You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed that the speedtest client only does a single ping measurement before the download and upload tests, but fails to measure latency under load while the transfers are running. This is a mistake: measuring latency under load is a really simple way to get an indication of whether the user suffers from bufferbloat, one of the most common problems affecting perceived user networking performance.
To see examples of how other speedtests incorporate bufferbloat measurements, see:
This really ought to be part of the test specification, but I couldn't find a way to open a bug against that, so opening this instead.
If you need more information about how best to incorporate bufferbloat tests into your speedtest, we are multiple people willing to help on the bufferbloat mailing list: https://lists.bufferbloat.net/listinfo/bloat
The text was updated successfully, but these errors were encountered:
thank you for your input!
On a first glance, this looks doable without server modifications. So the best place to start for a proof of concept is probably the web client rtr-nettest/rmbtws. I'll move the issue there.
cproof
transferred this issue from rtr-nettest/open-rmbt
Sep 4, 2019
Hi
I noticed that the speedtest client only does a single ping measurement before the download and upload tests, but fails to measure latency under load while the transfers are running. This is a mistake: measuring latency under load is a really simple way to get an indication of whether the user suffers from bufferbloat, one of the most common problems affecting perceived user networking performance.
To see examples of how other speedtests incorporate bufferbloat measurements, see:
For more information about bufferbloat see:
This really ought to be part of the test specification, but I couldn't find a way to open a bug against that, so opening this instead.
If you need more information about how best to incorporate bufferbloat tests into your speedtest, we are multiple people willing to help on the bufferbloat mailing list: https://lists.bufferbloat.net/listinfo/bloat
The text was updated successfully, but these errors were encountered: