Slashdot Mirror


Netflix Compares ISP Streaming Performance

boustrophedon writes "The Netflix blog compared streaming performance among 20 top ISPs for the past three months. A Netflix HD stream can provide up to 4800 kbps, but the fastest American ISP, Charter, could sustain only 2667 kbps on average. Most Canadian ISPs beat that, with champ Rogers providing an average of 3020 kbps. Clearwire, Frontier, and CenturyTel were in the doghouse with under 1600 kbps."

209 comments

  1. Wrong. by AnonGCB · · Score: 0

    I've got Verizon FiOS, and though I know it's not that common, but I can get steady 3.7 MB/s streams.

    --
    http://CryoLANparty.com/ A lan I'm staff on!
    1. Re:Wrong. by Anonymous Coward · · Score: 1

      I guess you don't understand the meaning of "average".

    2. Re:Wrong. by Rinnon · · Score: 4, Insightful

      I've got Verizon FiOS, and though I know it's not that common, but I can get steady 3.7 MB/s streams.

      I'm not going to suggest that you are incorrect, but I am going to suggest that your single piece anecdotal evidence is not nearly enough to discredit the report Netflix put together.

    3. Re:Wrong. by Morty · · Score: 2

      Is there a value to looking at "averages" over large areas?

      I'm in Howard County, MD. On Comcast, I get about 8Mbps sustained after an initial 20Mbps. This is typical in my area. My neighbors using Verizon FiOS will typically see even higher throughput. However, this graph, presumably containing large parts of the US, has Verizon as slower than Comcast, and both are much lower than what I see. A person trying to make a decision on ISP service in my area would be misled.

    4. Re:Wrong. by macpacheco · · Score: 2

      The article uses Mbps (megabits per second) and not MB/s (megabytes per second).
      3.7MB/s is a LOT of bandwidth, it's 30Mbps. Not even FULL HD video uses that much bandwidth.
      You probably meant 3.7 Mbps.

    5. Re:Wrong. by Edzilla2000 · · Score: 1

      With Fios, he probably meant what he wrote.

    6. Re:Wrong. by technomom · · Score: 1

      Yes, I'd like to know if they were putting both DSL and Fios numbers together for this.

    7. Re:Wrong. by sanosuke001 · · Score: 1

      Time Warner is listed at 2400kbps; that's 300KB/s. I regularly grab large files at 2-3MB/s (big B). I'd like to know more about their testing methods because these seem a bit suspect. I wouldn't be surprised if it were throttling of the netflix content by said providers, though.

      --
      -SaNo
    8. Re:Wrong. by macpacheco · · Score: 2

      Full HD streams using H264 run around 10Mbps for poorly compressed video (small publishers without a video engineer doing the work).
      Look at wikipedia:
      http://en.wikipedia.org/wiki/High-definition_video
      It says that the heaviest video streams from Netflix use 5Mbps.
      I just played a few of the 1080 Full HD streams I have. The highest bandwidth they use is 10Mbps.
      30Mbps for a video stream is completely crazy, even if your ISP service is that fast, you won't be able to sustain that speed, there will be serious drops.
      Only Blue Ray uses that kind of bandwidth. Not even DVD.
      Maybe he's talking about video content accessed directly from Verizon. Then it's not internet streaming video, it's VOD content coming directly from your ISP.

    9. Re:Wrong. by uolamer · · Score: 1

      I agree.. Time Warner in my area, their lowest package I think will go above 300KB/s. I have their 2nd lowest package and I get at the very least 1.5MB/s, usually it is 1.7-1.8MB/s

      --
      s/©//g
    10. Re:Wrong. by Anonymous Coward · · Score: 0

      > 3.7MB/s is a LOT of bandwidth

      Err, that's a data-rate. Bandwidth is a measure within the frequency domain.

      It's like saying that your car has 250 tyres when you mean horsepower. No-one would be that daft.

    11. Re:Wrong. by Suki+I · · Score: 1

      I guess you don't understand the meaning of "average".

      It has 3 common meanings and the one they are using in TFA is not mentioned above.

    12. Re:Wrong. by tgd · · Score: 1

      No, but I can toss in another piece of anecdotal evidence -- I rarely can sustain streams to Netflix that fast, but I can easily pull data from other sources *much* faster than that.

      Now, its possible my ISP is trottling Netflix, but I don't believe they are that sophisticated.

      The problem with any of Netflix' information is that we can't tell if the problem is at the ISP or at peering points, or at Netflix itself. We also can't tell what speeds the customers in question actually are paying for. (Thus making the statistics basically useless beyond Netflix playing politics...)

    13. Re:Wrong. by Gadget_Guy · · Score: 1

      Time Warner is listed at 2400kbps; that's 300KB/s. I regularly grab large files at 2-3MB/s (big B). I'd like to know more about their testing methods because these seem a bit suspect.

      Well, I suspect that they tested it by streaming from their servers, not downloading files from other unnamed servers. There are so many varying factors that you can't just assume that it is throttling if you don't get full speed access to Netflix. If you try tracerouting to different sites you can find wildly different network topology resulting in vastly different download speeds. Perhaps the pipe going to the popular site Netflix has much higher usage that the one going to the site from which you download your files.

    14. Re:Wrong. by papasui · · Score: 2

      It's not ranking YOU. It's ranking them as a company, which includes all tiers of their service.

    15. Re:Wrong. by dreamt · · Score: 1

      Which is, of course, part of the reason BluRay is superior to Netflix (or just about any other) streaming :). And DVD wouldn't use that much because DVD isn't HD.

    16. Re:Wrong. by mikael_j · · Score: 1

      Why is 30 Mbps sustained transfer rate "completely crazy"? I have a 100/100 Mbps connection and I routinely have higher sustained download speeds when downloading things from servers that have the bandwidth for it. And when it comes to p2p the main problem I have seems to be that I'm using ZFS with RAIDZ and doing 100 Mbps worth of random writes while simultaneously seeding various things at 10-20 Mbps generally makes it pretty slow (I noticed after doing some cleanup of the filesystem that as long as I keep utilization under 75-80% this isn't really a problem, it's when the filesystem is beginning to fill up that performance starts to suffer which makes all other processes that rely on disk I/O sluggish too).

      --
      Greylisting is to SMTP as NAT is to IPv4
    17. Re:Wrong. by Anonymous Coward · · Score: 0

      I've got FIOS and I'm just outside Manhattan, I can sustain the full 4800kbps with room to spare. I've use Netflix on my PS3, which actually supports 1080p with 5.1 surround sound on a few titles. I've played a movie while browsing the web without any issue on many occasions. They only list "Verizon" on the chart, I suspect it's Verizon DSL and not FIOS.

    18. Re:Wrong. by r_naked · · Score: 1

      I've got Verizon FiOS, and though I know it's not that common, but I can get steady 3.7 MB/s streams.

      I'm not going to suggest that you are incorrect, but I am going to suggest that your single piece anecdotal evidence is not nearly enough to discredit the report Netflix put together.

      Well here is some more for you...

      I have Brighthouse cable in the Tampa, FL area, and I get ~4.8MB/sec (yes that is a big M for mega and a big B for bytes).

      On a different note, I would suggest anyone that has Netflix streaming to check out VuDu. I don't know if it is available for anything other than the PS3, but they offer 9Mb (that would be a little b for bit) streams that look great. There is still the occasional pixelation on real high speed scenes, but I have to guess they are using h264.

      Bottom line, there are plenty of providers out there that can stream HD (well what Netflix / VuDu call HD) in real time. AND there are some that could stream *real* HD in real time.

      --
      -- http://anonet.org -- The internet the way it was meant to be. Check it out, you may be surprised.
    19. Re:Wrong. by crashumbc · · Score: 2

      Exactly what I was thinking for Verizon and maybe some others... A lot of Verizon's DSL connections are either 3.0 or 1.5 capped which would destroy their throughput...

      that said probably 80+% of Verizon's service IS DSL not Fios, so its actually probably lifting their numbers higher then they should be, not the other way around. Last I heard Verizon had stopped rolling out Fios (this may just be internet rumor mill crap)

    20. Re:Wrong. by jellomizer · · Score: 1

      Yes BluRay is superior... However streaming is good enough for most cases. As a lot of us wants to watch a movie and not count the pours on the main characters skin. So the value of BluRay over a lower bandwidth connection isn't much of an issue. Once people get an average of 100mbs for their network then we may seen BluRay quality streaming media. But by then there will be something else that will top that.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    21. Re:Wrong. by ElectricTurtle · · Score: 1

      It seems very likely by the way the ISPs' performance is distributed. You can see all the ones at the top are primarily cable companies, followed by providers of mostly DSL, then things even slower by nature like clearwire 802.16 or whatever they use. Verizon is solidly placed between the cable and DSL groups. So either they have the best DSL in the whole damn universe (which my personal anecdotal experience would not support), or their stats are getting bumped by the inclusion of FiOS with the DSL performance metrics.

      --
      I support the Slashcott and will not be reading or commenting from 2/10/14 to 2/17/14. Beta is steaming pile of dog shit
    22. Re:Wrong. by budgenator · · Score: 1

      Well that's the point of the study, the ISPs are starting to throttle netflix, it's unlikely to change as long as a Democrat is at the helm and they are saving metric boatloads of money streaming rather the using the USPS. The ISPs are thinking if netflix used to spend Millions on postage from delivery and we're now doing the delivery, shouldn't they pay us millions now? To counter this Netflix is trying to shame the ISPs by making their poor performance public. To be honest I just changed from Comcast cable internet to AT&T DSL and the DSL feels like the Cable during prime-time for have the price, pretty much like the graph shows.

      --
      Apocalypse Cancelled, Sorry, No Ticket Refunds
    23. Re:Wrong. by elrous0 · · Score: 1

      It likely means that Time Warner (along with many other ISP's) is throttling Netflix or streaming video in general. I know my ISP does (Netflix starts out with a faster speed and then drops suddenly after a few minutes, and this happens consistently).

      --
      SJW: Someone who has run out of real oppression, and has to fake it.
    24. Re:Wrong. by njfuzzy · · Score: 1

      Verizon offers DSL in far more areas than FiOS. I suspect the results would be more meaningful if their two products were separated.

      --
      My Photography - http://ian-x.com
      The Deathlings (comic) - http://thedeathlings.com
    25. Re:Wrong. by curunir · · Score: 1

      The key word is likely sustained. You might be getting 2-3MB/s the majority of the time and may not be noticing the minute or two that the transfer speed drops significantly because you're either not watching it closely or not downloading anything at the time. But when streaming, if transfer speeds drop at all and you exhaust what little buffer there is, you've either got to adjust to the available bandwidth or stop and refill the buffer.

      My personal experience watching Netflix on Comcast and Charter bears this out...it's quite common for the vast majority of a movie or show to be very high quality but have short periods where the quality level dips or the streaming stops for a bit. So while I too frequently get big-Bs/s on downloads, the sustained transfer rate is lower because they can't keep that level of performance over the 2-hour period that I watch the movie.

      I don't think this is about Netflix specifically since I see the same dips in quality when watching espn3.com as I do with Netflix on both providers. And considering that both providers have paid espn to allow me to use espn3.com, I find it unlikely that they'd do any type of throttling to discourage using it as it would be much easier (and cheaper) to just not pay for it.

      --
      "Don't blame me, I voted for Kodos!"
    26. Re:Wrong. by Coren22 · · Score: 1

      Most definately. My FiOS is 25/25 service, and I have seen very nice 2.5MB sustained without issue. Even the different levels of FiOS would vary widely as you can get from 10Mb down to 100Mb down, that is a huge difference in effective streaming speed.

      --
      APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
    27. Re:Wrong. by AnonGCB · · Score: 1

      I know my megabits from my megabytes. Don't worry

      --
      http://CryoLANparty.com/ A lan I'm staff on!
    28. Re:Wrong. by nabsltd · · Score: 1

      Full HD streams using H264 run around 10Mbps for poorly compressed video (small publishers without a video engineer doing the work).

      Actually, 1080p24 works just fine at 10Mbps average bitrate using H.264 and a two-pass encode, which can get essentially transparent re-encodes of Blu-Ray sources (at least based on PSNR and SSIM results).

