This new Government site was in a news article on AOL today. We can test our ISP actual speed for download and upload. It just went into service on a couple days ago.
My standard comcast tested out at about 3mbps both ways. not bad, I guess. It sure beats my old dial up connection last year.
download 7814 a second time 21182 kbs almost 250% difference
upload 3038 second time 3911 almost 30%
comcast. consistency is their byword
WOW, testing a couple more times with both test engines i get 6.7 and 3.2 mbps and then 12 and 2.3 mbps. pretty consistant…
3mbps both waysthats not a good downstream speed. another thing to keep in mind is that the way our network shapes traffic these tests are not always accurate. we allow high throughput bursts for large file downloads that would make your speeds look faster than they are.
Heres the bottom line. Speedtests and ISP speedtiers are relative and are almost useless without the proper context. We have a 50mb/s speed tier available now but guess what? That file you downloaded from yourfavoritesite.com yesterday at 300k/bs with your 6mb/s speedtier would download today at a whopping 300kb/s with a 50mb/s speed tier! why? because no matter how fast you are able to receive data, your only going to get it as fast as the site you are connecting to is willing to give it to you which is highly regulated. paying for a higher throughput speedtier is only good if you have mutiple, simultaneous connections or a designated point to point in network connection.
[quote=“logans_daddy, post:4, topic:2763”]
3mbps both waysthats not a good downstream speed. another thing to keep in mind is that the way our network shapes traffic these tests are not always accurate. we allow high throughput bursts for large file downloads that would make your speeds look faster than they are.
Heres the bottom line. Speedtests and ISP speedtiers are relative and are almost useless without the proper context. We have a 50mb/s speed tier available now but guess what? That file you downloaded from yourfavoritesite.com yesterday at 300k/bs with your 6mb/s speedtier would download today at a whopping 300kb/s with a 50mb/s speed tier! why? because no matter how fast you are able to receive data, your only going to get it as fast as the site you are connecting to is willing to give it to you which is highly regulated. paying for a higher throughput speedtier is only good if you have mutiple, simultaneous connections or a designated point to point in network connection.[/quote]
+1 its called bandwidth throttling
bandwidth throttling
yes, but keep in mind that your provider does not do it(at leat not anymore!). we actually got spanked by the FCC for “throttling” P2P traffic among other things. that is the exact reason for the bandwith cap that was implemented a couple of years ago if anyone remembers. instead of restricting specific types of traffic we have limited everyone to 250GB of throughput per month.
theoretically, the only time your throughput would be less than what you pay for is during peak times when the amount of traffic is beyond the capacity of the switches and routers. however, when this happens your likely not to notice unless your down/uploading content deemed to be unimportant. quality of service(QoS) measures “shape” traffic flows to make sure that your IPvoice packet along with other critical data packets are given priority over p2p and other less important types of traffic.
this is really the only time that you might not get the full capacity of what you are paying for but is something that is certainly not unique to comcast. it is something that is built into every single network and is referred to as oversubscription.
how come you have to put in your address? HMMMMMMMMM
try this one to see how much they differ
Im not sure what you mean, but the only reason you would put in your address is if the service locates a server automatically for you. Most sites simply allow you to manually select your preferred location. I will give an example of what i am talking about.
Here is the route that a group of packets took that i sent to this site www.delreefclub.org. The number on the left is the hop number(router or mulitlayer switch), the next three numbers are the time in ms that a packet took to make the roundtrip between my pc and the last two items is the name and ip address of the router(or sometimes gateway). Every single router with the exception of the last three are in-network. This means that your speed tier is guarenteed all the way up to hop 13. The second it leaves our network there is nothing we can guarentee. The routers with an “ibone” prefix in the name means VERY, VERY fast wirespeed routing. This is our proprietary backbone that allows us to implement QoS from end to end(hop 1-13 in this case). Sometimes a packet might jump on a different network the moment it leaves our network. There is simply no way of telling. The path is NEVER fixed and describing factors that determine which path a packet takes would have me breaking my own record for longest post ever! ;D
Ignore the high round trip times and timeouts. Its related to my wireless connection and is not accurate. Even with the distorted number you can see that the longest trip on average is along the link between georgia and texas which makes sense.
