askywalker
join:2002-06-25
Markham, ON
| Rogers speedcheck is the only check that matters
Since my switch to the 74. series IP I have had really poor download speeds. Two locations both on Extreme - both have had consistent 6Mb speeds until the change. The speedcheck.rogers.com shows 6Mb, but toast, dslreports, giganews all show VERY slow dl (but fast upload). Opened trouble tickets at both sites. One of my sites is now working perfectly again. Other site is still poor, but Rogers closed the ticket (without telling me) - with a quote:
"Speedcheck is the only speed test that matters. It shows fine speeds - so you don't have a problem." I desperately tried to convince him that EVERYTHING on the net is slow, but wouldn't budge.
Does anyone have any experience with this? I have been happy with the Rogers service so far. But DSLReports is showing my speed as around 600 down/700 up. I have connected directly to the modem, and used 4 different computers. Giganews and TOAST.CA show consistently the same results.
Rogers told me they are delivering 5.5 - cause nothing matters except for speedcheck. If I don't like it, they suggested I take my business elsewhere. Any suggestions? I can't get Bell high speed DSL... |
|
BayStBoi
@rogers.com
| The slow down is likely related to the way the 74 block of IP addresses is being routed once it leaves the Rogers network. Rogers would have no control over this since they don't own / operate these routers, so not much that can be done it would seem. |
|
iceweasel
@rogers.com | They could give me back my old IP address,this price increase for crappier service sucks  |
|
HiVolt
30
Premium
join:2000-12-28
Toronto, ON
clubs:
 ·TekSavvy Solutions..
·Bell Sympatico
| reply to askywalker
I wouldn't be surprised if its not some mis-configured overly aggressive packet shaping. The Rogers speedcheck is fine because its likely bypassed by the packet shaping.
--
·.¸¸.·´¯`·.¸¸.·´¯`·.¸¸.·´¯`·.¸¸.·´¯`·.¸¸.·´¯`·.¸¸.·´¯`·.¸¸.·´ |
|
x0marx
join:2006-04-15
Guelph, ON | reply to askywalker
Mines just the opposite, my upload speed went down and download stayed about the same since I got the 74.* IP. |
|
Sanek
join:2006-08-10
Kanata, ON
| reply to askywalker
I have a 70.* IP and Rogers speed check shows 5.94 Mbit / 812 Kbit, while all other speed checks - even Canadian ones including VideoTron show ~4.7 Mbit / 790 Kbit. VideoTron does give me better results than the rest, but not by much.
Is this the problem with the network or is Rogers causing this?
I'm on 6 Mbit / 800 Kbit Rogers Extreme service. |
|
sbrook
Premium,Mod
join:2001-12-14
H0H 0H0
 ·Rogers Hi-Speed
Host:
Rogers
Bell Canada
| This is quite common. Remember that as soon as packets leave Rogers internal network, they are subject to whatever routing that the backbone routers chose to give them with whatever delays may be enroute. They can end up going throug a lot of overloaded routers.
So, results that show slow speeds from outside Rogers are not uncommon.
One good one to try is speedtest.cogeco.net This will verify that at least one of the pipes used to get out of Rogers network is OK. (It's usually within a couple % of the Rogers speedtest results)
The reason that Rogers speedtest is the only one that matters is for that very reason. There's only one way that it's still a Rogers issue and that's if the pipes that it leaves Rogers network on are overloaded or misrouting. At one time, we saw a lot of that when they used Teleglobe, but now it's much better, considering they have their own pipes now to the Washington area, the New York area, and I believe also to Chicago. |
|
askywalker
join:2002-06-25
Markham, ON
| sbrook - I tried »speedcheck.cogeco.net/ but no page there. Do you have another link.
I have found that I cant get over 1 MB dl from any of the test sites - except the Rogers speedcheck. Since I don't dl anything from Rogers, I don't care that I have a 5MB connection to Rogers.