      If you can use the two-pass encode as a streaming source and guarantee a video buffer (like Blu-Ray), then it should work fine. For streaming of live 1080p24 (or worse, higher framerates), then you probably need 15-20Mbps average to maintain quality on the single-pass encode. Netflix should be fine for video at 3-4Mbps for 720p24, but they won't be able to afford the bits to allow the audio to be anywhere close to Blu-Ray quality.

      The primary reason Blu-Ray sees average bitrates of 20-30Mbps for 1080p24 is because they can. Once the disc size (25GB or 50GB) is decided on, it doesn't cost any more to fill it with bits. Unfortunately, it actually reduces quality in the scenes that need it because of the 35Mb limit for any single second of video. With a lower average bitrate, then hard-to-compress frames could use the equivalent of 60-80Mbps and not overflow the video buffer.

    29. Re:Wrong. by gknoy · · Score: 1

      Is there a filesystem better suited to p2p?

    30. Re:Wrong. by gknoy · · Score: 1

      And yet, when talking about network throughput, "bandwidth" is the accepted term, despite having an older definition too. Similarly, "firewall" and "hub" mean different things, but we manage not to be confused.

    31. Re:Wrong. by gknoy · · Score: 1

      If Netflix can stream significantly faster to Canada than to the US, it sounds like it's not a problem internal to Netflix... though I could be wrong.

    32. Re:Wrong. by ffejie · · Score: 1

      I have an LG BluRay player that uses Vudu, and I agree - their service is excellent. Unfortunately, their pricing model (pay per movie) isn't my preferred method when I'm already forking over $xx/month to Netflix for "unlimited" movies, but I have used it a few times when they have a new release that I really am interested in.

      --
      Disagreeing with me does not mean you get to mod me troll.
    33. Re:Wrong. by macpacheco · · Score: 1

      The best filesystems for managing large files are extent based, tree based filesystems.
      I store all my large files using XFS and I'm very happy with it.
      XFS delivers read performance that is the same as accessing the hard drive directly.
      So far what other filesystems have online defrag.
      It also makes a huge difference if your download software can pre-allocate files before writing to them, this can result in zero fragmentation.

  2. Wow Verizon is around 2k? by BradyB · · Score: 1

    This is FiOS, this is SLOW...

    --

    Good is never enough, when you dream of being great!
    1. Re:Wow Verizon is around 2k? by RevWaldo · · Score: 4, Insightful

      There's Verizon FiOS and Verizon DSL. Is the measurement for FiOS, DSL, or both?

      .

    2. Re:Wow Verizon is around 2k? by kannibal_klown · · Score: 1

      There's Verizon FiOS and Verizon DSL. Is the measurement for FiOS, DSL, or both? .

      I'm wondering that too. I have a hard time believing that Fios is lower than *some* of those up there considering the speed offerings they have.

      I'm on a 50 megabit Fios connection right now, and from my regular download speeds it looks like I'm getting that speed for the sites that can support it.

      A few years ago I was on 3 megabit DSL, though I hear their DSL can hit 15 or 20 megabit with their residential versions.

      Granted, it's possible streaming is throttled or can otherwise not take advantage of the full (or near-full) stream.

    3. Re:Wow Verizon is around 2k? by AlphaBit · · Score: 1

      This is definitely not the FiOS that I have. I can stream Netflix and Hulu and be running 2 games and using voip without any noticeable slowdown. I could probably add a couple more HD streams and still have some headroom. And in practice, I get almost that same bandwidth upstream (35Mb/s)

  3. Reverse the tables by MrDoh! · · Score: 4, Insightful

    Very nice. Rather impressive to pre-empt the ISP's.
    "well, your competitor is able to provide better speeds to more customers, why are you whining? Oh? AND You charge more for lower service? Interesting. Well, lets let your customers decide for themselves with more facts who they want"

    It'd make sense at this point for an ISP with a bit of sense to make a nice deal with Netflix to improve things here, then everyone wins.

    --
    Waiting for an amusing sig.
    1. Re:Reverse the tables by fleeped · · Score: 2

      Very nice. Rather impressive to pre-empt the ISP's. "well, your competitor is able to provide better speeds to more customers, why are you whining? Oh? AND You charge more for lower service? Interesting. Well, lets let your customers decide for themselves with more facts who they want"

      It'd make sense at this point for an ISP with a bit of sense to make a nice deal with Netflix to improve things here, then everyone wins.

      Not very nice. Remember the two-tier thing in UK? Perhaps Netflix is trying to reverse that in its favor by hinting for 'arrangements'? Shady deals like that won't really work in favor of the consumers.

    2. Re:Reverse the tables by MrDoh! · · Score: 1

      If you've got the choice, and the figures to back things up? Perhaps.
      "ok, this company is offering internet for 40 bucks a month, and I get the option for an extra 5 bucks to get Netflix bundled in" (which a smart ISP here, if traffic IS that popular, might be worth getting in on.
      "or I can spend 30 bucks a month, and Netflix works, just not in HD, but I've the choice"

      Ok, ok, it's more likely the ISP's will charge extar for netflix, AND you'd need a seperate netflix account AND if you don't pay, they'll slow things down.

      --
      Waiting for an amusing sig.
    3. Re:Reverse the tables by thomasw_lrd · · Score: 1

      The only problem with that is the non-compete agreements in the US. Where I live, I can only get Suddenlink (which involves paying for cable), or Centurytel (which is expensive, but still cheaper than the other option). I can't get anybody else out here. I know I've tried.

    4. Re:Reverse the tables by Mousit · · Score: 5, Interesting

      ...Well, lets let your customers decide for themselves with more facts who they want"

      Though unfortunately for many customers (the majority I'm sure), "who they want" is a choice between that ISP or nothing, so it doesn't help them too much to simply tell them hey, they're getting screwed.

      However, I would like to see this broken down into smaller areas. By region, or even by city, rather than just the ISP as an overall average. I'd be very, VERY curious to see if the very same ISP performs significantly better in areas where there's some actual competition going on.

      That would be nice to wave around "look here, here's measured evidence of what they're doing in areas they don't have to compete".

      As an aside, I already kind of see this in my area. I have Time Warner broadband. In my personal location, they're the only choice; even DSL is not available to me. The highest service available is 15Mbps and I average 5-8 most of the time. However, in the sections where Verizon FiOS is also available and competing? Why, suddenly Time Warner's got a 50Mbps service available which averages 35-40! Imagine that..

    5. Re:Reverse the tables by uolamer · · Score: 1

      Exactly. for me it is either Time Warner or DSL. A few blocks south or east the only option is Time Warner. A few miles in some direction there is no DSL or Cable, people have to use Dial up, Satellite, or 3G. While I do I have pretty good service, I would have much faster, better and possibly cheaper if there was competition.

      --
      s/©//g
    6. Re:Reverse the tables by johnncyber · · Score: 1

      Where I live, I can only get Suddenlink (which involves paying for cable)...

      I don't know where you live, but I know that Suddenlink does offer internet without bundled cable here in West Texas. A buddy of mine just canceled his cable tv while keeping his 10 Mb internet.

    7. Re:Reverse the tables by Anonymous Coward · · Score: 0

      Might have to look into that.

    8. Re:Reverse the tables by cvtan · · Score: 1

      Where I live we can only get Time-Warner and Frontier DSL. Having some other ISP give better service is interesting, but gives me no leverage as others have pointed out. Maybe we should go back to the public utility model. For some reason I'm asked to choose an electricity supplier even though there is only one electric infrastructure, but I have little choice in internet access. Have to say Roadrunner service has been fine, but pricey.

      --
      Sorry, but gray text on gray background is making my eyes bleed.
    9. Re:Reverse the tables by hedwards · · Score: 1

      I have the same problem around here. I'm stuck with Comcast who can't be bothered to provide any service on a reliable basis and Qwest which is reliable but slow. Seattle was planning on doing its own municipal fiber back in 2005, but Qwest pretty much told them not to do it implying that they'd be doing it. To date I have yet to see any evidence that they will ever be bringing faster speeds here as they've stagnated for the last decade. My connection speed has gone from 4mbps to 5mbps over the last decade.

      I'd like something faster, but my main complaint is latency and from what I've read Comcast isn't any better in that respect. Right now I'm paying $50 plus tax whereas Qwest charges in other regions $55 plus tax for 40mbps service via fiber.

    10. Re:Reverse the tables by CastrTroy · · Score: 1

      Funny thing is, I'm only paying for 3 Mbit/s internet with Rogers in Canda. However, ever since I started using Netflix, my connection suddenly started going at 10 Mbit/s. And not just for Netflix, but for the entire internet. Maybe Rogers got some inside information that Netflix was going to be collecting and releasing this information, and they upped the speed of all customers who they noticed accessed netflix. The change in speed has lasted too long, even through multiple power outages, and was too close in time, I recall it happening within a month of when I started using Netflix for it to be a coincidence. Maybe it is a coincidence. But it seems a little shady.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    11. Re:Reverse the tables by Belial6 · · Score: 1

      Do you have 'residential' or 'business' Comcast? Anyone can get 'business' Comcast. It is only ~$10 above the regular residential rate, which is what it looks like your paying. The service is dramatically better, and they don't do the filtering and port blocking that they do on the residential accounts since it is a service intended to be use by businesses.

    12. Re:Reverse the tables by Belial6 · · Score: 2

      Well, this kind of report can do a lot to bring awareness to the masses about what is going. Most people don't know what their speed really is, and thus don't care. They take the marketing terms 'fastest' and 'unlimited' and assume that they are getting the best that is available. People do care about their TV though. If they find out that their ISP is breaking their TV, they are going to be outraged. This is particularly true if it a cable provider that is breaking their TV.

    13. Re:Reverse the tables by Bengie · · Score: 2

      ISPs should have to advertise their "dedicated" speed as their speed and the "up to" speeds ad burst

      eg. 16/2 with no cap, but once you add in over subscription, it's more like 2/1 dedicated with up-to 16/2 burst

      If an ISP has a data cap, that should have to calculate into the "dedicated" speed.

      eg. 16/2 with 100GB cap would list as 160Kb/160Kb with up to 16/2 burst. I listed 160Kb for up and down because most caps include up and down, so it would be split evenly.

    14. Re:Reverse the tables by Maxo-Texas · · Score: 1

      At some point, it has to make sense to have some legal way for ISP's to cache the shows and movies locally. Perhaps in an agreed on encrypted form which their netflix users can download.

      You could probably set up a bank of 40 terabytes for under $2000 since it's not critical data you could use consumer drives.

      --
      She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
    15. Re:Reverse the tables by omnichad · · Score: 1

      Except that the dedicated speed is 0/0. Best effort delivery means that you'll very likely get something above 0/0, but it's not guaranteed at all.

    16. Re:Reverse the tables by Phoghat · · Score: 1

      I also have Time Warner with FiOS available. My download speed is quite good even with the regular base service according to the website Speedtest.net I get a constant download of 10.36 Mb/s. Things could be worse

      --
      Think of how stupid the average person is, and realize half of them are stupider than that.
  4. Pertinent? by Anonymous Coward · · Score: 0

    Do these charts represent the speed from the ISP to the consumer? Is it possible that an ISP could make it appear to Netflix that the speed is higher, while slowing down the transfer to the actual user?

  5. Ah Rogers by Kris+Warkentin · · Score: 4, Informative

    Yeah, my cable modem is stupid fast....only problem is, running full tilt I can go through my monthly bandwidth cap in eleven and a half hours.

    Fortunately, for the moment, the overage cap is $50 so if you download a bunch some month you just say, "Woohoo, unlimited bandwidth." For example, in January I downloaded 750MB which put me 625MB over my cap and would have cost an extra $780. Ridiculous no? And now the CRTC (equivalent of FTC) has ruled that the major ISPs are allowed to pass usage based billing fees onto third party providers which means there will be no more unlimited plans and the billing cap will likely go away.

    Basically, Rogers and Bell want you to watch their channels, not use Netflix, AppleTV, etc. And the wretched hive of scum and villany known as the CRTC is letting them do it.

    Not much point in fast internet if you can't use it.

    --

    In Soviet Russia, hot grits put YOU down THEIR pants.
    1. Re:Ah Rogers by Kris+Warkentin · · Score: 1

      Oops...meant FCC, not FTC.

