internet speeds confusion
I did a few speed tests (at diff. times of diff. days) and came up with ~1.15 down(including one off of my IPs page). The upload speed was just over those advertised. I called them up and they ran a test coming up with 4.3 mbps. I asked about the difference, nobody knew, they didn't even know the test was available (on their page). I ran their website test again still hovering around 1.1.
Kinda forgot about it for a couple days, starting watching some videos (Top Gear) and it seems everything is loading much faster than usual. I've run a few more tests and now I'm at ~2.15 consitently for the last few days.
I'm sure it's simple, but can anyone explain? Thanks.
Last edited by tdister; Dec 6, 2005 at 02:31 PM.
Here's a couple links regarding speed:
http://www.dslreports.com/speed
http://www.dslreports.com/faq/258
Nothing simple about speed and the internet. So many variables out there beyond your or your ISP's control. Your cable internet provider might be sharing the connection, causing your actual download speeds to be lower due to traffic on your section (node). If your neighbors are all banging away on the internet, that might impact you.
I just ran a test here, got 1.74 on one test, and 2.3 on another (testing from different sites). Test was 30 seconds apart. Should have a 3mbps DSL connection here. Sometimes it just isn't there.
Hope those pages help clear up the confusion.
I understand that my line is shared and will be diff. at diff. times, but why does their test say 4.3? Are they using a different test? It would seem to be false advertising to say "You can have 4mbps,,, test you current speed now" all in the same line if they're diff standards. I find it hard to believe the fluctuation are that severe and only get over 4 when they run the test just by luck. I think they bumped me up a notch after I called to ask about it, neither computer has loaded this fast before. Ever.
So is it worth upgrading my service if I can't get my advertised speed now? I don't expect it to be 5mbps, but would it increase the same 25%, or just suck more $$ out of me?
Thanks for the help.
Last edited by tdister; Dec 6, 2005 at 03:30 PM.
the tests are going to never be dewad on, cause theres no telling how many computers are hitting the same test server at the same time, distance from you to the test server, number of hops to the test server.
It makes perfect sence that your cable cop's test is faster since youre on the same network as them. there are no external hops over different backbones.
There are only a very, very select few cable modem users that get a guaranteed amount of bandwidth, and they pay a hefty premium for that allowance. 99.9999% of cable modem users get "best effort" profiles which will do just that...provide the best effort at giving you your alloted bandwidth.
Without going to far, here is a little insight:
- All cable modem traffic is on shared bandwidth--both upstream & downstream. To over-simplify: Your upstream is shared with your immediate neighbors. Your downstream is shared by everyone in your general area. This will all depend on how well your local cable company manages its capacity.
- Download/upload speeds depend on every piece of network gear & transport they pass through between you & your download source/upload destination. So, any bottleneck between point A & point B will be seen on your side.
- Speed tests are very subjective. In addition to the first two bullet points, your speeds can also be affected by your modem itself. Out of date firmware or DOCSIS spec could drastically reduce what your modem is actually capable of. To be more specific (warning: geek talk ahead): A DOCSIS 1.0 modem generally will not perform as well as a DOCSIS 1.1 or 2.0 capable modem in a DOCSIS 1.1/2.0 environment. DOCSIS 1.1 & 2.0 capable modems, in a similarly capable RF plant environment, gain access to more advanced methods of data packaging. So, if you are running a 1.0 modem, your modem will not be capable of taking advantage of these lovely little data packaging methods. (Horribly over-simplified, but, hey, what do you expect?)
- The health of the local cable plant has a very intimate relationship with your speeds. Poor RF health will corrupt the signals to/from your modem leading to packetloss. And packetloss will lead to slower speeds, as one side or the other (if not both sides) will have to continually resend data that got corrupted in transit across the cable plant.
- There are virtually no sites on the Internet that will give you anything close to the download speeds that the average CM/DSL subscriber is provisioned for. The ONLY thing I can think of that a subscriber can run that will actually make use of their entire downstream bandwidth is a Bit Torrent client, and this is only because you would be downloading little pieces of a file from tens, if not hundreds, of other users, at the same time. Outside of using Bit Torrent extensively, all of the cool, high bandwidth service levels are pretty much just advertising at their finest...
Trending Topics
Ford Trucks for Ford Truck Enthusiasts
I never got a call from anyone, but I think I'll not bug them anymore. I guess it pays to ask questions. Now if I'm waiting it's because of the other end, and most videos will load faster than they play. I used my grandparents dial-up last weekend and got too frustrated after about 5 min, and that was just checking my email! LOL
-Matt
have your DSL provider check your line for stresses. Check you actual download at a couple of site while downloading good size file. it should stabilize depending on your speed of service at leat at 100k. don't let them give you the BS that that DSL can't run any faster than 786 and cap you.








