Page 6 of 8
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 03:01
by rekaa02511
vnb wrote: ↑15 Nov 2025, 00:21
OKAY
Very important update! :
I found the culprit to my issue : upstream UDP jitter
Here is the proof :
upstream jitter capture.PNG
So even though my XGS-PON Fiber connection is 2Gbp/s UP and Down capable, this is only a 100mbp/s symmetrical test.
As you can see the result above the RX latency (downstream) stays under 1 ms but the TX latency (upstream) is never below 10ms and the value varies a lot which is not good. So waveform bufferbloat and most other "tools" to measure latency are not giving us the full pictures and can somehow overshadow the "raw results" of your ISP... Case closed. Now time to "fight" with my ISP wish me luck
Conclusion
: i found the culprit to the low in height yellow lines in the CS margin graph after months of research and testing, Valve network engineer and other ppl on twitter and reddit that consider themselves "expert" on the matter were not able to give me a conclusive answer.
well good news if you have the same issue as me (too much yellow lines low in height in cs margin graph) even tho you have no packet loss or bad network values, well the issue is
upstream UDP jitter as shown in my screenshot above.
whats the app to test it on?
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 03:04
by themagic
If you search for Jitter then it says that under 30ms is very normal and good and for VOIP same..
What is considered good jitter
Ideal: Under 30 ms is considered good and generally provides a smooth, responsive gaming experience.
Acceptable: Jitter between 30 and 50 ms can start to negatively impact gameplay.
Poor: Higher levels of jitter can lead to significant lag, disconnections, and poor quality of service.
Just increase and put some Jitter on your RX and so everything aligns perfect and then you have 10+ms Jitter down 10+ms up...together still under the 30ms mark.../Ironyoff
But wish good luck with your ISP...
Pls keep share and tell and if you achieve something in that direction...
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 03:26
by themagic
rekaa02511 wrote: ↑15 Nov 2025, 03:01
vnb wrote: ↑15 Nov 2025, 00:21
OKAY
Very important update! :
I found the culprit to my issue : upstream UDP jitter
Here is the proof :
upstream jitter capture.PNG
So even though my XGS-PON Fiber connection is 2Gbp/s UP and Down capable, this is only a 100mbp/s symmetrical test.
As you can see the result above the RX latency (downstream) stays under 1 ms but the TX latency (upstream) is never below 10ms and the value varies a lot which is not good. So waveform bufferbloat and most other "tools" to measure latency are not giving us the full pictures and can somehow overshadow the "raw results" of your ISP... Case closed. Now time to "fight" with my ISP wish me luck
Conclusion
: i found the culprit to the low in height yellow lines in the CS margin graph after months of research and testing, Valve network engineer and other ppl on twitter and reddit that consider themselves "expert" on the matter were not able to give me a conclusive answer.
well good news if you have the same issue as me (too much yellow lines low in height in cs margin graph) even tho you have no packet loss or bad network values, well the issue is
upstream UDP jitter as shown in my screenshot above.
whats the app to test it on?
http://startrinity.com/InternetQuality/ ... ester.aspx
https://www.gutefrage.net/frage/spielra ... itter-hoch
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 05:42
by dervu
You should also check if it happens on speeds that CS2 actually uses.
You can use iperf3 too, just have to find server that accepts UDP.
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 06:04
by vnb
dervu wrote: ↑15 Nov 2025, 05:42
You should also check if it happens on speeds that CS2 actually uses.
You can use iperf3 too, just have to find server that accepts UDP.
it happens regardless of the speed. i even tried with 1 mbp/s = same thing. And this is directly connected to my ISP infrastructure hence the 1ms RTT.
on iperf3 you have too many variables like which route / tier 1 isp you go through / transit ip, etc. And results vary too much on public servers due to how many ppl use them; you would need a lot of data over an extended period of time to draw a conclusion. Instead in my case i go directly my router --> ISP infrastructure with nothing in between. So if in this path which is the shortest possible i have already issues; imagine with a longer path like on iperf3... Definitively a ISP issue without a doubt.