      --

      In Soviet Russia, hot grits put YOU down THEIR pants.
    2. Re:Ah Rogers by Anonymous Coward · · Score: 0

      my cable modem

      -snip-

      For example, in January I downloaded 750MB which put me 625MB over my cap and would have cost an extra $780.

      I hope you meant GB and not MB. If it was over 3G then a 125mb cap seems possible but restrictive, but you said cable modem.

    3. Re:Ah Rogers by Anonymous Coward · · Score: 0

      And eleven and a half hours to go over 125MB hardly qualifies as a "stupid fast" cable modem...

    4. Re:Ah Rogers by Anonymous Coward · · Score: 0

      I so dread if they start having caps where I live. Some months I exceeded 1TB in traffic, last month I was about 1.5TB.. Damn pirates using all the bandwidth..

    5. Re:Ah Rogers by Malc · · Score: 1

      For example, in January I downloaded 750MB which put me 625MB over my cap and would have cost an extra $780

      Was this on your phone (sounds a bit cheap for Rogers Wireless :P), or did you mean GB?

    6. Re:Ah Rogers by Jason+Levine · · Score: 1

      Either you meant GB instead of MB or your cap is only 125MB. Even when Time Warner Cable was proposing ridiculously low caps, they chose 5GB. If any ISP decided that users should only download 125MB per month, they'd have a huge uprising on their hands.

      --
      My sci-fi novel, Ghost Thief, is now available from Amazon.com.
    7. Re:Ah Rogers by Kris+Warkentin · · Score: 1

      Sorry guys...it was early and I hadn't had my coffee. I meant GB.

      --

      In Soviet Russia, hot grits put YOU down THEIR pants.
    8. Re:Ah Rogers by Kris+Warkentin · · Score: 1

      GB....hadn't had my coffee yet. ;-)

      --

      In Soviet Russia, hot grits put YOU down THEIR pants.
  6. Re:Please say it ain't true ! by Rosco+P.+Coltrane · · Score: 1

    Well, that's not too hard, a sizable portion of the rest of the world is faster than America.

    --
    "A door is what a dog is perpetually on the wrong side of" - Ogden Nash
  7. Faster by Anonymous Coward · · Score: 0

    Why aren't we getting 10x this? For heaven sake we have had terabit optical data transmission technology for over a decade now. Isn't it about time to start getting REAL high speed to door steps?

  8. I'm not exactly color blind... by astern · · Score: 5, Informative

    ...but how in the hell is anyone supposed to pick the colors out of those graphs, at least three of them are the same shade of sky blue.

    I'd like to see this redone as the graph is certainly compelling, just a little bit more readable.

    --
    If the world isn't beating a path to your door you're doing something wrong.
    1. Re:I'm not exactly color blind... by Anonymous Coward · · Score: 0

      I am colorblind and I had to have my wife come over. She still had difficulty!

    2. Re:I'm not exactly color blind... by Anonymous Coward · · Score: 0

      Not being a troll here, but you might really, actually be in some stage of color blindness. To me the three shades of blue are all quite distinct.

      The two green lines are the most similar to my eyes, but still not something I'd whine about.

    3. Re:I'm not exactly color blind... by Eclectic+Engineer · · Score: 1

      You're not the only one. I wish more authors would put some thought into their graphs. In particular, I find it helpful when the legend is in some sort of order that corresponds to the graph. At least then I can correlate labels by position.

    4. Re:I'm not exactly color blind... by Anonymous Coward · · Score: 0

      I am colorblind, so I'm lucky they colored mind black.

    5. Re:I'm not exactly color blind... by Anonymous Coward · · Score: 0

      Top to bottom: Comcast, CenturyTel, Cable One.
      (you may be color blind)

    6. Re:I'm not exactly color blind... by jellomizer · · Score: 2

      We normally can differentiate 8 colors easily on an RGB Screen
      Red
      Yellow
      Green
      Cyan
      Blue
      Magenta
      Black
      White

      With a background color we are limited to 7 graph points.
      I would have recommended after they first looked at the data they saw that they were in 3 distinct bands High, Medium, and Low. Then seporated the graphs in those bands. using the more primary colors.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    7. Re:I'm not exactly color blind... by Eravau · · Score: 1

      The three seem fairly distinct to me. You probably just need a better monitor... or some color calibration performed on the one you have. But anyway... for reference, the top "sky blue" line is Comcast, middle is Cable One and bottom is CenturyTel.

    8. Re:I'm not exactly color blind... by Machtyn · · Score: 1

      I wouldn't point to colorblindness so quickly when dealing with monitors.

    9. Re:I'm not exactly color blind... by cvtan · · Score: 1

      The shades of blue are distinct to me. It would be good practice to decorate the chart lines so that they would be distinct even in a monochrome print (data markers, dashed lines, etc.). With so many chart lines, this is difficult.

      --
      Sorry, but gray text on gray background is making my eyes bleed.
    10. Re:I'm not exactly color blind... by kannibal_klown · · Score: 1

      The three seem fairly distinct to me. You probably just need a better monitor... or some color calibration performed on the one you have. But anyway... for reference, the top "sky blue" line is Comcast, middle is Cable One and bottom is CenturyTel.

      I can tell some of them apart, it just takes me longer. Is that Cablevision or Charter? Is that CableOne or CenturyTel? I just have to judge them for a second to figure out. At least they're not close to each other or criss-cross.

      I'm a big fan of colors + patterns; such as solid vs dashed.

    11. Re:I'm not exactly color blind... by Anonymous Coward · · Score: 0

      I had no issue with differentiating the colors... perhaps, you are--in fact--color blind?

    12. Re:I'm not exactly color blind... by Tromad · · Score: 1

      Someone on reddit fixed it: http://i.imgur.com/58vph.png

    13. Re:I'm not exactly color blind... by Eravau · · Score: 1

      There's no doubt that a little bit of pattern added to the colors would make it more universally legible despite computer or optical specs.

    14. Re:I'm not exactly color blind... by TheoMurpse · · Score: 1

      Yeah. Let's face it, Open Office makes terrible looking graphs. I'm surprised a billion-dollar company can't afford to make nicer looking graphs.

      All that being said, I have Time Warner, and I'm pleased with what I'm getting for $30/mo. I can hold steady at 1MB/s if the server can upload that fast. I don't care too much about up, as I upload very little. But, for what it's worth, I can sustain 50KB/s.

  9. Where is this leading... by Rinnon · · Score: 1

    Dear Netflix,

    Subtle suggestion noted. Perhaps we can discuss the details of our offering you a superior service sooner rather than later. We would like to propose that in exchange for offering you a faster connection to our consumers, by prioritizing your traffic over others, you openly endorse us as your ISP of choice in the regions we serve. I think you'll agree that we can both come out ahead in this particular arrangement. You of course, will be free to make similar agreements with other ISP in regions that we do not service, and we will not consider this to be an issue.

    Yours Truly, [Insert ISP Here]

    1. Re:Where is this leading... by ducomputergeek · · Score: 1

      Dear Netflix, Subtle suggestion noted. Perhaps we can discuss the details of our offering you a superior service sooner rather than later. We would like to propose that in exchange for offering you a faster connection to our consumers, by prioritizing your traffic over others, you openly endorse us as your ISP of choice in the regions we serve and pay us the sum of $5 per subscriber per month for the privilege. We will of course be offering this service to our customers for the sum of an additional $19.99 per month. This additional service fee will allow our customers free streaming to On Demand Video Service. You can make this On Demand Service, powered by Netflix, for the mere sum of $15M per month to co-locate within our network. We will of course then be willing to offer you the sum of $1.99 per subscriber per month. I think you'll agree that we can both come out ahead in this particular arrangement. You of course, will be free to make similar agreements with other ISP in regions that we do not service, and we will not consider this to be an issue. Yours Truly, [Insert ISP Here]

      Fixed that for you.

      --
      "The problem with socialism is eventually you run out of other people's money" - Thatcher.
  10. Great to live in the US... by Anonymous Coward · · Score: 3, Insightful

    Fantastic. The worst ISP in Canada is still faster than the best ISP in the US.

    Also, while interesting, this is basically useless to the average US consumer. It's not like you get a chance to choose between those 16 US ISPs. In the US, you're lucky if you get to choose between 2 of them.

    1. Re:Great to live in the US... by Lev13than · · Score: 2

      Fantastic. The worst ISP in Canada is still faster than the best ISP in the US.

      Yes, but if you try to watch Netflix on a Rogers account you'll blow through the download cap by the second act of the first movie. You'll also be paying more for the privilege, and Ted Rogers will personally come to your house at night to empty the coins out of all your pockets and leave your milk out on the counter.

      --
      When you have nothing left to burn you must set yourself on fire
    2. Re:Great to live in the US... by spammeister · · Score: 1

      You know, Ted Rogers is dead right? I'm pretty sure it's ok to shoot dead things (zombies) though...

      --
      I tried to think of a good sig, and this wasn't it.
    3. Re:Great to live in the US... by Eravau · · Score: 1

      Yeah... but the numbers look a little skewed because the Canadian ISPs are using metric bytes, which are a smaller unit than the Imperial bytes used in the U.S... much in the same way that the standard Canadian kilometer measurement of distance is shorter than the U.S. mile or liter smaller than the gallon. ;)

    4. Re:Great to live in the US... by Amouth · · Score: 1

      actual metric bytes would be 10 bits (metric is base 10) where as normal bytes are 8 bits.. so 1000kbs (metric) is 25% larger than 1000kbs (standard)

      sorry you had them backwards..

      --
      '...if only "Jumping to a Conclusion" was an event in the Olympics.'
    5. Re:Great to live in the US... by Eravau · · Score: 1

      Doh!

  11. ridiculously low?? by Dilligent · · Score: 1

    Uh what?? I do always get sustainable 12 Mbit out of my DSL line here in Germany (which is advertised as 16MBit, but due to bad lines only goes to 12). I get 7Mbit out of my prepaid cheapo 10 €/Mo unlimited (but slows down to GPRS after 1 GB of usage) mobile plan (vodafone) and am soon switching to 32Mbit cable at home (which i've been told is sustainable as well).

    What exactly is the big deal with getting customers rates exceeding 4 Mbit??? Gotta be kidding me...

    1. Re:ridiculously low?? by servies · · Score: 1

      It's always the same with the US, they think they're way ahead but in reality they're lagging ages behind ;-)
      Within half a year I'll have the possibility of a 100/100 Mbit/s connection (actually I picked the 50/50 solution, don't need more at the moment). Within a year it will be possible to get a 200/200 connection. All unlimited...
      And the Netherlands are not even leading the bandwidth race...

    2. Re:ridiculously low?? by Yosho · · Score: 1

      Internet access in the US sucks terribly, that's what. I pay $50 / month for 6 MBit cable, and most of the time I'm lucky if I can sustain a connection at 3 or 4 MBit. I have two alternatives in my area, which are either slower (but cheaper) or much more expensive (and only slightly faster).

      --
      Karma: Terrifying (mostly affected by atrocities you've committed)
    3. Re:ridiculously low?? by paintballer1087 · · Score: 2

      The biggest problem with this isn't the ISPs, but the customers. I was working for a nationwide cable ISP for about 3 years, and we offered speeds of 1.5Mbps, 3Mbps, 12Mbps, and 25Mbps. Which do you think the majority of people chose? The 1.5Mbps for $15-$20/Month. Most people didn't want to spend the extra $15/month for the 12Mbps connection, and then complained about the problems when they tried watching 3 streams from Netflix and online games at the same time. There's no way they would up their speed though, it's the ISP's fault that the speeds were so slow. There's just no winning when everyone goes for the cheapest product available. It's the same reason Walmart is doing so well... Most Americans go for low price over quality.

    4. Re:ridiculously low?? by servies · · Score: 1

      You can get 20Mbps ADSL2 (unlimited) for somewhere around €20 in the Netherlands, cable goes for a minimum of something like €50 (including (digital) TV and telephone, internet speed will be somewhere in the same region in that case) so the problem is still (partially) the ISP.

    5. Re:ridiculously low?? by theantipop · · Score: 1

      As has been noted elsewhere, this graph lumps all sorts of tiered speeds in together, and also lumps every geographic location together. I think from netflix's end, this is trying to show that ISPs are providing subpar service.

