Internet seemingly throttled despite expected ping and speedtest-cli outputs

Not very many, and only the bottom two showed up when I was actually doing the test, and certainly not every time the 3-5 second buffer was happening.

I hadn’t noticed nmcli connection show had included device names, but had replaced with the profile to where IPv6 had been disabled correctly.

Ok, 9 retransmits per second is not optimum, but also not alarming. So it is not the only cause for transport delay. DNS, IPv6, some minor optimizations did also not fix the issue. I am pretty sure there is nothing wrong with your pc networking setup. It is unclear why the rendering load is much higher in your second profile compared to the first. That would be an issue with firefox or the rendering interface VA-API. You could test google chrome to see if there are differences.

After DNS and IPv6 the issue doesn’t seem nearly as bad on chromium actually. The infinite loading buffering is much better, but loading a heavy page like youtube is still slow. I’m not sure how to share the chromium profiles. I have them downloaded as .jsons but cannot upload them as such.

I would not be surprised if Chrome shows better speed with youtube. There have been persistent complaints over the years that google has “optimized” youtube for their own browser and throwing sand in the gears for other browsers. That is an ongoing browser war. Disgraceful.
When i run youtube on chromium it is obvious that it is faster than firefox, but that’s not for free as i don’t get the ads blocked and they spy on me. I personally don’t use chrome.

What’s the problenm with uploading? If it is just too big, json compresses very well. Use something like pcloud or google drive sync.com whatever you prefer.

Not certain which is which but I did one google images scroll test and the other loading a youtube page.

Agreed, we need regulations. By standard I use librewolf, but for the tests thought it would be simpler to show on plain firefox.

Both profiles show that scripting load is the major part in the timeline. The youtube profile suggests that adblock may help a little. You may try to see if that really improves to block doubleclick.net stuff. The youtube profile shows clearly the bottle neck is script processing. So even with faster and lossless internet connection you would not gain much.

How would I test or improve scripting load specifically? Is that the difference between the speeds I’m getting and what this PC used to get on windows and other PCs on the network get?
I have access to a different router/ISP today so thought I’d rerun all the tests. This one seems to have a route for IPv6 so haven’t bothered disabling it, but the change to 1.1.1.1 DNS is applied here too.
https://share.firefox.dev/3X4xeCi

[laura@pigeon ~]$ dig google.com  
  
; <<>> DiG 9.20.0 <<>> google.com  
;; global options: +cmd  
;; Got answer:  
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 54228  
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1  
  
;; OPT PSEUDOSECTION:  
; EDNS: version: 0, flags:; udp: 1232  
;; QUESTION SECTION:  
;google.com.                    IN      A  
  
;; ANSWER SECTION:  
google.com.             137     IN      A       142.250.179.238  
  
;; Query time: 10 msec  
;; SERVER: 1.1.1.1#53(1.1.1.1) (UDP)  
;; WHEN: Wed Aug 14 16:11:14 BST 2024  
;; MSG SIZE  rcvd: 55
                               My traceroute  [v0.95]  
pigeon (2a00:23c6:c20d:5701:5634:a113:2c4c:d3d9) -> google.com2024-08-14T16:17:10+0100  
Keys:  Help   Display mode   Restart statistics   Order of fields   quit  
                                             Packets               Pings  
Host                                       Loss%   Snt   Last   Avg  Best  Wrst StDev  
1. 2a00:23c6:c20d:5701:1a82:8cff:fe67:8f10  0.0%   114    1.3   2.6   1.2  27.1   3.3  
2. 2a00:2302::1100:203:371                  0.0%   114    6.3   6.9   5.6  16.1   1.9  
3. (waiting for reply)  
4. 2a00:2302::1100:100:4f                  92.0%   114    8.7  10.3   8.7  14.5   1.8  
5. peer2-et0-1-7.slough.ukcore.bt.net       0.0%   114    9.7  11.8   9.1  31.5   4.5  
6. 2001:4860:1:1::8cc                       0.0%   113    9.1  10.6   8.6  23.7   2.8  
7. 2a00:1450:8165::1                        0.0%   113   10.9  12.0  10.4  25.6   2.4  
8. 2001:4860:0:1::54ce                      2.7%   113   12.6  10.5   9.0  17.9   1.8  
9. 2001:4860:0:1::7dfc                     31.9%   113   10.2  11.8   9.9  23.3   2.4  
10. 2001:4860::c:4002:1123                   0.0%   113   11.3  14.2  10.7 114.7  11.0  
11. 2001:4860::9:4001:2716                   0.0%   113   10.4  15.8   9.8  86.9  12.8  
12. 2001:4860:0:1::7dc3                     38.1%   113   11.6  12.2  10.6  22.7   2.3  
13. 2001:4860:0:1::537b                      0.0%   113   13.4  11.1   9.6  17.8   1.8  
14. lhr48s28-in-x0e.1e100.net                0.0%   113   12.2  11.5   9.7  18.2   2.3
[laura@pigeon ~]$ iperf3 -t 8 -c iperf.as42831.net -p 5300-5400  
Connecting to host iperf.as42831.net, port 5300  
[  5] local 2a00:23c6:c20d:5701:5634:a113:2c4c:d3d9 port 36220 connected to 2a01:a500:6  
::2 port 5300  
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd  
[  5]   0.00-1.00   sec  2.88 MBytes  24.1 Mbits/sec    0    137 KBytes          
[  5]   1.00-2.00   sec  2.12 MBytes  17.8 Mbits/sec    0    159 KBytes          
[  5]   2.00-3.00   sec  1.88 MBytes  15.7 Mbits/sec    0    166 KBytes          
[  5]   3.00-4.00   sec  1.75 MBytes  14.7 Mbits/sec    0    180 KBytes          
[  5]   4.00-5.00   sec  2.25 MBytes  18.9 Mbits/sec    0    180 KBytes          
[  5]   5.00-6.00   sec  2.25 MBytes  18.9 Mbits/sec    0    187 KBytes          
[  5]   6.00-7.00   sec  2.12 MBytes  17.8 Mbits/sec    0    197 KBytes          
[  5]   7.00-8.00   sec  2.25 MBytes  18.9 Mbits/sec    0    245 KBytes          
- - - - - - - - - - - - - - - - - - - - - - - - -  
[ ID] Interval           Transfer     Bitrate         Retr  
[  5]   0.00-8.00   sec  17.5 MBytes  18.3 Mbits/sec    0             sender  
[  5]   0.00-8.13   sec  16.5 MBytes  17.0 Mbits/sec                  receiver
[laura@pigeon ~]$ ip a  
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen  
1000  
   link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00  
   inet 127.0.0.1/8 scope host lo  
      valid_lft forever preferred_lft forever  
2: eno1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN group d  
efault qlen 1000  
   link/ether 48:ba:4e:b3:ea:e6 brd ff:ff:ff:ff:ff:ff  
   altname enp2s0  
3: wlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group defau  
lt qlen 1000  
   link/ether f8:94:c2:b8:fd:6c brd ff:ff:ff:ff:ff:ff  
   inet 192.168.1.85/24 brd 192.168.1.255 scope global dynamic noprefixroute wlan0  
      valid_lft 86172sec preferred_lft 86172sec  
   inet6 2a00:23c6:c20d:5701:5634:a113:2c4c:d3d9/64 scope global dynamic noprefixroute  
   