C:\Users\Logans_Daddy>tracert delreefclub.org
Tracing route to delreefclub.org [174.132.114.66]
over a maximum of 30 hops:
1 2 ms 6 ms 1 ms TEW-639GR [192.168.10.1]
2 * * * Request timed out.
3 8 ms 8 ms 9 ms 68.85.152.45
4 12 ms 17 ms 11 ms te-1-1-ur01.frenchtown.de.panjde.comcast.net [68
.86.209.114]
5 9 ms 11 ms 8 ms te-1-1-ur01.tyboutscrnr.de.panjde.comcast.net [6
8.86.209.118]
6 9 ms 138 ms 9 ms 68.85.35.42
7 21 ms 18 ms 8 ms te-0-1-0-7-ar01.newcastle.de.panjde.comcast.net
[68.86.153.246]
8 38 ms 15 ms 18 ms pos-0-4-0-0-ar01.401nbroadst.pa.panjde.comcast.n
et [68.86.210.209]
9 14 ms 24 ms 20 ms pos-1-4-0-0-cr01.mclean.va.ibone.comcast.net [68
.86.91.245]
10 33 ms 31 ms 30 ms pos-1-15-0-0-cr01.atlanta.ga.ibone.comcast.net [
68.86.85.69]
11 54 ms 52 ms 51 ms pos-1-14-0-0-cr01.dallas.tx.ibone.comcast.net [6
8.86.85.153]
12 50 ms 57 ms 51 ms pos-0-2-0-0-pe01.1950stemmons.tx.ibone.comcast.n
et [68.86.86.150]
13 51 ms 90 ms 53 ms theplanet-cr01.dallas.tx.ibone.comcast.net [75.1
49.228.2]
14 66 ms 61 ms 52 ms te7-1.dsr02.dllstx3.theplanet.com [70.87.253.18]
15 * * * Request timed out.
16 69 ms 135 ms 85 ms po2.car01.dllstx6.theplanet.com [70.87.254.142]
17 53 ms 71 ms 50 ms gator636.hostgator.com [174.132.114.66]
Trace complete.
Great post, Shawn. My bad, when I first opened the link, I didn’t expand the page and didn’t see that you could just click on the test button, only saw the addy part. verdict_in :ZZZ)
To be honest i didnt even look at the link. I just figured it wasnt allowing you to manually choose your server.
Public speed test servers were the bane of my existence for several years when i worked exclusively with business class customers. No matter how hard i would try to explain this concept to a suscriber the wouldnt believe it could be anything other than a Comcast problem. There are just so many myths and rumors about how our technology works that no one wants to believe the truth of the matter. My favorite of all time was when Verizon was putting in all of their fiber for FIOS. During this time one of my main responsibilities was to “babysit” their contractors to ensure that they wouldnt damage criticial components of our infrastructure while installing their own. At least once a day i would have someone come up to my truck with a goofy grin on their face asking “so, when is comcast going to get around to installing fiber?”. The first dozen or so times i would politely explain that we have had a fiber network for over 10 years in New Castle to their disbelief. The best part was always when they would imply that i was being dishonest! After putting up with this for a couple of weeks i changed my tactics a little! I would answer their question with a question and ask them if they were a Comcast customer. If they replied yes, i would politely explain the same answer because we care about our customers(honestly, we do!!). If they replied no i would just roll up my window and turn my radio back up! ;D
Oly Chit Batman! Lots of tech talk way over my head. But that’s good. That gubmint site is just surveying by location for overall geographic performance through the whole network. for compliance regs enforcement or improvement no doubt. But im quite sure there are lots of Al-Gore-Rhythms employed by all parties involved, to best serve the question posed and the answer returned throught the internets. So there will be lots of variability based on size of file, and yes, the upload rate of the website we are tapping.
But if we test it a bunch of times, it gives us a range of instantaneous performances. not bad to know. Internet flow performance seems to be variable by design, in bursts. Not like my favorite bartender who fills my glass smooth and steady, first time, every time, so as not to make my beer foam up. I hate it when that happens.
I guess that means Bartenders are better Servers than ISP’s?