      And while I can't say for sure, I think there's a lot of traffic throttling happening behind the scenes with US ISPs. I have a 15mb Time Warner line that will usually see 1.5MB/s or more file transfers (HTTP, Bittorrent and however Steam transfers data), but has a lot of trouble sustaining an HD stream from Netflix. Heck, even the pseudo-HD stream from Comedy Central's Quantserve hosting hiccups regularly. So take these graphs for what they are: indicative of average sustained link speed between customers and Netflix.

    6. Re:ridiculously low?? by coldsalmon · · Score: 1

      Yes, the US has rather slow broadband compared to many similarly-situated countries. Consider this ranking: http://speedtest.net/global.php#0

      Interestingly, according to speedtest.net the US is slightly faster than Canada. I could speculate that the US's lag in broadband speed is due to the large size of the country and difficulty in providing broadband to rural areas, but I am sure that Russia has an even bigger problem in that regard and they are significantly faster (on this list). Also, this does not explain why even urban areas have embarrassingly slow broadband.

    7. Re:ridiculously low?? by initdeep · · Score: 1

      except that again, people want to look only at overall country-wide population density, which is a poor metric for a case like broadband.
      instead they should be looking at the distribution of people and the population density within that distribution area.

      if i have a 100 square mile country, with 100 people in it, overall population density says that i have 1 person per square mile.
      reality might be very different if instead i see that all 100 of those people live within the same 1 square mile and the other 99 square miles of the country are completely empty.

      this example would make it very easy to provide "uber high speed" network access to those 100 people and yet look like the country has really low population density when placed on one of those nifty "comparison" charts.
      And then there would be people using my little fake country to "prove" that the US is "far behind" the rest of the world in broadband adoption and speed.

      it would all be bullshit, but people would spew idiocy like they do now anyway.

    8. Re:ridiculously low?? by hedwards · · Score: 1

      Where are you providing service though? I live in Seattle which is one of the best connected cities in the country and I can't get any service that's faster than 5mbps. Supposedly Comcast goes faster but from my experience that's for a burst and you're lucky if you've got any service on any given day. None of the major ISPs are willing to advertise what their speeds are in various places, just that fraudulent up to number.

    9. Re:ridiculously low?? by Anonymous Coward · · Score: 0

      I have charter, and I average about 25mb/s I am signed up for the 16mb/s plan.
      These speeds represent how much netflix streams through the ISPs. If netflix isn't trying to stream anything faster than 2.8mb/s, then nothing will be above 2.8mb/s. So the reason these are so low, is because of netflix, not the ISPs.

      This is more a comparison between them at how easily data can be streamed through them.

    10. Re:ridiculously low?? by paintballer1087 · · Score: 1

      I guess since I'm not there anymore I can talk about it. It was Cox Communications in Cleveland, OH. The speeds were "up to" but consistently remained at (within 10%) advertised speeds, and many times were much higher than that. The 25Mbps speed was $59.99/ month, with 5Mbps upstream (if I remember correctly). One thing the company was very good at was increasing bandwidth to meet demand, if they sold the service they made sure they could provide it. Only once or twice did I see problems in which they couldn't meet the demand, which were due to sales members pushing specials and huge increases in the highest tier sold. There was a couple of weeks in which speeds were slower for customers, but we did node splits in the affected areas and had it back up to normal ASAP. I can't speak for regions other than Cleveland, but the company as a whole received high customer satisfaction ratings for the service.

  12. Boo Hoo by Anonymous Coward · · Score: 0

    It's better than what I get from Netflix.
    A big fact 0 kbps because of region restrictions.

  13. Honorable mention by eedlee · · Score: 1

    Interesting to see Verizon, Time Warner and Comcast all in the top five. It seemed certain there would be at least one throttling between those three. We've had our Netflix on the first two and never had a problem (yet).

  14. ISPs are doing dumb penny pinching by macpacheco · · Score: 2

    Fiber optical very high speed equipment (used behind the xDSL/Cable copper network and behind the wireless towers) has never been as cheap as its now.
    Long range gigabit ethernet stuff is dirt cheap.
    10 Gigabit ethernet, which allows 2 thousand 4.8Mbps streams is already very affordable for carriers.
    A pair of fiber strands can carry 16 10 Gigabit links easily, that's enough for 32000 top speed streams.
    Long distance fiber optical cables typically have at least 36 strands. Some reach as high as 144 strands. Do the math and you'll see fiber capacity is almost never the issue.

    But then there's a very interesting FACT. If you exclude p2p and video streaming, 10Gbps link can provide bandwidth for one million users. That's right, typical users that don't run P2P or use video streaming/download services require just 10kbps bandwidth.

    That's the conundrum. Heavy users (p2p and video streamers) are responsible for such a disproportional share of bandwidth consumption that most providers just don't care about the quality of service they deliver for such users.
    They might actually prefer you go elsewhere given the tight margins.

    You should be glad if you can pay US$ 10 more to have higher quality service. At least that way you ISP can't complain that your piggybacking on the average light user. Honestly asking for a discount if you use little bandwidth isn't very reasonable either. A very large portion of the ISPs cost is last mile stuff that doesn't change if you're using zero bandwidth or maxed out in the xDSL case. Copper / coaxial cable stuff requires most maintenance, as they are most subject to ice storms, lighting, traffic accidents, ... Fiber can have redundancy.

    ADSL is the best option for video, as long as you're close enough to your ADSL provider DSLAM that you can get fast enough speed. Cable is only better if you're far from the nearest ADSL DSLAM. DSLAM is the counterpart to the ADSL modem. Your ISP should be able to estimate your max link speed with your address and tell you when you're too far for your selected speed.

    Just trying to demystify some facts. Most large ISPs / carriers won't discuss this stuff openly. They prefer you don't know the real facts. I'm not trying to judge what is right or wrong from either side.

    1. Re:ISPs are doing dumb penny pinching by macpacheco · · Score: 1

      Obviously fiber is better....
      GEPON can reach crazy speeds.
      But most users are still years away from that kind of service.

    2. Re:ISPs are doing dumb penny pinching by Bengie · · Score: 1

      "ADSL is the best option for video, as long as you're close enough to your ADSL provider DSLAM that you can get fast enough speed. Cable is only better if you're far from the nearest ADSL DSLAM. DSLAM is the counterpart to the ADSL modem. Your ISP should be able to estimate your max link speed with your address and tell you when you're too far for your selected speed."

      My mom has fiber to the house, she lives about 2 blocks from her ISP's wiring building, she gets a 40ms ping to her first hop which has a DNS name from her town, her town only has about 200 people. xDSL doesn't auto-magically make things better. Over subscription is an issue on every network. The question is, where's the bottle-neck?

      "They might actually prefer you go elsewhere given the tight margins"

      Recent article was talking about how High Speed Internet brings in several times more profit than TV. To bring this into perspective, I use Charter Comm. I paid $35 for extended cable. I was trying to cut corner for bills, so I was planning on dropping extended cable for just basic cable. They cut my bill by $15 AND added Showtime/HBO. I now pay $20/month for 100 digital channels plus Showtime/HBO. I still only watch the same 2 channels and only when I go to bed, but they're still making a profit on me.

      If they're making a profit from $20/month plus ST/HBO, then how much profit were they making from $35 and just extended cable? I pay $46/month on top of that for internet. and they make several times that profit from HSI.

    3. Re:ISPs are doing dumb penny pinching by m0masb0y · · Score: 1

      Interesting math, "...has never been as cheap as its now", and "...very affordable for carriers" costs what?
      Who is the most in "...most providers just don't care about the quality of service they deliver for such users".

      There are a great many assumptions made about network providers. Most of them are not fair representations of any known facts. Vilifying the network provider is really easy. But, they aren't the enemy. If they can afford to provide you a service you request, they will. And, they will usually care about all their customers because that is how they stay in business. If more customers communicated effectively with their network provider the world would be a much better place. :-)

      --
      A smile and a hug never hurt anyone ;-)
    4. Re:ISPs are doing dumb penny pinching by macpacheco · · Score: 1

      I have ADSL2+ service at 12000/1200 alignment speed.
      I can ping other users on the same local area at about 15-20ms. I can ping the access router at less than 6-11ms.
      I can ping sites at the my country's central internet hub (São Paulo), about 1000 miles away in fiber distance with 30-40ms.

      Over subscription is the standard. All carriers/ISPs do it.
      For example, say an ISP have exactly 1000 customers with 10Mbps down each. That's 10Gbps downstream total sold bandwidth.
      I can guarantee that one 10Gbps link is an absolute overkill for that case.
      Even one 1Gbps link might barely reach full capacity utilization. Still overkill.
      Over subscription doesn't even mean running the backbone links to capacity. It just means that you don't need to purchase one Mbps for each Mbps sold.

      The actual issue is the level of over subscription.
      As long as there's no dropped packets and lag is low enough, there's no need to upgrade.
      If the links are internal to the same ISPs network or is a zero cost (peering) link, and it runs on leased/owned fiber it should be kept with spare capacity, since upgrading doesn't increase monthly costs.

      The issue of over subscription starts to surface when the link is paid for its speed monthly. See the Comcast link with Tata bandwidth charts leaked. That is most likely the pair of last resort links, to destinations far away from Comcast's network, where it doesn't have peering arrangements. What is not depicted on those charts is queuing level (number of packets in the transmit queues at each side) in peak times. As long as pings aren't prioritized, it can be measured by pinging the ip on the other side of the link.

      Typically residential ISPs will wait until the link is projected to reach 1% packet loss before ordering an upgrade, and use QOS to give priority to higher paying customers (dedicated full links and business links). 1% packet loss is already nothing to take lightly, an upgrade should be ordered right away. Some higher cost carriers might opt to run links at 50% capacity, with pairs, so that if one link fails, the remaining one can handle the traffic. That's usual with tier 1 / 2 providers, since all their traffic is dedicated (those carriers only provide wholesale and other very high speed IP links, usually minimum 100Mbps dedicated).

      Now, the 2% documented packet loss with Comcast links with Tata telecom, that's completely unacceptable.
      Even 1% packet loss can prevent streaming/downloads at full speed.

    5. Re:ISPs are doing dumb penny pinching by macpacheco · · Score: 1

      I purchased 2 years ago a pair of single mode bi-di (single strand) SFPs with 20Km range for less than US$ 1000 including a pair of 8 port managed gigabit switches to install the SFPs on. The pair of SFPs alone costed about US$ 300 each. 120Km bi-di SFPs back then costed about US$ 1500 per pair.
      Widespread FTTx projects are causing gigabit ethernet optics to drop in price very fast. 2 years is a *long* time in telecom.
      The aggressive price reductions are now spreading onto 10 gigabit XFP stuff.
      If you know what to look for, you can get a very good sense on prices just searching eBay for new equipment for vendors that sell thousands of units through eBay. I just found new SFPs for 120Km for US$ 135 each on eBay ! US$ 270 a pair. That's a 80% price reduction in just 2 years.
      Just 10 years ago any kind of high speed WAN optical links costed in the range of ten thousand dollars in STM/SONET equipment alone. You just didn't deploy an optical network without STM/SONET. Today L2 switches can handle redundant link recovery with rSTP.
      Large L2 and L3 ethernet switches (L3 switch = pure ethernet router) from vendors outside the Cisco/big gang are eating away into their tradition of charging tens of millions of dollars on huge routers with enormous profit margins.
      With VLAN priority tagging routing is no longer needed in intermediate ISP/Carrier network hops. Routing is only needed at the borders where you service customers and where you exchange bandwidth with peers and buy upstream bandwidth. That can reduce router requirements 3 or 4 fold. Real world examples of single hop IP routing is Global Crossing and TIM Seabone. One hop from South America to Europe. One hop from South America to the West coast of USA. And the next hop is already at another network or the customer of said network.
      I'm not vilifying anybody, just informing people on factual information.
      It's important to remember that an ISP / Carrier can't redo its network every year or two or three. Investment needs to be recouped.
      But there's a lot to be said about what ISPs do when they are in a monopoly position. Those are definitely evil doers. Comcast seems to be one of them. Oi, Telefonica and Embratel here in Brazil are such examples.
      Telefonica had such a serious technical problems that it lead Brazil's FCC equivalent from suspending sales of ADSL products until the situation was solved after a streak of very serious problems affecting millions of ADSL users until the problem was solved. They had multi day DNS unavailability that essentially shutdown their network.
      Complaints on Comcast are pretty serious as well.
      We don't need to be enemies of nobody. However we have to understand that businesses exist to make money. And sometimes the easiest way to maintain a substantial profit margin involves delaying necessary investments and preventing those delays from being know by their customer base, even when they are badly needed. Politicians lie all the time, most time if they spoke the truth they would loose votes either way. Large corporations are typically in similar situations, where at least manipulating the truth is seen as absolutely normal (but of course never admitted).
      Add to that the tendency of the population to preferring feel good misinformation over the truth (typically the truth is very ugly), and it leads to the current wasteful government and high profit corporate environment. Backroom lobby deals.
      Only no bullshit information can enlighten consumers / voters. For instance the US is on its way to being bankrupt in less than 20 years. No political party has a realistic plan to avoid that. And the population can't come to terms that even the tea party solution might make things worse. The only solution is efficiency. Dumb cost cutting is bad, as well as dumb spending.