      valid_lft 294sec preferred_lft 114sec  
   inet6 fe80::ea87:b0d8:9233:da26/64 scope link noprefixroute    
      valid_lft forever preferred_lft forever  
[laura@pigeon ~]$ ip r  
default via 192.168.1.254 dev wlan0 proto dhcp src 192.168.1.85 metric 600    
192.168.1.0/24 dev wlan0 proto kernel scope link src 192.168.1.85 metric 600
[[laura@pigeon ~]$ dig @1.0.0.1 pickles.org AAAA  
;; communications error to 1.0.0.1#53: timed out  
;; communications error to 1.0.0.1#53: timed out  
;; communications error to 1.0.0.1#53: timed out  
  
; <<>> DiG 9.20.0 <<>> @1.0.0.1 pickles.org AAAA  
; (1 server found)  
;; global options: +cmd  
;; no servers could be reached](<[laura@pigeon ~]$ dig google.com AAAA

; %3C<%3E> DiG 9.20.0 <<>> google.com AAAA
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 13876
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 1232
;; QUESTION SECTION:
;google.com.                    IN      AAAA

;; ANSWER SECTION:
google.com.             78      IN      AAAA    2a00:1450:4009:822::200e

;; Query time: 9 msec
;; SERVER: 1.1.1.1#53(1.1.1.1) (UDP)
;; WHEN: Wed Aug 14 16:26:54 BST 2024
;; MSG SIZE  rcvd: 67
[laura@pigeon ~]$ ping google.com  
PING google.com (2a00:1450:4009:81e::200e) 56 data bytes  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=1 ttl=58 time=9.13 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=2 ttl=58 time=12.2 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=3 ttl=58 time=10.5 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=4 ttl=58 time=10.0 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=5 ttl=58 time=10.4 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=6 ttl=58 time=9.82 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=7 ttl=58 time=10.1 ms  
64 bytes from lhr25s32-in-x0e.1e100.net (2a00:1450:4009:81e::200e): icmp_seq=8 ttl=58 time=9.94 ms  
^C  
--- google.com ping statistics ---  
8 packets transmitted, 8 received, 0% packet loss, time 7008ms  
rtt min/avg/max/mdev = 9.128/10.256/12.176/0.824 ms
[laura@pigeon ~]$ speedtest-cli  
Retrieving speedtest.net configuration...  
Testing from BT (86.138.126.41)...  
Retrieving speedtest.net server list...  
Selecting best server based on ping...  
Hosted by VoiceHost Ltd (London) [105.78 km]: 16.467 ms  
Testing download speed.............................................................................  
...  
Download: 21.63 Mbit/s  
Testing upload speed...............................................................................  
.......................  
Upload: 18.87 Mbit/s

The java script load depends on the engine in your browser. Internet Explorer had it’s own, earlier versions of edge used it too. Google came up with it’s V8 engine, and Edge now uses that too. SpiderMonkey is the engine from Mozilla, also used in Gnome Shell. Apple, like KDE utilize KIT, today Apple developed it to JSC. They are all different in power. adblockers can help, if they reduce the amount of java code that is loaded. How much that helps depends on the percentage of ads in a web page. The third key with influence is the cpu.
Your profiling results show again that your cpu is busy all the time with java interpretation. renderer load is low, network is low. If you say other pcs in your network do not see these 20 s until content is displayed, that could be related to other cpu. I have no explanation for a difference with windows on your laptop and a change to EOS when you run the same browser. I think i am going to test that.

I FIGURED IT OUT!!! I was trying to figure out how to get my konsole to open faster since I had ~1.3s launch. This post had a similar issue and said power save was the problem, so I poked around in settings and found Power Management > On AC Power > Other Settings > Switch to power profile was set to Leave unchanged. I switched it to Performance and INSTANTLY my entire laptop became ~1s faster at every action. All web page loads, all program starts, tabbing between applications, switching discord channels, EVERYTHING. Hooray. Thanks a ton for all the diagnosis support, sorry the question was malformed.

3 Likes

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.