I did a test with giganews and found that I could keep adding connections to get to a total of 5.5 MB - it is as if they are throttling each of my connections to the Internet. I downloaded 5 GBs from giganews very reliably at 5.5 MB - but needed to have 6 connections open to get to that speed. I would suspect it could be a PC or router problem, so I tested with my notebook directly connected to the modem with the same result. |
|
sbrook
Premium,Mod
join:2001-12-14
H0H 0H0 | Ya gotta READ it right - it's speedtest not speedcheck ! 
here's a clickable ... »speedtest.cogeco.net |
|
askywalker
join:2002-06-25
Markham, ON
| i need that head slapping emoticon now... sorry bout that! Thanks for the link.
I just ran that test and it gave me 5341/723. Isn't it strange that I can tunnel to my other Rogers connection (in the same city but different location) and get these same speeds from the DSLReports tests, Toast and Giganews? |
|
sbrook
Premium,Mod
join:2001-12-14
H0H 0H0 | reply to askywalker
BUT what is your IP address at the two locations? 74.x.x.x seem to have slower routings than other Rogers IP ranges.
Are you connected via a router at either? (esp. wireless) |
|
askywalker
join:2002-06-25
Markham, ON
| Both locations got switched to the 74.117 series the same weekend - second and third octet are different. One on Friday, the other on the Sunday. Both immediately had their dl rates killed - though there was no packet loss and the upload speed stayed at the standard Extreme rate. Opened tickets on both locations - one got fixed - the other didn't, and as I said Rogers has closed the ticket and 'hung up on me' - they refuse to talk about it any further as 'the speedcheck.rogers.com is the only test that matters'.
While on the original support calls I disconnected the router and connected a notebook directly to the modem - exact same results. It is a wired network.
Funny thing is - the connection is VERY reliable - just really slow. The only time it is very obvious is doing some daily work downloads that are large - suddenly they take an hour instead of several minutes - and I have a giganews account that now gets around 400k instead of the previous 5.5MB. My bet is many of my neighbours would have the same issue if they looked into it, they just don't really notice.
I spent four+ hours with support. They didn't notify me that they closed the ticket, and it took several calls back before they could find my ticket (their ticket system was down etc..). Then they finally call me back only to tell me they won't talk to me anymore - that I don't have a problem. Fun stuff. |
|
Dustyn
Premium
join:2003-02-26
Ontario, CAN
|  reply to HiVolt
said by HiVolt  :I wouldn't be surprised if its not some mis-configured overly aggressive packet shaping. The Rogers speedcheck is fine because its likely bypassed by the packet shaping. I get very comparable results accessing the Net Access speed test servers which hosts DSLR. Packet shaping has little to no effect on results.
--
»/forum/wall |
|
SW_anon
@rogers.com
| reply to askywalker
Don't bother... the Rogers speedcheck doesn't even matter to Rogers themselves. I got steady 3Mbps instead of 5Mbps (Express) on their very own speedcheck. All they said was that it's only up to 5Mbps and we aren't guaranteed any speed. They said there was nothing wrong. |
|
sbrook
Premium,Mod
join:2001-12-14
H0H 0H0
·Rogers Hi-Speed
Host:
Rogers
Bell Canada
| reply to askywalker
Call Rogers and ask to have a customer service manager call you back. They'll try to "solve the problem" again, but be insistent that you want a call back.
Explain that support has NOT fixed the problem on one line when they did fix the other and fobbed you off with a non-answer, closed tickets that weren't fixed and now refuse to speak with you.
Whether the manager will get someone to actually work on the problem is another matter, but Rogers does need to get the message that support is crap. |
|
Fredra
Undesirable Alien
join:2000-04-08
Nepean, ON
·Rogers Hi-Speed
1 edit | reply to askywalker
I have tried Rogers Speedcheck and IMHO it is neutured to some degree.
Try this Visualware speed check
»myspeed.visualware.com/?email=ma···85004C57
It may look the same as rogers, but it gives more information with the results, which you can interpret anyway you like.
They also have a d/l version.
