That is it ...

Well done .

That patch from 7:10pm to 8:10pm is appalling, with the rest of the evening unhealthy, but it's time for bed for me so not going any deeper.

Though there don't seem to be any re-syncs and possibly no actual loss of PPP connection. With luck others will comment in the morning long before I'm around.

In reply to a post by pluralist:

In reply to a post by Leaper1981:

*BQM

I'm not sure what that post is for, but it doesn't do anything . It's just text preceded by the asterisk.

I understood it was to correct "BMQ" in the previous post.

As an aside to Leaper1981, there is an "Edit" button next to your previous posts so you can correct them, at least for a limited time after posting.

Guess what!

I hadn't noticed the typo.

Just like you didn't notice:

In reply to a post by Leaper1981:

BTW the *BQM was because I had written BMQ in the previous post

Got there in the end 😂

The reason I was checking was because I had got access to a remote gaming service & for the first 4 days it ran fine then after that I kept getting kicked out of the remote computer so although I knew remote gaming can have issues from their end it had ran fine for almost a week so I didn't think it was them. The point where the red bits are coming down was while using the online gaming service. I feel like this might have been occurring for sometime but was never really an issue as before using this service all I was really doing was general Internet browsing so if it cut out and came back on quickly it didn't really cause me a problem. I felt like that graph was saying something wasn't right but just wanted to check.

Actual loss of sync or shows as the red going all the way to the bottom and staying solidly there for several "ticks". I've not seen that word used before on these forums, but I'm using it to define the frequency of each entry along the x-axis. Each entry represents 100 pings. Averaging at 100 seconds.

Once you have registered an IP address for monitoring, we will ping that IP address sending small ICMP echo requests. We send a ping each and every second and average 100 seconds of pings for each point displayed on the graph.

A loss of just PPP can show as a spike all the way to the bottom, though I'm not sure it always does.

Edited by pluralist (Sun 09-Jan-22 18:06:05)

Bob remember as VM02 use DOCSIS on their cable networks there is no concept of sync. or PPPoE as there is on Openreach. DOCIS networks have delay characteristics that aren't necessarily directly equivalent to DSL.

Hehe !

My ignorance of DOCSIS beyond seeing it achieve speeds over copper that Openreach found impossible is so great that I never realised what you post. I have nothing to forget .

Seriously, a very good point of which I was totally ignorant. What effect it has on the interpretation of a BQM I haven't yet thought about. Can you help the OP?

That graph is absolutely atrocious, the red spikes from the top are packet loss and when streaming your game you're looking at 30% packet loss and upwards. I'm actually surprised it worked at all and just gave you some intermittent disconnects.

One thing is highly likely on a VM network, and that's oversubscription, or in their words 'high utilisation'. It's also impossible for you to fix, it requires VM to split up your area ('segment it') so that you're sharing the bandwidth with fewer users, or implement DOCSIS 3.1 which in practice means you must be issued a Hub 4 rather than a Hub 3.

The VM forums have some helpful users who can talk you through checking the signal levels on your router in case that's the problem. The bad news is it's less likely, the good news is that's an easier fix that most general VM technicians can resolve quickly.

I suspect that the signal levels won't be an issue. The BQM looks generally OK before the period of gaming, and deteriorates heavily during the period the OP says he is gaming, roughly in the period from 19:00 to 20:30.

There are some further 4 or 5 nasty spikes in the period following roughly every half hour up to midnight. We don't know if further game play or heavy internet usage was happening then or something else?

To my untrained eye it looks like a very severe case of bufferbloat - perhaps in combination with a CMTS segment that is already quite heavily subscribed. It would be useful to understand what the sustained and peak data transfer rate (down / up) is during gaming. If the OP is completely soaking up all the available bandwidth, either down (or more likely up) on the connection then that will ultimately result in a BQM like that.

I would run some further bufferbloat tests to see how bad the connection is at other times of the day. OP could try:

https://www.waveform.com/tools/bufferbloat