2) by githaron on Monday December 15 2014, @03:02PM
Ping (latency) is used to measure how fast a piece of data gets from point A to point B once it is actually sent. Bandwidth measure how much data can be sent at one time.
(Score: 2) by Immerman on Monday December 15 2014, @06:16PM
Only if the amount of data being sent is sufficient to saturate the available bandwidth.
The two primary considerations for "internet speed" are bandwidth and latency - pings measure the latency.
Think of it like pouring marbles through a tube, one byte per marble. Bandwidth tells you is the diameter of the tube, and obviously the fatter the tube the faster you can pour your marbles into it. But latency tells you the length of the tube, and if the tube is really long it may take hours for your marbles to reach the other end, even if the tube was so large that you could dump in a whole wheelbarrow at once.
A comparably extreme real-world example would be space probes. The data link with a Mars probe may offer a multi-Mbps bandwidth capable of streaming video and scientific data, but the 3 minute minimum speed of light delay between our planets means the delay between sending a signal and getting a response can never be less than six minuets. Utterly unsuitable for gaming, or even a video-chat, despite having more than enough bandwidth.
For purely terrestrial data links lightspeed typically isn't a consideration, but instead there are routing delays on every device the packet hops to. Plus delays caused by having the signal routed through congested routes, where it will have to spend time waiting in line for the opportunity to make the next hop.
(Score: 4, Informative) by SGT CAPSLOCK on Sunday December 14 2014, @10:37PM
I have one of these great rural Internet connections provided by my local monopoly, "TotalHighSpeed.com".
The service is a WISP (Wireless ISP) which uses Motorola Canopy equipment, and I happen to be a pretty hardcore gamer, but I'd still trade this connection for literally anything else. I've even tried begging AT&T for a 128k ISDN connection to replace this, but they won't bite.
Although my ping times are pretty great, they're indescribably unstable. I can ping 40 to a server in QuakeWorld, but it's always with random spurts of 200+ ping. Aside from that, the real problem is the packet loss. It's like there's a random number generator deciding if each packet I want to send/receive is worthy of making the full journey.
Add to that the fact that my connection is a maximum of something like 90 kilobyte/sec (I have no idea really since it's so unstable) and that it costs more than $100 USD a month... Well, at least you'd get some low pings in games if we could trade!
My problems are typical for WISPs in rural areas which have zero motivation or reason to improve things. They know that I have literally no other options.
(Score: 0) by Anonymous Coward on Monday December 15 2014, @08:16PM
I just completed a 3.5 mile PTP link using modern Cambium gear (what Canopy turned into) and the latency for standard ICMP pings is always under 10ms. The link is very reliable and tested out at 49mbit in each direction. It is running 5.8ghz so it is highly directional and requires LOS. The radio hardware reports a 32,000 ns (nanosecond) radio flight time for the data, so the multi-millisecond latency is coming from higher layers.
You may be seeing variable latency because you are in a P-MP deployment (e.g. you are competing for spectrum and time slots against other radio subscribers)
To be fair, all network designs eventually have a resource contention problem and produce latency spikes as certain nodes are forced to re-transmit denied/dropped data.
WISPs have two main concerns: the ratio of active subscribers on a given AP/sector, and the backhaul connection that services each AP.
(Score: 2) by Arik on Sunday December 14 2014, @10:57PM
Latency, reliability, and even upload limits are all more important.
If laughter is the best medicine, who are the best doctors?
(Score: 2) by frojack on Monday December 15 2014, @03:35AM
Actually, that's not true.
For most people, download speed determines how fast the page loads, how smoothly the movie streams, etc.
A page request is small, just a URL in a wrapper of TCP. The page itself is much bigger.
And latency doesn't matter in this case either. For email, web browsing, latency is seldom an issue.
It can be an issue in gaming.
Reliability is important, but its also seldom an issue.
So you see, the internet is actually designed quite intelligently, contrary to an uneducated opinion, or corporate suspicion.
Download speeds are almost universally slower than upload speeds, because A) you don't need upload to be all that fast, and B) you won't be able to effectively ping flood someone else with a single connection. Those who need balanced or enhanced upload are usually running a web server of some kind, and they can buy that type of service.
No, you are mistaken. I've always had this sig.
(Score: 2) by c0lo on Monday December 15 2014, @05:13AM
Err, um... punoɹɐ ʎɐʍ ɹǝɥʇo ǝɥʇ ? Maybe?
This is why it badly need to be fixed: pay or we'll throttle it to the sky and one step further (err... sorry, I mean we'll be "Quality-of-Service"-ing it).
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by frojack on Tuesday December 16 2014, @04:22AM
Doh!
Yeah, other way around...
No, you are mistaken. I've always had this sig.
(Score: 2) by Arik on Monday December 15 2014, @03:02PM
If laughter is the best medicine, who are the best doctors?
(Score: 2) by SuperCharlie on Monday December 15 2014, @12:07AM
Its hard to get good pings when there are 50 hops from isp, out and about, noname servers, god knows where and then finally to your destination. Im an old fart, so take it as you may.. but I distinctly remember doing tracerts in win95 and 98 and hitting places in 10 hops pretty regular. I am surprised now if I see less than 15-20.
(Score: 3, Informative) by Snotnose on Monday December 15 2014, @02:54AM
Lets say you need to get something from point A to point B. The something could be a letter (low bandwidth) or a 1,000 washing machines (high bandwidth). Lets say one moving company has people hanging around waiting for a load to come in. Load shows up, they're out the door. That's low latency. Another company, when a load comes in, gets on the phone looking for someone to take it. That's high latency.
Why shouldn't we judge a book by it's cover? It's got the author, title, and a summary of what the book's about.
(Score: 3, Interesting) by Geezer on Monday December 15 2014, @10:55AM
The telcos never actually invested anywhere near what they've already received in subsidies, so I'm not confident we'll see any change in behavior.
Case in point: my local telco/isp has been getting rural access subsidies from Day 1. I'm still stuck on a 1.5Mb DSL connection, and my provider has no plans to upgrade my end of the county...ever. Cable is not available.
No worries, I just chalk it up as an ancillary cost of not living like a lab rat in an apartment block or dealing with suburban homeowner's associations.
It just pains me that this is apt to be just another suckling session for providers at the federal teat.
(Score: 3, Insightful) by Sir Garlon on Monday December 15 2014, @01:41PM
I don't follow. If the telcos have made / will make enough investment in infrastructure to deliver 10 Mbps, then isn't that satisfying the social contract? And if they just pocket most of the money like you say they haven been (and I have no reason to doubt you), then won't the FCC cut off that revenue? So I can't see this policy change as anything but a forward step.
If anyone should be griping, it should be people like me, because it's my surcharges that are paying those subsidies. I have no problem with paying a tax to ensure you get a network connection. My problem is that if I have to pay that tax, then I want to see a non-profit organization or the county government building and administering the network. If the telcos don't want to build a connection for you, then they should not get to collect fees from you after I pay to have it built. (Or, alternatively, the subsidy should provide loans, not grants.)
[Sir Garlon] is the marvellest knight that is now living, for he destroyeth many good knights, for he goeth invisible.
(Score: 2) by Geezer on Monday December 15 2014, @04:10PM
Indeed, you are injured via the surcharges. It's just insult on injury that the charges are not honestly being used as intended.
I say honestly because no, the social contract has not been met if the partial performance is in bad faith.