  15. Different Services need to be split by cwtrex · · Score: 4, Insightful

    The different connections need to be split.

    For example, Verizon needs to have:

    Verizon DSL 768kbps - 1Mbps
    Verizon DSL 1.5Mbps - 3Mbps
    Verizon DSL 4Mbps - 7Mbps
    Verizon DSL 10Mbps - 15Mbps
    Verizon FIOS 15Mbps
    Verizon FIOS 25Mbps
    Verizon FIOS 50Mbps

    Obviously a low end DSL connection is not going to be the same as those who can order the 10-15Mbps DSL connection. And it is likely that the DSL 10-15Mbps connection is going to be different from the FIOS 15Mbps.

    To group all of those connections into one Verizon line is completely misleading. And if they didn't take measurements from all of those connections, then then that makes the results even more suspect as the graph doesn't specify what type of connection they chose to test with.

    1. Re:Different Services need to be split by papasui · · Score: 1

      All the service providers have multiple tiers of service, not as misleading as you seem to think it is.

    2. Re:Different Services need to be split by cwtrex · · Score: 1

      My point is that those different services need to be split or similar services for each tested and stated with specifics mentioned for each. Verizon was simply an example.

    3. Re:Different Services need to be split by silas_moeckel · · Score: 2

      This is the collected stats from the real world. Picking out what tier of connection they have would be nearly impossible unless it's encoded in the reverse DNS. Associating an IP address with a particular carrier is easy anything past that is near impossible to determine.

      --
      No sir I dont like it.
    4. Re:Different Services need to be split by dkf · · Score: 2

      The different connections need to be split.

      Not really, especially as each of those ISPs has lots of complexity in their service portfolio so doing the split would be exceptionally tricky (and make the whole graph impossible to read, instead of just plain difficult). What I find more interesting is that the average level of service of the worst performing Canadian ISP would put it in the middle of the top performing group of US ISPs. Maybe it's because the US has a much lower population density than Canada? Or maybe it's because US ISPs prefer giving their customers the shaft to actually providing a decent and value-for-money service...

      --
      "Little does he know, but there is no 'I' in 'Idiot'!"
    5. Re:Different Services need to be split by Gates82 · · Score: 2

      Mod parent WAY up.

      The first thing I thought when I looked at the graphs, what service plan are these people on. Most I know have the cheapest package they can find with connection speed upper bound at 1.5-5mbs. At those speeds the throughput Netflix is reporting look pretty good. I have a 20mbs connection (from an ISP not listed in the report) and over the summer I routinely streamed MLB.com (@ 8mbs), the wife would have a Netflix movie on her laptop, and the kids watching some show from Netflix as well. Aggregate bandwidth requirements were roughly 15mbs.

      It fits: you get what you pay for, though I don't defend the prices or lack of sustained rate one typically finds with ISP's. I just found that good hardware, a decent home network configuration, and staying away from the cheapest bottom rung plan tend to go a long way.>/p>

      --
      So who is hotter? Ali or Ali's Sister?

    6. Re:Different Services need to be split by beanpoppa · · Score: 1

      Netflix regularly sends emails asking how the quality was when watching 'x'. It wouldn't be a big deal for them to allow customers to input their ISP/Type/Rate plan into their account so that they can compile more specific statistics.

    7. Re:Different Services need to be split by elashish14 · · Score: 1

      Yes, but penalizing Verizon (or company of choice) for having slow data rates will give them an incentive to upgrade them. I know many people choose slower data plans, but if I were Netflix, I would judge based on what's fastest around the area, within some reasonable price (say, $40/month). I wish they disclosed a bit more about their testing process.

      --
      I have left slashdot and am now on Soylent News. FUCK YOU DICE.
    8. Re:Different Services need to be split by initdeep · · Score: 1

      yes, but do all service providers offer tiers of service SLOWER than the highest speed netflix streaming offers?

      DSL providers = yes.
      Cable providers may not = yes (my local cableco's lowest offering is 12mbps)
      wireless carries = yes.

      so it is in fact, a POSSIBLE poor metric to use.

    9. Re:Different Services need to be split by Anonymous Coward · · Score: 0

      I agree that breaking out the connection speed would be next to difficult. But they could have at least included Max and Min values for each provider. Though, this might be misleading too unless the data was scrubbed to remove corporate connections.

      David

    10. Re:Different Services need to be split by Anonymous Coward · · Score: 0

      Even just a separation between Verizon FIOS and DSL would be nice, without all of the tiers.

    11. Re:Different Services need to be split by Anonymous Coward · · Score: 0

      Actually, they really do need to be split. If Netflix wants to publish data and conclusions based on that data, the data needs to be accurate and useful for answering the question at hand. Currently it is not. If you're going to only take the average of Netflix HD streaming rates per ISP and then compare them country-by-country, you have to translate that into kbps / dollar based on the average price the consumer is paying for the broadband access at the very least.

      But more importantly, I think a big reason Netflix published this data is they're trying to stave off complaints from consumers about crappy HD streaming quality by saying, "Look it's not in our control, your ISPs suck". By taking a per-ISP average and ignoring service tiers, they're basically inventing a false set of data to back up that claim. I call bullshit. I have Verizon FIOS at 50/20 speeds, and against worthy network peers (DSL test sites anywhere in the US, my own servers on gigabit+ backbone uplinks at various datacenters) I can max out those speeds anytime, anyday.

      But when I go to play a Netflix "HD" movie, it's not unusual that I experience bad compression artifacts and/or horrid motion-blur side effects from low-bitrate compression. The problem could be my ISP if they're intentionally filtering Netflix bandwidth to try to strangle money out of them (but that's unlikely, I haven't seen any news about Verizon taking those sorts of tactics). Most likely the problem is on the part of Netflix and/or their CDN providers, and I'm not going to let this bullshit chart try to convince me otherwise.

  16. So where is Netflix hosted again? by Targon · · Score: 1

    The problem here is that you have the difference between the speeds the ISPs can run at and the speeds from the customer to Netflix. I am using the normal Cablevision 15Mbps service, and I DO see speeds up at around 11Mbps. Now, if I have to go through 5 other ISPs to get from Cablevision to the Netflix servers, the problem is the connections between ISPs is where the limits are, not Cablevision itself. The same may apply to the other ISPs out there, where you get the speeds, and you can get to many web sites at the high speeds, but getting to Netflix is where there could be problems.

    Back in April, it seems that Netflix moved to the Amazon hosting....could it POSSIBLY be that they are not seeing great speeds due to Amazon?

    1. Re:So where is Netflix hosted again? by Anonymous Coward · · Score: 0

      Not really. From TFA:

      "As we use a number of CDNs, and our clients can adapt to changing network conditions by selecting the network path that’s currently giving them the best throughput, Netflix streaming performance ends up being an interesting way to measure sustained throughput available from a given ISP over time, and therefore the quality of Netflix streaming that ISP is providing to our subscribers. Obviously, this can vary by network technology (e.g. DSL, Cable), region, etc., but it's a great high-level view of Netflix performance across a large number of individual streaming sessions."

    2. Re:So where is Netflix hosted again? by RingDev · · Score: 1

      Netflix has content delivery networks all over the country with partnerships with most major ISPs. The stream you are getting from them is likely not even coming over the back bone of the internet, but through a CDN linked directly to your ISP regional center.

      -Rick

      --
      "Most people in the U.S. wouldn't know they live in a tyrannical state if it walked up and grabbed their junk." - MyFirs
    3. Re:So where is Netflix hosted again? by Anonymous Coward · · Score: 0

      Netflix uses CDNs (Content Delivery Networks). These are typically distributed over the backbone. Your not going between ISP when you use Netflix. Unless your using an ISP that doesn't own it own wires.

    4. Re:So where is Netflix hosted again? by Anonymous Coward · · Score: 0

      No one said who was at fault or that there was even a fault to argue. These are real world aggregate numbers. It doesn't matter if your get 100MB/sec from point A or B. A lot of people use Netflix and they are showing average numbers in a real world situation that their customers are getting. These are facts with no tilt or slant and it does not contain an editorial or opinions that attempt to guide your opinion in any one direction as to what the results actually mean.
      Apparently many people here are not used to having to analyze data and draw their own conclusions from it and they rely on others to do it for them.

  17. They have to be lumping DSL and Fios together by wernox1987 · · Score: 1

    There is a huge difference between the two, it kind of invalidates the who study.

  18. to how many people? by mapkinase · · Score: 1

    "A Netflix HD stream can provide up to 4800 kbps"

    To how many people simultaneously? Somewhere must be an evidence (assumption) that Netflix servers work at less than 100% capacity. I am completely unfamiliar of these. Saying "A Netflix HD stream can provide up to 4800 kbps" and missing the total number of users trying to connect is not saying much about that.

    --
    I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
    1. Re:to how many people? by bay43270 · · Score: 1

      Are you asking how scalable Netflix's server farm is? They use Amazon's servers. Between the size of Amazon's data centers and Netflix's ability to write Amazon a check, I don't think they will be hitting capacity any time soon.

    2. Re:to how many people? by omnichad · · Score: 1

      I'm thinking that's just the max encoded bitrate they have available for streaming.

  19. Re:Please say it ain't true ! by realityimpaired · · Score: 1

    Wouldn't be the first time. *coughs* 1812 *coughs*

  20. Not Much Difference by hazzey · · Score: 1

    All that this shows is that there isn't much difference between the ISPs. They had to scale the chart (it doesn't start at 0) just to show the differences. As Netflix commented in the linked post, their HD streams are much higher (4.8kb) than these graphs. Of course the graph is just an average, so it doesn't speak to how HD users are affected.

    1. Re:Not Much Difference by guruevi · · Score: 1

      Yes it does, RTFA. They filtered out only HD-capable devices on HD-capable networks with HD-capable movies. They can stream 4.8Mbps but on average only seem to hit 2.4Mbps. You can notice the same thing if you use eg. YouTube or another streaming service at 1080p for long periods of time. I have TWC which advertises me a 10Mbps line which I can seem to burst into for the first 5 minutes of a YouTube video (I watch a lot of e-sports and I have a 27" High-Res screen which YouTube automatically selects 1080p for, those games are typically 15-45 minutes long) however after 7 minutes (and it is timed almost to the second) all of a sudden the stream starts stuttering and needs to be buffered and I fall back to having only 2-3 Mbps throughput to the YouTube server (new streams to other servers and other services are fine). It's not my router, it's not my computer, it's TWC actively throttling my connection even though I am on an unlimited plan. The way around that (according to them) is buying TurboBoost.

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
  21. Wut? by Flambergius · · Score: 1

    I'm I reading this wrong? That's the limit that the ISP can reliably provide, right? Or are those numbers lower than ISP's max because many clients have low-end broadband connection (2M xDSL or something). My ISP can supply sustainably about 5 times that much. I'm on the other side of the Atlantic, but USA can't be that far behind?

    --
    Computers are useless. They can only give you answers - Pablo Picasso
    1. Re:Wut? by Anonymous Coward · · Score: 1

      We are indeed that far behind

    2. Re:Wut? by drinkypoo · · Score: 1

      First, we ARE that far behind. Second, some of these ISPs may be throttling Netflix.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  22. great! by Charliemopps · · Score: 1

    This is exactly what we need. They should run this test daily and have a website dedicated to it.

