@herrshaun wrote:
It's important that Rogers knows it's not just tied to something like a Call of Duty release, because if they can find a reason to pass the buck, they'll do it...
Yeah, I agree totally. It's one thing for a TPIA provider to be impacted if they have under-provisioned at the POI. For Rogers, this should be a non-issue.
Unless of course Rogers has carved off several hundred Gb/s from the core routers and dedicated that capacity for something else, like maybe 5G deployment? "Look at how fast our 5G is!!!" All speculation on my part of course....
It doesn't make sense that the latency and packet loss problem has suddenly arrived out of the blue. As IPTV deployment ramps up for example, I would think you would see a gradual decrease in performance if Rogers didn't expand its data capacity, or didn't preplan for the required capacity. I don't believe it would cause a sudden drop. I doubt that the installers are doing enough installs every day to cause what appears to be a sudden network performance drop. Something else has been going on over the last two months that Rogers isn't revealing (take your pick), and whatever that is, it appears to be hitting the GTA with greater impact.
@Datalink wrote:
It doesn't make sense that the latency and packet loss problem has suddenly arrived out of the blue. As IPTV deployment ramps up for example, I would think you would see a gradual decrease in performance if Rogers didn't expand its data capacity, or didn't preplan for the required capacity. I don't believe it would cause a sudden drop. I doubt that the installers are doing enough installs every day to cause what appears to be a sudden network performance drop. Something else has been going on over the last two months that Rogers isn't revealing (take your pick), and whatever that is, it appears to be hitting the GTA with greater impact.
Yeah, this is a weird issue that has been impacting everybody who is on Rogers' infrastructure, that (my feeling as well) occurred out of the blue because somebody did something not realizing that something profoundly bad would happen. (Will retest this evening to see if the problem in my corner of the world has been fixed.)
Yesterday, the TPIA providers potentially got hit twice: due to local issues at the CMTS and bandwidth constraints at the POI with the traffic spike from game downloads.
Just also agreeing that traffic spikes should not have brought any part of the Rogers infrastructure to its knees.
@m4te here are some items to consider:
1. Bell appears to route a good majority of its traffic thru Chicago for some reason. End result, the Rogers and Bell routing are probably completely different.
2. Even though you and your friend are situated close together, you might be connected to a completely different neighbourhood node and CMTS. The neighbourhood nodes don't show up in a trace. If the modem is in Gateway mode with a direct connection, or, the modem is in Bridge mode with a router behind it, and with a direct connection to the router, the second hop in the trace is the CMTS. If you and your friend were to compare a trace to the same point, have a look at the IP address in hop #2 to see if its the same address. If so, you're both on the same CMTS. I would expect the routing to be the same to the same endpoint. If not, ok, then you're on a different CMTS, in which case the routing will be slightly different, but, I'd expect to see the two routes converge after one or two hops.
3. Probably the only staff who could tell you if you were both on the same neighbourhood node would be tech support or perhaps @RogersAndy or @RogersMoin or @RogersTony. You would have to send one of them the MAC address of the two modems in order to track down the connected neighbourhood node and the connected port on the node. Even if you were on the same node, you could be on different ports, which is highly probable if the neighbourhood node is nearby. Each node port would support some number of connected modems (500, 1000, or 1500 ??) and have its own noise issues and cable issues to contend with. All of those factors combined could result in different modem performance observed by yourself and your friend. I think the biggest factor would be the total number of modems connected to each port. If your node port had a significantly higher number of supported modems, then you would end up with less transmit time slots and higher latency. This is down into the weeds level, but, knowing how many modems are connected to each port would provide some idea of why there might be a difference in latency. Beyond that, differences in noise levels experienced by both modems could have an impact on the data rates. If you were to compare both modem's signal levels, QAM levels and signal to noise , that might provide another part of the puzzle.
When you compare your gaming performance to your friend on Bell, there are a couple of points that might work in his favour:
1. VDSL can definitely give you lower latency, as see in the following chart:
https://www.dslreports.com/forum/r32680851-
2. Bell's routing to the servers in question might just be in his favour.
The bigger question, I think, when you're trying to compare numbers and understand the differences, is the routing and peering issue. I've said this before, and I'll say it again (hello moderators), the gaming community here in the forum could really make use of an Ask the Expert session with someone who is knowledgeable in Rogers routing beyond the borders, and knowledgeable with Rogers peering partners. The additional aspect of this is the throughput at the exchanges that Rogers is connected to. For example, for Torix, the Toronto Internet Exchange, the exchange web site indicates that Rogers has a 40 Gb/s link, while Teksavvy has a 100 Gb/s link. If you look at the following chart, you can see the throughput rates at Torix. Keep in mind that it doesn't provide the whole story of total ISP peering throughput rates. This is just for Torix connected companies, at Torix:
Here's the data from the peering database. Note that this is the pubic data, so there is more that can't be seen"
https://www.peeringdb.com/net/307
So, these are a couple of examples, and maybe Rogers exchange links elsewhere provide enough throughput, or maybe not. At the end of the day, the customers don't care, unless their throttled by a lack of throughput due to the choices that Rogers has made in its routing and peering and throughput data rates. This appears to be a closely guarded secret squirrel issue, and as a result, its a total black hole for the gamers who use Rogers. Fwiw, TPIAs appear to be willing to discuss issues like this, or at the very least attempt to sort out issues due to routing, larger incumbents appear to do nothing. That's from observations of posts from users with various ISPs expressing frustrations with their routing.
Hope this helps.
This is just brutal. Roger needs to get their act together.
Hello @LimboMan,
I definitely get your frustration as everyone has their breaking point. In all honesty, we'd be pretty devastated to lose your valued business after 25+ years!! 😟
If at all possible, we'd love for you to send us a Private Message @CommunityHelps so we can do a deeper dive into your issue. We can run some tests at our end but may also need to collect some data from you as well. After the necessary tests are run we'll be able to determine if this is an isolated issue or if it's related to the latency issues impacting some of our users who have posted here.
For more information on how our Private Messaging system works, please CLICK HERE.
Kindest of regards,
RogersLaura
-
207227 views
-
156 Likes
-
-