I am still on 24.xx.xx.xx,so I am unable to make comments on the 74.xx.xx.xx.
However, try the above check and see what you get as opposed to the Rogers diluted version.
Hope that hepls.
Cheers
--
The Endless |
|
sbrook
Premium,Mod
join:2001-12-14
H0H 0H0
·Rogers Hi-Speed
Host:
Rogers
Bell Canada
1 edit | The need for a speedtest is normally to validate that your connection to Rogers network is at the advertised rates.
As such, it is important that the server you're connecting to has few network delays to ensure that you get an accurate result. Therefore, it is best to use Rogers speedcheck. The so-called neutering doesn't devalue the accuracy of the speedtest.
The visualware speedtest, whether a downloaded or web version, like other most other speedtests is connecting to a site somewhere across the internet and is subject to network congestion and routing issues outside Rogers control and slow speeds or long roundtrip times to these servers mean nothing in terms of problems with your own connection. All those additional items provided by the visualware site speedtest only serve to show that the internet is full of quirks, and not that your Rogers connection is faulty when you get a bad result.
I have tried the visualware site on two occasions now ...
Once a few weeks ago, when I got an appalling result (surfing to most places was fine and Rogers speedtest was fine) which meant that the problem was out on the net and not Rogers problem (could be visualware's server was overloaded too).
Then today, apart from a 78mS roundtrip time, the speeds were the same as Rogers, which meant that I had a relatively good connection and the visualware server was not overloaded.
Using these diagnostic tools without a full understanding of the significance of the numbers and the nature of the network can lead you on wild goose chases. They must be backed with the use of other tools such as tracerts etc. to begin to make an accurate determination of if and where there are problems. |
|
askywalker
join:2002-06-25
Markham, ON
| Thanks for all the ideas and thoughts. I ran some tests
Results: (dl/ul)
visualware: 459/740
DSLReports (NJ): 497/717
Giganews: 637
speedcheck (Rogers): 4790/742
speedtest (Cogeco): 5382/723
My other location: (also Rogers Extreme)
visualware: 3999/762
DSLReports (NJ): 2014/732
Giganews: 3871
speedcheck (Rogers): 3120/763
speedtest (Cogeco): 4425/725
sbrook - I completely understand what you are saying - in fact I am a VoIP guy, have done some Cisco certification and have a fairly decent grasp of networking and the Internet. I will work on getting further help from Rogers - their support is really hit and miss. There is most definitely a problem - it is a tricky one though - as my connection to Rogers is good - just seems something is out of wack with the node I am on, or some traffic shaping or something that is working against me. I still believe they are implementing upgrades and having some troubles (74.*), and that my problem will end up getting sorted out eventually - just that I need the connection soon. My work will probably end up having to get my a wireless connection, as we almost did this a couple of years back before Rogers starting bumping up their bandwidth. |
|
sbrook
Premium,Mod
join:2001-12-14
H0H 0H0 | Try posting the results of a tracert to www.rogers.com and www.visualware.com |
|
askywalker
join:2002-06-25
Markham, ON
| Good news - convinced Rogers to send out a new modem, and it has solved the problem.
New results first - results from this morning in brackets
visualware: 5940/794 (459/740)
DSLReports (NJ): 3388/771 (497/717)
Giganews: 6011 (637)
speedcheck (Rogers): 5950/795 (4790/742)
speedtest (Cogeco): 3468/263 (5382/723)
Finally back to normal... Now the downloads I need to do from the office will be back to 15 minute waits instead of 2.5 hours! 15 minutes is still too damned slow - can't wait till 18 Mbps...
Not sure what it was that was the problem. Tech thought it was a faulty modem - I still think it had something to do with the contract to the modem sent from Rogers - but whatever - I am back to wasting my time with other things now. I do think others will have similar problems - as it happened during the 74.* switch to two locations that I manage. Hope others can learn from the massive time that I had to spend convincing Rogers that there was a problem.
Thanks for all the ideas and help! |
|
