What's new

R7800 bufferbloat test

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

ducdeptrai1234

Occasional Visitor
Lately I have more devices in my place using VoiP. I want to check if my R7800 has any issue with bufferbloat as some reports phone call issues. I came across DSLReport, I have used it before but I could not use its bufferbloat. The bufferbloat area just disappeared without any error notice. Anyone got the same issue?

screencapture-dslreports-speedtest-2021-03-07-11_14_31.png
 
All i can see is upload is limited or ISP plan itself. How many such devices are using voip and is QOS on?
 
1615265585952.png

I dont see such issues
 
I already did that with the website. However, I realized I do speed test far from the router, I do it via wireless extender (tri-band EX7500.) No matter how I tried, I got B at best. So let it be. Btw, my internet speed is ~220 Mbps and I used yeah congestion control.
 
I already did that with the website. However, I realized I do speed test far from the router, I do it via wireless extender (tri-band EX7500.) No matter how I tried, I got B at best. So let it be. Btw, my internet speed is ~220 Mbps and I used yeah congestion control.
That’s the thing with speed and buffer tests: it measures the connection between the server and the device from which you are doing the test. So it involves all intermediates (router, extender...) but the end device as well (phone, tablet, laptop...) so the buffer bloat result is for all the weakest device (or accumulation).

My results are B also, but from WiFi iPad.

However, if I do the Quick Test described here : https://www.bufferbloat.net/projects/bloat/wiki/Tests_for_Bufferbloat/
Pinging from the router, and doing the speedtest from a device connected through Ethernet, I get full speed speedtest (950 Mbps down, 940 Mbps up), and the ping goes to 7ms to 20ms during the test.
 
So it involves all intermediates (router, extender...) but the end device as well (phone, tablet, laptop...) so the buffer bloat result is for all the weakest device (or accumulation).
Also any intermediate servers. Modems, your ISP server etc. Buffer bloat test depends on concrete your environment. Not the only router.


Voxel.
 
Pinging from the router, and doing the speedtest from a device connected through Ethernet, I get full speed speedtest (950 Mbps down, 940 Mbps up), and the ping goes to 7ms to 20ms during the test.

Latency always increases when you load a circuit. You need to check your hardware stats to see if it is on your end. Latency really grows when you max a circuit as you are depending on buffers.
 
You might want to try different queue disciplines too.

For example, I get better buffer bloat scores using Stochastic Fairness Queueing (sfq) versus Fair/Flow Queueing + Codel (fq_codel). These are options I have on my UTM (Untangle).

In addition I have 300 Mbps down and 30 Mpbs up; however, via QoS in my UTM I have tweaked the WAN bandwidth max download to 285 and upload to 28.25. You typically set these to 85%-95% of your actual line speed.

Its a game. Different times of day may yield different results. I typically score an A using the sites mentioned here however.
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top