Oh, and this is a useful tool to measure the phenomenon: https://www.waveform.com/tools/bufferbloat

This was how I could test that the sfq configuration made an improvement. I get an "A" grade now.

lm@lm-System-Product-Name:~$ iperf3 -c ams.speedtest.clouvider.net -R -p 5203
Connecting to host ams.speedtest.clouvider.net, port 5203
Reverse mode, remote host ams.speedtest.clouvider.net is sending
[ 5] local 192.168.0.111 port 34674 connected to 194.127.172.176 port 5203
[ ID] Interval Transfer Bitrate
[ 5] 0.00-1.00 sec 96.1 MBytes 806 Mbits/sec
[ 5] 1.00-2.00 sec 104 MBytes 876 Mbits/sec
[ 5] 2.00-3.00 sec 109 MBytes 914 Mbits/sec
[ 5] 3.00-4.00 sec 110 MBytes 926 Mbits/sec
[ 5] 4.00-5.00 sec 110 MBytes 927 Mbits/sec
[ 5] 5.00-6.00 sec 110 MBytes 926 Mbits/sec
[ 5] 6.00-7.00 sec 110 MBytes 927 Mbits/sec
[ 5] 7.00-8.00 sec 110 MBytes 926 Mbits/sec
[ 5] 8.00-9.00 sec 110 MBytes 927 Mbits/sec
[ 5] 9.00-10.00 sec 110 MBytes 926 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.04 sec 1.06 GBytes 907 Mbits/sec 242 sender
[ 5] 0.00-10.00 sec 1.06 GBytes 908 Mbits/sec receiver

iperf Done.

lm@lm-System-Product-Name:~$ ping -c 10 185.60.112.157
PING 185.60.112.157 (185.60.112.157) 56(84) bytes of data.
64 bytes from 185.60.112.157: icmp_seq=1 ttl=245 time=14.5 ms
64 bytes from 185.60.112.157: icmp_seq=2 ttl=245 time=14.6 ms
64 bytes from 185.60.112.157: icmp_seq=3 ttl=245 time=14.3 ms
64 bytes from 185.60.112.157: icmp_seq=4 ttl=245 time=14.3 ms
64 bytes from 185.60.112.157: icmp_seq=5 ttl=245 time=14.1 ms
64 bytes from 185.60.112.157: icmp_seq=6 ttl=245 time=15.0 ms
64 bytes from 185.60.112.157: icmp_seq=7 ttl=245 time=14.6 ms
64 bytes from 185.60.112.157: icmp_seq=8 ttl=245 time=14.0 ms
64 bytes from 185.60.112.157: icmp_seq=9 ttl=245 time=14.2 ms
64 bytes from 185.60.112.157: icmp_seq=10 ttl=245 time=14.0 ms

--- 185.60.112.157 ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9013ms
rtt min/avg/max/mdev = 14.043/14.356/14.959/0.295 ms
lm@lm-System-Product-Name:~$ ping -c 10 pingbox2.thinkbroadband.com
PING pingbox2.thinkbroadband.com (80.249.99.171) 56(84) bytes of data.
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=1 ttl=57 time=7.93 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=2 ttl=57 time=7.83 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=3 ttl=57 time=8.00 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=4 ttl=57 time=8.23 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=5 ttl=57 time=7.77 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=6 ttl=57 time=7.92 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=7 ttl=57 time=7.93 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=8 ttl=57 time=7.54 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=9 ttl=57 time=7.46 ms
64 bytes from ip99-171.thdo.ncuk.net (80.249.99.171): icmp_seq=10 ttl=57 time=7.33 ms

--- pingbox2.thinkbroadband.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9014ms
rtt min/avg/max/mdev = 7.326/7.791/8.232/0.262 ms
lm@lm-System-Product-Name:~$

Edited by adslmax (Thu 15-Jan-26 18:44:27)

Candlerb, Adslmax isn't going to get that in a tp-link, but wouldn't QoS do the a very good job instead..

In reply to a post by Taras:

Candlerb, Adslmax isn't going to get that in a tp-link, but wouldn't QoS do the a very good job instead..

I try enabled QoS and set download limit as 1000 and upload limit as 95

But still high spiked. But I don't know to be honest!

I am going to put G.fast back on and try both downstream and upstream latency to see if Cerberus network is low latency! If still on high latency, maybe to do with my PC or something else!

Will post back soon when I connected back to G.fast.

Edited by adslmax (Thu 15-Jan-26 19:54:11)

what dns servers are you using ?

In reply to a post by Taras:

what dns servers are you using ?

Ah good point! How to check this?

Here is G.fast - still same on high spiked on latency (upload) so, I ruled out FTTP, nothing to do with FTTP now not caused any high spiked. Must be somewhere!!!! Maybe DNS server!

https://www.speedtest.net/result/18720759795.png

In reply to a post by Taras:

what dns servers are you using ?

DNS will have nothing to do with incoming icmp traffic.

Openreach Gfast Modem directly to PC using W10

https://www.speedtest.net/result/18721152703.png

https://www.thinkbroadband.com/speedtest/17685158238...

The upload still high latency. Probably to do with Wellington Fibre Handover to network are fine with low latency for down and high latency for up.

Both same results FTTP or Gfast. No different whatsoever.

Anyway the case now closed!

FTTP and Gfast are fine.

Seem Banger on Plusnet TBB BQM on his FTTP 1000/115 are lots better than mine on TBB BQM as it should be like this for any full fibre. https://www.thinkbroadband.com/broadband/monitoring/...

Oh well. I leave it for now. Will switch to Plusnet next year January 2027.

Edited by adslmax (Thu 15-Jan-26 23:13:47)

In reply to a post by PCJM40:

In reply to a post by adslmax:

Sadly can't use Plusnet Hub Two for FTTP

Haven't you already tried two routers? really these 160ms spikes are either caused by devices connected to the FTTP router or they are the ISP, I can't see them being caused by the Openreach infrastructure.

Could be something interfering with the ONT or router,