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 07:58
by paddyhhh
vnb wrote: ↑15 Nov 2025, 00:21
OKAY
Very important update! :
I found the culprit to my issue : upstream UDP jitter
Here is the proof :
upstream jitter capture.PNG
So even though my XGS-PON Fiber connection is 2Gbp/s UP and Down capable, this is only a 100mbp/s symmetrical test.
As you can see the result above the RX latency (downstream) stays under 1 ms but the TX latency (upstream) is never below 10ms and the value varies a lot which is not good. So waveform bufferbloat and most other "tools" to measure latency are not giving us the full pictures and can somehow overshadow the "raw results" of your ISP... Case closed. Now time to "fight" with my ISP wish me luck
Conclusion
: i found the culprit to the low in height yellow lines in the CS margin graph after months of research and testing, Valve network engineer and other ppl on twitter and reddit that consider themselves "expert" on the matter were not able to give me a conclusive answer.
well good news if you have the same issue as me (too much yellow lines low in height in cs margin graph) even tho you have no packet loss or bad network values, well the issue is
upstream UDP jitter as shown in my screenshot above.
If I open the home screen of this software while running a test, there is neither instant ul or dl jitter.
When opening the measurements (raw) tab, rx instant jitter is fluctuating 0-30ms.
Idk if this test should be trusted
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 13:36
by vnb
paddyhhh wrote: ↑15 Nov 2025, 07:58
If I open the home screen of this software while running a test, there is neither instant ul or dl jitter.
When opening the measurements (raw) tab, rx instant jitter is fluctuating 0-30ms.
Idk if this test should be trusted
probably a you issue tbh. something must be wrong on your pc. I tried on two different high end PC and even a midrange laptop and results were the same. whatever non sense you say is not happening with me.
I was about to share the result of iperf which show similar numbers...
BUT Anyways, the "deniers" on this forum are everywhere like the plague. so i will avoid sharing more; because as soon as i share
(from good faith because i have nothing to gain, u know)
ppl like you come and say "ye but this ye but that" y'all are just fucking annoying as fuck and people will stop sharing their finding because there always be annoying mf that comes to deny.
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 14:01
by paddyhhh
vnb wrote: ↑15 Nov 2025, 13:36
paddyhhh wrote: ↑15 Nov 2025, 07:58
If I open the home screen of this software while running a test, there is neither instant ul or dl jitter.
When opening the measurements (raw) tab, rx instant jitter is fluctuating 0-30ms.
Idk if this test should be trusted
probably a you issue tbh. something must be wrong on your pc. I tried on two different high end PC and even a midrange laptop and results were the same. whatever non sense you say is not happening with me.
I was about to share the result of iperf which show similar numbers...
BUT Anyways, the "deniers" on this forum are everywhere like the plague. so i will avoid sharing more; because as soon as i share
(from good faith because i have nothing to gain, u know)
ppl like you come and say "ye but this ye but that" y'all are just fucking annoying as fuck and people will stop sharing their finding because there always be annoying mf that comes to deny.
People like me?
You should chill a bit, I pointed out a potential bug of this software, nothing else.
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 14:25
by vnb
paddyhhh wrote: ↑15 Nov 2025, 14:01
People like me?
You should chill a bit, I pointed out a potential bug of this software, nothing else.
yes people exactly like you.
Tell me what did you add to the topic? 0 data, nothing.
You just come to deny then dip out. After a quick glance at your profile it seems to be a usual behavior of yours.
There is no bug in the software lmao, go fix ur shit
Re: CS2 Margin Graph
Posted: 15 Nov 2025, 14:49
by dervu
Here are my results from iperf3 on one of French servers for UDP, yet it still happens on some servers. That only indicates in my case issue would lie somewhere between my ISP and specific servers.
Code: Select all
Run | Direction | Transfer [MBytes] | Bitrate [Mbit/s] | Jitter [ms] | Lost/Total | Loss [%] | Out-of-order
----+-----------+-------------------+------------------+-------------+-----------------+----------+-------------
1 | Sender | 179 | 100 | 0.000 | 0/130201 | 0 | N/A
1 | Receiver | 179 | 99.4 | 0.018 | 0/130201 | 0 | N/A
2 | Sender | 180 | 100 | 0.000 | 0/0 | 0 | N/A
2 | Receiver | 179 | 100 | 0.022 | 291/130627 | 0.22 | 9