- last edited by
RogersMoin
Have been experiencing some latency issues, primarily while gaming, but I have noticed some odd hitches in other general activities, though just as intermittent. The plan at the house was upgraded to 1.5Gbps but they left the XB6 modem from the original plan. Given that was leaving some performance on the table, and that the modem was pretty old, it's since been upgraded to an XB8 but there is still this lingering problem.
I've tried changing a lot of settings on the PCs here, but there is an intermittent problem with latency that is quite noticeable in games.
I've gone through a bunch of older forum posts about certain things to check and run - the modem stats seem okay, bufferbloat is hit and miss with some of these tests and times of day, but something stood out as I went through running some traceroutes - the ping to the pool was wildly different at times.
So I let a ping run to it for a bit and you can see it's not really consistent.
So, I have to wonder what is the next course of action? Is any part of that first hop a result of something on my end, or is it something in the neighborhood?
*Added Labels*
@divito Unfortunately, your ping stats are pretty normal for cable Internet. It is not at all unusual to have occasional latency spikes exceeding 100 ms, and the annoying part for gamers is that the network performance is not consistent, especially when the DOCSIS network in your area is under load.
When you ping a well-connected Internet site, such as 8.8.8.8, you will typically see RTTs in the 20-35 ms range.
Perform a traceroute to 8.8.8.8 and you will also see response times in 20-35 ms range. Since the traceroute response times are cumulative for all the hops up to that point, that's confirmation that pretty much ALL of the latency that you see comes from clearing the DOCSIS infrastructure and the first-hop router beyond your gateway.
Perform that same traceroute multiple times, and you eventually see that the response time for the first hop beyond your gateway will be 100 ms (or more) but probes to subsequent hops are back to being in the 20-something ms range... confirming that all the latency spikes are being caused by Rogers' cable Internet infrastructure.