    Better yet, the FCC should be running the test.

    1. Re:great! by Confusador · · Score: 2

      Better yet, the FCC should be running the test.

      "The Consumer Broadband Test, currently in beta, is the FCC’s first attempt at providing consumers real-time information about the quality of their broadband connections."

      http://www.broadband.gov/qualitytest/about/

  23. ridiculous by Anonymous Coward · · Score: 0

    This is ridiculous. Amost no consumer internet connection will have a "sustained" speed of 4.5Mbps. That's the reason video streaming sites use buffering. There is always going to be dips and peaks in the bandwidth.
    They also may not have tested with the highest-tier service offerings, but only the average consumer level.

    1. Re:ridiculous by omnichad · · Score: 1

      Sustained, in this sense, would probably simply be not having to down to a lower quality stream.

  24. Re:Please say it ain't true ! by mister_playboy · · Score: 1
    --
    Do what thou wilt shall be the whole of the Law ::: Love is the law, love under will
  25. yep by 4d3fect · · Score: 0

    there's my isp waaay at the bottom, no surprises there.

  26. Comcast at #2 by Anonymous Coward · · Score: 0

    How surprising is that? Comcast is routinely bashed for performance - maybe Netflix paid them off?

    1. Re:Comcast at #2 by macpacheco · · Score: 1

      Netflix aparently purchases conectivity at least with L3 networks.
      L3 was forced to pay Comcast in order to avoid Netflix traffic from going through their last resort upstream links currently with Tata telecom.

  27. the least worst of... by TheTrollToll · · Score: 0

    Well lucky me i have the least worst ISP for netflix in america.

  28. These numbers are averages, so they're bullshit by lseltzer · · Score: 3, Insightful

    The larger the ISP, the more they’re penalized by the more rural regions which are limited to DS3 45 Mbps circuits feeding a whole town.

    1. Re:These numbers are averages, so they're bullshit by papasui · · Score: 4, Informative

      I'm gonna toss this out there. I've been designing isp networks for the last 10 years, including some of the ones in the list. I haven't seen an area that only has a DS3 as it's backbone in about 5 years, and even then it was 3 of them. And yes, I have worked in some extremely rural areas where the entire subscriber base has been less than 100.

    2. Re:These numbers are averages, so they're bullshit by silas_moeckel · · Score: 1

      Your point is? Larger ISPs should not be penalized because they screw over there rural subscribers? If the DS3 is the choke point put something bigger in. US ISP's are claiming all sorts of things to avoid upgrading there networks and often being shortsighted when doing those upgrades by putting the cheapest thing in right now but not the cheapest in the long run. In your own example there is no reason to have a DS3 in a ISP network, a single pair of dark fiber can get you what you need today and with one time costs get you to 1.6tbit/s (160 10ge links) today. Telco's are free to put in all the fiber they care to and cable companies can often do the same. The drive to keep this quarters profits high leads to bad decision making.

      --
      No sir I dont like it.
    3. Re:These numbers are averages, so they're bullshit by Telvin_3d · · Score: 1

      This is not a population density or rural/urban issue at all. If it was then the Canadian ISPs wouldn't have averaged so much higher. Canada has the lowest population density in the world and has a significant rural population. However, its lowest ISPs would sit in the top few on the American graph and the top Canadian ISPs would actually sit above the American graph.

  29. Re:Please say it ain't true ! by h4rm0ny · · Score: 1

    We are talking about AMERICA, yeah, the GREAT AMERICA, beaten down by the canajians???

    Please say it ain't true, AMERICA !!!!!!!!!!!!!!!!!!!!

    I once had a meeting with a software partner from a Canadian telecomms company when I worked in that industry who said: "Most of our country is below freezing for half the year. We're good at technologies that mean we don't have to go outside."

    --

    Aide-toi, le Ciel t'aidera - Jeanne D'Arc.
  30. AT&T: Uverse, DSL, etc. by Luyseyal · · Score: 2

    The AT&T numbers will suffer somewhat due to Uverse (VDSL) and regular DSL being sold at varying speeds. While I like the idea of 18-24Mbps, I can't reasonably afford it at this time.

    -l

    --
    Help cure AIDS, cancer, and more. Donate your unused computer time to worldcommunitygrid.org. Join Team Slashdot!
  31. works great on Fios by Anonymous Coward · · Score: 0

    I have been streaming Netflix HD over the minimum service level of Fios (5Mbps) for 3 years now with nary a hickup. We use it in place of cable (Fios does not provide TV in our area).

  32. Re:Please say it ain't true ! by crashumbc · · Score: 1

    And with the exception of Canada, the US is bigger then all those countries above it combined...

  33. Rogers has Brutal CAPS & Throttling. by Analog-X64 · · Score: 1

    What this article forgets to mention is Rogers has Brutal bandwidth caps and speed throttling, and is very very expensive.

    1. Re:Rogers has Brutal CAPS & Throttling. by Dorkmaster+Flek · · Score: 1

      That's why I use TekSavvy's cable service instead of Rogers. Rogers has bandwidth caps so ridiculously low that there's literally no point to having a connection that fast. It's obscene how broken Internet service is in North America across the board. Both the FCC and the CRTC seem disinclined to do anything about it, however the whole usage-based billing thing is gaining quite a bit of attention in Canada right now. I'm glad to see people actually getting upset about it. Maybe it will have some impact after all.

      --
      I like to think of online DRM as something akin to a college -- you pay for lessons until you learn something.
  34. please use real speeds by KiloByte · · Score: 4, Insightful

    only problem is, running full tilt I can go through my monthly bandwidth cap in eleven and a half hours.

    We badly need a "truth in advertising" law that would make it illegal to label a "100Mbps connection" with a 5GB monthly cap as anything above the 16331bps it really is (yes, less than 16kbps, this is not an error). Providing a bigger burst is ok but only if that's clearly marked as such.

    Toss in something about the scam that lets ISPs call 100Mbps down/128kbps up by the bigger number. If you want to use just one number, you'd need to print the lower one. Anything else is deceiving the customer.

    --
    The creatures outside looked from Alt-Right to Antifa; but already it was impossible to say which was which.
    1. Re:please use real speeds by clodney · · Score: 1

      We badly need a "truth in advertising" law that would make it illegal to label a "100Mbps connection" with a 5GB monthly cap as anything above the 16331bps it really is (yes, less than 16kbps, this is not an error). Providing a bigger burst is ok but only if that's clearly marked as such.

      Toss in something about the scam that lets ISPs call 100Mbps down/128kbps up by the bigger number. If you want to use just one number, you'd need to print the lower one. Anything else is deceiving the customer.

      If you are attempting to compute the speed by dividing the bandwidth cap over the cap period, you end up with a meaningless figure. I don't use my computer 24/7, and the fact that I get 20Mb service for the hours that I want it, and never worry about my bandwidth cap means that for me (and no doubt the vast majority of consumers) the speed and the total volume need to be quoted separately.

      Same with up/down speeds - asymmetric connections don't bother me, because my usage is asymmetric as well.

    2. Re:please use real speeds by MikeBabcock · · Score: 1

      Cogeco's website has now clearly marked the bandwidth limits for each broadband speed here in my part of Canada.

      Sadly, here's the list (from their site):

      $25 - 3Mbps - 10GB/mo
      $32 - 7Mbps - 30GB/mo
      $42 - 14Mbps - 60GB/mo
      $77 - 16Mbps - 125GB/mo
      $60 - 30Mbps - 125GB/mo (limited availability)
      $100 - 50Mbps - 150GB/mo (limited availability)

      The frustrating package difference for me is from $42/mo to $77/mo for almost no speed increase but an extra 65GB/mo. Interestingly, the over-use charges for the $42/mo service is capped at $30 (making it $72/mo if you download hundreds of gigs a month).

      I have no problem with this so long as its clearly marked as it is now. What I do have a problem with is that some of these limits are insanely low for today's usage. Downloading Qore each month on the Playstation 3 takes a gig or two. Downloading game demos takes another 1.5GB each on average. And that's not including my PC usage, video streaming, or actual online play.

      --
      - Michael T. Babcock (Yes, I blog)
    3. Re:please use real speeds by Kris+Warkentin · · Score: 1

      Well, it's not exactly a burst speed - I can run flat out at 3MB/s all day long. I see your point though - with the cap it amounts to a much slower overall rate if I want to stay off my limit.

      --

      In Soviet Russia, hot grits put YOU down THEIR pants.
    4. Re:please use real speeds by arth1 · · Score: 1

      Toss in something about the scam that lets ISPs call 100Mbps down/128kbps up by the bigger number. If you want to use just one number, you'd need to print the lower one. Anything else is deceiving the customer.

      It's worse than that. Those numbers are the "up to" numbers, and not the CIR.
      Personally, I would much rather have a 1000-1500/256-256 line than a 0-6000/0-768 line.

      Worst of all is, perhaps, the "speedboost" marketing term. They sell that they throttle your service after a while as an advantage? Yeah, for them, it is.

  35. Re:Verizon by TaoPhoenix · · Score: 1

    Yeah, I have a Verizon 3.0 DSL line.

    Recently, with the exact same plan, Hulu suddenly seems a lot slower. I just recently had to start buffering the show again starting about a month ago, but all I can tell is my line is the same it has been. I don't know if there's advanced throttling going on to support the whole Hulu Plus push.

    --
    My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
  36. Well, I am sure one of... by Anonymous Coward · · Score: 0

    I am sure one of the ISPs will consider it harassment and sue.

    Maybe.

  37. What is the minimum bit rate for non-HD? by Anonymous Coward · · Score: 0

    I've been able to get an acceptable non-HD stream from Netflix using Clear.

    Yadda, yadda anecdote, of course.

  38. I see all the big ones but... by SamuraiHoedown · · Score: 1

    Where is Xfinity? I don't see it on this list.

    1. Re:I see all the big ones but... by coldsalmon · · Score: 1

      Xfinity is Comcast. In case you weren't being sarcastic.

  39. Hm. by JustAnotherIdiot · · Score: 1

    Yet another reason Canada is better than the US, I miss being able to make fun of them.

    --
    What do I know, I'm just an idiot, right?
  40. Kill Bots? by Anonymous Coward · · Score: 0

    Beware Canada!

    According to the graph, Canada is producing Kill Bots at an alarming rate!

  41. Re:Please say it ain't true ! by S.O.B. · · Score: 1

    I once had a meeting with a software partner from a Canadian telecomms company when I worked in that industry who said: "Most of our country is below freezing for half the year. We're good at technologies that mean we don't have to go outside."

    Your friend was either exaggerating or was misinformed.

    Most major cities in Canada, and hence most people, only get about 4 or 5 months of below zero temperatures and cites like Vancouver may only get a few days. You have to go north of the Arctic circle before you get cities that have six months below freezing. Even Whitehorse has average temperatures above 0 for 7 months of the year.

    That's not to say that when it does get cold it isn't $#^%$# miserable but it doesn't last half the year.

    The reason we're good at communication technologies is not because it's too cold, it's because we have 35 million people in a country bigger than the U.S.

    --
    Some of what I say is fact, some is conjecture, the rest I'm just blowing out my ass...you guess.
  42. Misconceptions by m0masb0y · · Score: 2

    Netflix streams at rates that effectively consume the entire subscriber subscription rate for hours on end. It is obvious from the postings which are being made that most who are posting are end users with no network design knowledge. So, here goes.

    Netflix streams at 1.0-6.0 Mbps at a fairly constant clip the entire time a movie is playing. There are exceptions to this behavior. But, for the most part, that is the picture the network engineering folks see. Now, take a rural telco of 5000 access lines that offers DSL to 100% of its customer base. Today 2000 of the 5000 lines are subscribed and running ADSL2+FAST and 90%+ can attain speeds of at least 3 Mbps. Most can attain much higher speeds.

    Of the 2000 DSL subscribers, lets assume that only 1% are Netflix customers (arguably a low number). If 1% are Netflix customers, then twenty customers will be consuming 60Mbps of bandwidth at 100% capacity constantly. If 10% are Netflix customers, then 200 customers will be consuming 600Mbps of bandwidth at 100% capacity constantly. This rate is twice the rate of the purchased bandwidth capacity for the entire ISP. And this assumes only 3 Mbps service.

