Hi everyone,
I'm struggling with inconsistent gameplay in Counter-Strike 2 (feeling "delayed" and experiencing micro-stuttering) despite having a 1Gbps Fiber connection. I've performed extensive diagnostics and I would appreciate a professional look at my data.
My Setup:
Router: GL.iNet GL-MT6000 (Flint 2) running OpenWrt.
Optimization: SQM (Cake) is enabled and tuned. Bufferbloat test results in an A+ grade (+1ms delay under load).
Connection: 100% Ethernet (no Wi-Fi involved in testing).
The Problem:
My local network is rock solid (Hop 1 stays consistently at 0.4ms - 0.6ms with 0.1ms Jitter). However, my ISP's border router (bgp.161.kos.beskidmedia.pl) shows frequent and massive latency spikes.
Key observations from my logs:
While the average ping to the target (8.8.8.8 or Valve servers) is around 7-9ms, the Worst (Max) value at the ISP's BGP hop frequently hits 120ms - 240ms.
These spikes occur even when my local router shows no latency increase at all.
My ISP claims this is just "ICMP Rate Limiting/CPU Protection," but I feel these spikes as desync/jitter during real-time applications (CS2 and Live Streaming).

Is this level of jitter (Avg 9ms vs Max 240ms) on a border router normal for Fiber? Could this be a peering issue or a congested port at the ISP?
Thanks for any insights!

Hey sanders,

Thanks for reaching out on the PingPlotter forums and sending your PP2 file!

From what I can see inside the data collected by PingPlotter, there doesn't seem to be any sign of an issue. That isn't to say that there are no issues, but rather that the data collected so far doesn't show us what the issue is.

You mentioned the data going to an intermediate hop, your ISPs border router (bgp.161.kos.beskidmedia.pl). Very often, intermediate hops can show numbers that imply your network connection is far worse than it actually is. This isn't a bug, but rather a feature of how these routers handle data. Many devices will down-prioritize certain traffic (such as timed-out ICMP packets) in order to make sure that they can forward data where it needs to go. You can read more about the link below:

https://www.pingman.com/kb/article/one-poorly-responding-router-5.html

As for actually checking if this connection is bad, you may consider opening a target directly to bgp.161.kos.beskidmedia.pl. Since it would be the final hop in its own trace, the data that you get would give you much better insight as to how your connection to the router is.

You may also want to confirm that the Valve IP-address you're tracing to is the one that you're actually connecting to in-game.

When dealing with ISPs, some useful articles to check out may be:

https://www.pingplotter.com/fix-your-network/build-a-case/
https://www.pingplotter.com/manual/building_a_compelling_case/

I hope that this clears things up for you. If this leaves you with any questions, please feel free to follow back up, or reach out to us at support@pingman.com!

Thanks,
Zackary

Hi Zackary,

Thank you for taking the time to review my PingPlotter data and for your detailed explanation.

I would like to add some context to my setup. My internet connection is running in bridge mode because I also have IPTV services from my ISP. The tests were performed under idle conditions — there was no network activity at the time (no one was using Wi-Fi or watching TV).

I understand your explanation regarding intermediate hops and ICMP rate limiting. I will also verify the exact Valve server IP I’m connecting to in-game and consider testing directly to the ISP edge router as you suggested.

Thanks again for your help and guidance.

Best regards,
Przemysław