    Now lets assume that those 200 customers are paying $29.95 for their 3 Mbps service. The total revenue for those 200 customers would be $5,990.00 per month. That amount is almost $3000.00 per month less than the cost of the upstream port capacity without any transport fees. Oh and the bandwidth being purchased for the entire ISP is only 300Mbps redundant capacity. BTW - There is no regulatory recovery for Internet bandwidth. So, please don't play the telco monopoly card.

    So, there are a bunch of end users who think you are getting the short end of the stick from the ISPs. Think again. Just because Netflix wants a free ride to the end customer and the end customer wants to use 100% of their purchased link on the cheap doesn't make the ISPs the villians. With this kind of network utilization, the ISPs are no longer able to make any money on their services. As well, they are now having to rebuild their networks to make sure they can support over the top video services like Netflix. And, Netflix is rating the ISPs???!!! That is like the fox guarding the chicken coop. Why would anyone want to hear from Netflix how well ISPs perform for their service when Netflix isn't paying for the network or bandwidth.

    --
    A smile and a hug never hurt anyone ;-)
    1. Re:Misconceptions by RingDev · · Score: 1

      Actually Netflix is paying for the CDN services, so the ISP isn't paying for upstream. Heck, in the case of Comcast, Netflix is even paying THEM too accept the CDN connection instead of going over their upstream providers.

      The ISPs have over sold their lines. No surprise. It's not possible for an ISP to provite 100% (2000 customer in your example) 3Mbps service consuming 6Gbps of bandwidth for $30,000 in revenue. They can't even provide an order of magnitude LESS than that for $30,000.

      Historically though, thanks to usage patterns, ISPs could grossly over sell their pipes, because significant quantities of viewers wouldn't all be hitting their cap constantly.

      Usage patterns have changed. And ISPs will need to alter their sales patterns and pipe sizes to deal with it. Surprise, surprise, they can't keep selling the same 10 year old pipe to more and more users who have higher and higher demands.

      -Rick

      --
      "Most people in the U.S. wouldn't know they live in a tyrannical state if it walked up and grabbed their junk." - MyFirs
    2. Re:Misconceptions by gamricstone · · Score: 1

      What percentage of an ISPs overall operating costs does bandwidth normally represent? With 2000 customers at $29.95 thats just shy of $60,000 monthly. Your "upstream port capacity" costs less than $9,000 per month currently. You forgot one key part of your post, the part which explains why the ISP cannot purchase more bandwidth in order to satisfy its customers.

      The problem is always greed, not technical limitations.

      --
      The economic anarchy of capitalist society as it exists today is, in my opinion, the real source of the evil. - Einstein
    3. Re:Misconceptions by halltk1983 · · Score: 1

      Hardware and distribution costs. DSLAMs cost money. People to answer the phones cost money. Expanding service costs money. The backhauls to the COs cost money. In the example ~ 10-12% of the cost goes to the bandwidth. Most of the rest is taken by other parts of delivery. Margin on that would be quite slim as it is. So it's greed to not want to lose money on a product?

      --
      Watch for Penguins, they eat Apples and throw rocks at Windows.
    4. Re:Misconceptions by gamricstone · · Score: 1

      Nope it sure isn't greed to make money off a product or service, but I've never seen any numbers even HINTING that they need more money in order to provide a functional, profitable, and quality service. The fact is you have provided no evidence that the margin is in fact 'quite slim' and thus extra bandwidth cannot be purchased or installed.

      --
      The economic anarchy of capitalist society as it exists today is, in my opinion, the real source of the evil. - Einstein
    5. Re:Misconceptions by halltk1983 · · Score: 1

      Can't post numbers without violating my NDA, but I work for a small wISP, and I assure you that the margin per customer is relatively small by percentage.

      --
      Watch for Penguins, they eat Apples and throw rocks at Windows.
  43. Re:Verizon by Anonymous Coward · · Score: 0

    We have two Verizon DSL lines at work for our clients to use. During certain parts of the day, both can barely stream Youtube videos without buffering. Verizon support SUCKS. When I call, they refuse to acknowledge I have two separate business DSL lines with static IP addresses and both have problems. They start with having me reboot the "computer", checking the eth cable, they have sent me new DSL routers, this list goes on. I finally gave up and our guests live with crappy speed. Both are attached through their own M0nowall based router and there are no rogue PC's taking up the bandwidth. They just suck. It might be a quick fix on their end in the CO but I just don't have the hours it takes to be on the phone with them.

  44. Blame ignorance and then Excel by swb · · Score: 1

    The first thing is to blame the chart maker for not having any education in how to visually display quantitative information and for not putting any thought into people trying to read the chart and differentiate between a number of very similar colors.

    There's a number of things the chart could have that would have helped -- ensure no two similar colors are close (black and brown should be in the top band instead of 3 blues), callouts to indicate which line was which instead of relying on a color-coded legend.

    The next thing to blame is Excel. Excel should be reprogrammed so that once the first six or eight easily distinguished colors are used, additional chart elements should switch to a patterned line/fill (and a different color). This by itself would solve all the problems and make uneducated chart makers look a lot smarter.

    1. Re:Blame ignorance and then Excel by Rude+Turnip · · Score: 1

      I don't think Excel is to blame. I've made charts with lots of different data series and Excel 2007 automatically used a set of colors that made each series stand out from the other.

  45. only get that if they will peer/colo with you by Revek · · Score: 1

    There would have to be some serious peering/colo for any isp to be able to provide that. We will have to raise our rates. All of us have to raise our rates. We simply have more people using it than we used to. Even two years ago our little isp didn't ever top out our 45 meg connection. Today we have upgraded two two times and still have major loads during the evening. I am the one man tier2 at this company. I hear a lot of people upset that they can't get the supermega(tm) stream from nexflix. We also sell media(tv). You can asume that the telcos and backbone providers will be the only ones who will profit by this. I would also really love to crash netflix by giving them what they want im betting they couldn't handle giving all their customers that speed.

  46. People are starting to understand it... by RulerOf · · Score: 3, Interesting

    label a "100Mbps connection" with a 5GB monthly cap as anything above the 16331bps it really is

    I completely agree.

    I'm getting more and more pissed off as greater numbers of people are beginning to understand what monthly transfer caps are, and then proceed to voluntarily or forcedly believe the outright fucking lie that these assholes are perpetuating (and that most folks here also believe!):

    [bullshit]

    Enforcing a limit on data transfer over a given period of time is a very direct and extremely effective method of completely alleviating the problems that can be caused by a small number of users consuming most or all of the bandwidth on any given, shared network segment.

    [/bullshit]

    The statement that you just read, enclosed in [bullshit] tags, is 100% bullshit.

    It's a problem with bandwidth, not a problem with transfer. Don't ever believe the utter lie that these two concepts are inherently and directly correlated. While they can be correlated, they do not have to be, and, in true Slashdot spirit, they are most certainly not causal.

    Lastly, if you don't understand WHY what I say is true, think of it this way:

    Take a look at the gigabit switch sitting on your desk (or pretend you have one). You've used it very lightly. You just browse the web through it. Maybe some games. No youtube, no torrents, no downloading. You've owned it for a while now, and you've transferred about 10 gigabytes of data through it in that whole time.

    I own the exact same model, and I'm coming over your house later and swapping out switches with you, but the difference between your switch and mine is that I pushed 5 terabyes of data through mine every single day I've owned it.

    Given that neither switch is defective, when I switch hardware with you, will you notice the difference?

    The worst part about this whole thing is that bandwidth is a fundamental commodity and property of multi-segment interconnected networks in general (read: the Internet). It's so fundamental that, rather than paying specifically for the connection speed of a physical link into someone's network, ISPs pay specifically for bandwidth usage based on a well accepted model commonly called 95th percentile billing because of how fairly and accurately it reflects a given link's impact on the network. Overall transfer over a given period, while it may be calculated, is irrelevant because the amount of data pushed through a link simply doesn't fucking matter. Data transfer at all levels of a network is a function of bandwidth, not the other way around. Were you to graph it out as a function, as time approaches infinity, transfer does as well.

    On behalf of the ISPs though, this misconception and billing model is absolutely genius. If I literally possessed a LIMITLESS source of product (data transfer) and, irrespective of size, somehow managed to convince you that it was reasonable for me to charge you for a finite, expiring quantity of it, I'd laugh all the way to the fucking bank every time you came back for more.

    --
    Boot Windows, Linux, and ESX over the network for free.
    1. Re:People are starting to understand it... by snookums · · Score: 1

      It's a problem with bandwidth, not a problem with transfer. Don't ever believe the utter lie that these two concepts are inherently and directly correlated.

      How can they possibly not be correlated?

      Transfer = Bandwidth x Time

      In the case of a monthly billing cycle, Time is a constant (~2.6 Ms), so Transfer is directly proportional to bandwidth.

      Lastly, if you don't understand WHY what I say is true, think of it this way:

      Take a look at the gigabit switch sitting on your desk (or pretend you have one). You've used it very lightly. You just browse the web through it. Maybe some games. No youtube, no torrents, no downloading. You've owned it for a while now, and you've transferred about 10 gigabytes of data through it in that whole time.

      I own the exact same model, and I'm coming over your house later and swapping out switches with you, but the difference between your switch and mine is that I pushed 5 terabyes of data through mine every single day I've owned it.

      Given that neither switch is defective, when I switch hardware with you, will you notice the difference?

      This is a complete straw man. The reason why transfer caps are in place is not because the equipment is wearing out. My transfers this month don't affect anything next month. The analogy that you want is that you're coming over to my house and you're going to download your 5 TB through my switch. This is the reality of ISP networks -- shared infrastructure. Will that affect my experience? Quite possibly, because at times of heavy burst we'll be contending for the resources. If, on the other hand, I say "sure, come to my house but you can only have 500 GB" then you will, almost by definition, be contending for bandwidth less often than if I let you download 5 TB through my switch.

      ISPs pay specifically for bandwidth usage based on a well accepted model commonly called 95th percentile billing because of how fairly and accurately it reflects a given link's impact on the network. Overall transfer over a given period, while it may be calculated, is irrelevant because the amount of data pushed through a link simply doesn't fucking matter.

      How do you expect a consumer, who knows nothing about computer networks and just wants a fixed bill each month, to handle 95th percentile billing? It's completely unfeasible. It's very easy to say "Joe, you buy this plan and you can watch about 300 hrs of the youseTubes every month". Transfer is also very easy to track, and provides a nice "cut off" point, where the speed can be throttled by the ISP, or whatever they do to you over the cap.

      I know some ISPs, who normally run with monthly transfer caps, that have plans based on a rolling usage vs priority system. As you transfer more they lower the QoS on your packets. This means that if the network isn't congested you can use as much as you like. This is a pretty fair system, but it's very difficult to explain and advertise to most end users.

      If I literally possessed a LIMITLESS source of product (data transfer) and, irrespective of size, somehow managed to convince you that it was reasonable for me to charge you for a finite, expiring quantity of it, I'd laugh all the way to the fucking bank every time you came back for more.

      Now it's your turn with the bull shit. The ISP in no way has a limitless amount if data transfer. Each month they have exactly their total bandwidth x (number of seconds in the month) to sell, use it or lose it. So, they portion it up sell it to you on the exact same basis -- some GB per month, use it or lose it. Now, they do oversell, and they do tend to sell you more than you'll use - relying on the light email user not to use the full 50 GB that he's paid for. However, they really have to

      --
      Be careful. People in masks cannot be trusted.
    2. Re:People are starting to understand it... by RulerOf · · Score: 1
      The switch-swapping example is apt for those who don't understand the concept, really, but you pointed it out yourself:

      The reason why transfer caps are in place is not because the equipment is wearing out. My transfers this month don't affect anything next month.

      QED: transfer doesn't matter. Only bandwidth does.

      Like I pointed out, as time approaches infinity, potential transfer does as well. Bandwidth, on the other hand, remains constant. Transfer is the primary effect of using bandwidth irrespective of how much or how little of it that you do over any given period of time.

      Now for the most important concept: Bandwidth is only relevant at instantaneous points in time. How much available bandwidth you had an hour ago is completely irrelevant to how much bandwidth you have available for use now, or X seconds from now. Therefore, it follows that the effect (read: transfer) of using bandwidth an hour ago is likewise irrelevant to using bandwidth now.

      The problem isn't that billing for discrete units of transfer is a difficult concept to understand, it's that it's fundamentally wrong and similarly unfair. The ironic thing is that you actually do understand why, but I think your position on the argument is reversed because you benefit from the way things are under this model:

      I want to know two things: How much can I have, and how fast can I have it. I care very much about the burst rate of my connection. It's probably the most important thing to me, because I hate waiting. Other people like to fire and forget a bunch of movie torrents, not caring how fast they are, and watch them when they're done. That person will probably care more about how many they can fire off before they use more than their allocated share of the transfer pool.

      You value bandwidth as the actual commodity it is. It's a tangible thing that you understand, and you want as much of it as you can get whenever you happen to want it. I want my web browsing to be this way too, but I couldn't care less if my 1.5 GB file transfer took 1 minute or 10 because Joe Blow down the street wants to watch his YouTube or check his email.

      So, they portion it up sell it to you on the exact same basis -- some GB per month, use it or lose it. Now, they do oversell, and they do tend to sell you more than you'll use - relying on the light email user not to use the full 50 GB that he's paid for.

      In the absence of a transfer cap, that statement IS true. Additionally, the fact that I'm willing to wait 10 minutes instead of 1 to download that 1.5GB file shows a gracious, restrained use of that shared bandwidth pool. Yet with a transfer cap in place, that behavior gets completely and unfairly shat upon. And, in a complete twist of irony, when I breach the cap and pay extra, we end up with ME subsidizing YOUR connection, not the other way around.

      --
      Boot Windows, Linux, and ESX over the network for free.
  47. silly by buddyglass · · Score: 1

    They don't differentiate between offerings from each ISP. An ISP the majority of whose users are on slow 1.5 Mb/s connections (but with a few on super-fast 20 Mb/s connections) would fare poorly in this exercise compared to a competitor with of its users on mediocre 4.0 Mb/s connections. As a consumer interested in bandwidth, though, I'm going with the first ISP assuming their backbone is capable enough to actually feed that 20 Mb/s pipe.

    Where I live the two highest-performing options are Time Warner (Roadrunner Extreme) at 20 Mb/s for $50/mo and AT&T's U-Verse (Max Turbo) at 24 Mb/s for $65/mo. Currently I have AT&T's "Elite" DSL service which is 6 Mb/s for $40/mo. What I would find interesting is whether each of those can support the full 4800 kb/s Netflix stream, or if backbone issues crop up.

  48. Is Netflix airing the industry's dirty laundry? by joeszilagyi · · Score: 2

    Is the real offense here by Netflix, as evidenced by comments there on the site, is that they've exposed the dirty little secret of many ISPs--that they oversell capacity?

    --
    Dude, where's my packet?
    1. Re:Is Netflix airing the industry's dirty laundry? by Revek · · Score: 1

      No we don't oversell capacity. We don't want to top out. But a new thing has came in and it isn't helping. I said this in another post but you better get ready guys. If things keep going the way they have expect to pay more. The net here in Arkansas needs many more connects. not enough hubs screwed up routing and no clear standards of operation. AT&T is itself is composed of all its re-acquisitions. Ameritech and some other I can't remember the name of are who we deal with here in Arkansas. It was the same people I always talk to until one day the bills arrived AT&T. They don't communicate with each other that well it seems. They have few resources down here and are not investing in any real upgrades. We have upgraded the bandwidth here to keep a little buffer. Now we need a whole oc3 to even out. This has happened in less than two years. You gripe at all the big company's but you don't want to pay more. File sharers who won't buy a DVD in the five dollar markdown bin are usually the ones who complain the most. I have spent my last 2 months trying to fix this nightly suck and have concluded that its netflix or the rest of my customers. But since i can't charge netflix I guess ill have to charge our customers to pay for it. if i give them more they just will download at a higher res. and from that purty graph that the article had I don't have a chance.

  49. Numers are ok, Graph is skewed. by eepok · · Score: 1

    The graph doesn't accurately represent the data. Note the Y-Axis. It's values are as follows:
    - - - - - - - - -
    3000
    2800
    2600
    2400
    2200
    2000
    1800
    1600
    1400
    1200
    1000
    - - - - - - - - -

    Without taking the axis all the way down to ZERO, the graph gives the impression of greater disparity than there actually is. You see this frequently with tech review sites where they start a bar graph at "140fps" so that a difference of 3fps looks epic and the shelling out of $40 more on one video card is suggested.

    1. Re:Numers are ok, Graph is skewed. by sp0tter · · Score: 1

      That is typical. No reason to waste vertical space that shows little useful data. I would prefer that they zoom in on the area of interest anyway. As long as the scale is consistant, ie the y-values dont go 1000, 1100, 1500, 2000, 3000, 3100, 3200, 4000 to distort disparities. This is pretty much standard practice.

      --
      you don't eat crackers in the bed of your future--or else you'll get all scratchy
    2. Re:Numers are ok, Graph is skewed. by Dynedain · · Score: 1

      Is anything under 1Mbps even useful information? 768Kbps isn't even allowed to be called "broadband" anymore.

      --
      I'm out of my mind right now, but feel free to leave a message.....
    3. Re:Numers are ok, Graph is skewed. by eepok · · Score: 1

      Standard, yes... but still bad. It amplifies what could be negligible differences... mountains and mole hills, etc.

  50. Stay away from Frontier by tekgoblin · · Score: 1

    I have frontier DSL with only 3mb down and 1.5 at night... It is horrid sometimes I can only get 1 bar of quality from Netflix and this is their highest plan in my area even though I am right next to fiber lines to Purdue University. Frontier needs to wake up and build a better network IMO. Also Verizon shouldn't have sold out.

    1. Re:Stay away from Frontier by mikestew · · Score: 1

      I thought Comcast was full of crap when they ran scare ads about Verizon selling there fios to Frontier in Seattle. Just because it turns out they were right doesn't mean Comcast isn't full of crap, of course. With Verizon, it was always 25Mbps and I never touched the router. Since Frontier took over the bandwidth is mostly what they advertise, but not reliably so, and I reboot the router (same one that I used for Verizon) almost daily due to connection issues.

      I'm not so unsatisfied that I'd go back to Comcast (Frontier CEO would have to be caught killing puppies before I'd even consider it), but there was definitely a drop in quality since Frontier took over.

  51. I'm astounded by Fizzol · · Score: 1

    that's there's actually someone below Centurytel.

    1. Re:I'm astounded by BigSes · · Score: 1

      Atlantic Broadband is so low, it doesn't even make the chart! I kid, I know they don't meet the size criteria for analysis, but seriously, they suck.

  52. And this surprises.... by Anonymous Coward · · Score: 0

    the low throughput in the US is surprising because... ?

    It doesn't surprise me one bit. The good old USA has long ago given up being first, or the best in lots of things, except perhaps for making tons of $$ off of people who can least afford it.

    Why bother to have the BEST ISP in the country for your customers, when you can be the richest CEO of an ISP company?

  53. Help ... Which one is Frontier? by Sam_In_The_Hills · · Score: 1

    Could you tell me, by count from the bottom left, which one is Frontier? And but the way, I prefer chromatically challenged. Thanks.

    --
    Linux -- the Ultimate Windows Service Pack
  54. Network Neutrality? by Anonymous Coward · · Score: 0

    Its possible ISPs may be filtering streaming services like Netflix. That the bandwidth shown on the graph is only what the ISPs are allowing in speeds to customers. Maybe this data release by Netflix is their response to news from the other day-
    http://politics.slashdot.org/story/11/01/26/1820236/Senators-Bash-ISP-and-Push-Extensive-Net-Neutrality

  55. my ISP clearly states the caps by Chirs · · Score: 1

    I'm actually okay with giving burst speed and caps...it's pretty easy to transform this to constant bandwidth.

  56. Graph for the Color Blind by glodime · · Score: 3, Informative
  57. Re:Please say it ain't true ! by Coren22 · · Score: 1

    The reason we're good at communication technologies is not because it's too cold, it's because we have 35 million people in a country bigger than the U.S.

    I would even add to that, the majority of those people are clustered into a relatively few cities, whereas the US, people live all over the country. I would be curious to see a comparison of the percentage of population within 20 miles of a city between the US and Canada.

    --
    APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
  58. netflix isn't getting a free ride by Chirs · · Score: 2

    Netflix is paying for the bandwidth on their end, so it's not like they're getting anything for free. The issue is that all of the sudden people want to stream massive amounts of information and it's breaking all the assumptions that went into oversubscription ratios.

    The answer is for ISPs to be honest about what they provide. If there's a monthly cap, say so.

    That said, it would be good for netflix users to be able to pick a speed and have it stay there rather than sucking up all available bandwidth.

  59. I have a 1.5 Mbps Qwest DSL connection by Rick17JJ · · Score: 1

    I have a 1.5 Mbps Qwest DSL connection. I have been seeing those constant Qwest "Heavy Duty Internet" commercials on TV advertising speeds up to 40 Mbps. But, the only speed available where I live is 1.5 Mbps for $39.99 per month. I confirmed that by going to the Qwest webpage and typing in my telephone number to see what speeds are available for where I live.

    The comparison chart in the article, shows Qwest having a typical Netflix performance speed of somewhere around 1800 Kbps.

    According to my DSL modem, I am currently connected at a downstream rate of 1536 Kbps and an upstream rate of 576 Kbps. At least my download speeds are consistently close to that speed.

    I am not actually complaining, because I am happy that I am no longer stuck on dial-up. They finally made DSL available here several years ago, shortly after digging a 3 mile long ditch and putting in a couple of new underground conduits, and adding another new small windowless building for the DSL stuff. I live in a city of about 50,000 people, about 1/4 of a mile from the nearest small windowless Qwest building that has a DSL switch in it. Since I am so close to their nearest DSL switch, I am surprised that faster speeds are not available here.

    I do not have satellite or cable and am planning to sign up for Netflix. I plan to use the one at a time DVDs by mail plan for $9.99 per month. I do not know if the slowest Netflix steaming method would work here or not.

  60. NumBers are ok, Graph is skewed. by eepok · · Score: 1

    NUMBERS, damn it. >.

  61. Rogers by kyrio · · Score: 1

    Great, I can stream a whole 3 movies before I hit the bandwidth cap for the month!

  62. Contrary to popular beliefs by Anonymous Coward · · Score: 0

    I'm in Canada on Telus. I was and have been disappointed with their service for price for quite a while now. I had assumed that with more ISP's in the US, there would be greater competition, and therefore, better service for less money. What happened? Is it that each of these providers has their own little fiefdom, which they defend like a hungry wild animal over a freshly killed carcass, and anyone who tries to move in has a blood sport on their hands? You cannot argue that 'Canada is a very huge population in a tiny little space, so high speed broadband is easy to implement'. You just can't make that story stick, its too big a lie. I still think high speed internet is vastly more expensive in Canada than it needs to be (with far too little service to justify the cost). The US? How did they fall off the technological bandwagon (and I'm not talking about "oh, Freddy such-and-so has quadruple fiber links with a 50 terabyte monthly cap for 4 cents every 6 months", I'm talking about the average).

  63. Telus by jbr439 · · Score: 1

    I'm with Telus on the 6 Mb/s (dowload) plan. And I can get about 600 kB/s (that's bytes) when downloading a torrent, which is far in excess of what the netflix chart shows. Just my 2 cents.

  64. Frontier in the dog house you say? by Red_Chaos1 · · Score: 1

    Being that I work for a call center contracted to do tech support for Frontier I'm not the least bit surprised. There *are* customers who get like 7Mbps connections, but they're like 1 in 100 or worse. The vast majority are 1.344 or 1.5Mbps. The rest are 3Mbps or something ridiculous like 768K which balances out. The buyout of Verizon's copper did not help. VZ had shit running with bubblegum and shoestrings.

  65. Re:Please say it ain't true ! by S.O.B. · · Score: 1

    I would be curious to see a comparison of the percentage of population within 20 miles of a city between the US and Canada.

    I couldn't find that particular stat but according to Wikipedia 80% of the population lives within 150km (93 mi) of the U.S. border. NationalGeographic.com estimates it as 75% within 161km (100mi).

    --
    Some of what I say is fact, some is conjecture, the rest I'm just blowing out my ass...you guess.