Using it without discussing with me would seem particularly silly though, as they can't get useful results out of a 7530 router if another location is also pinging it given the throttling. I know from my chart it is mostly red with the occasional ping getting through, which would imply they're going to be seeing some packet loss they might not be expecting...

In reply to a post by SteveBushell999:

In reply to a post by dect:

I really surprised they didn't give you a SamKnows box for each as they could have learnt a lot from the comparisons.

Me too, I am a tester by trade, and I would have approached the whole thing very differently, right from the start!

Did you get sent the Samknows box by Zen support - and did set it up so you coud get your own results from it? I still don't have any access to the testing that the Samknows box Zen support sent me has done.

Having said that, on the basis of the results they've seen it sounds like I'm going to be migrated off Zen's network and back to BTW. It looks like from the Samknows info they're seeing an above average level of TCP re-transmissions on the downstream, and while it's not definitely cause and effect of the below par speed I get since migration, it has good legs I believe that the issues will go away once migrated. My performance went from basically perfect pre-migration (>900Mbps to almost any tester) to more often 550-750Mbps, with the occasional 900Mbps.

Will be interesting to see if it does change.

Out of interest, when you were migrated away from Zen GEA back to BTW, did you get sent an email like the following:
"We're sorry that you're leaving us and we'd like to thank you for being a customer.

Broadband Service: xxxxxxxxxxxx
Username: xxxxxxxxx@zen
We can confirm that your cancellation date will be: 31 August 2022."

It then goes on to detail how much I'll owe them for leaving early (nice!).

Finally ending with "Please note, if you're moving house or have recently upgraded/changed your service, don't worry, you can ignore this email. We have everything in hand and are processing your request.".

So you're none the wiser really as to whether this is a system error, or if you're actually getting cut off at the end of the month, or what. A bit of a mess really, sounds like a system in need of a bit of an overhaul.

Hi, Zen sent me the details for the SamKnows box, and I can log in to it. I have access to the past results, and all I can say is that the speeds are pretty constant, I am getting some disconnections, and the latency is increasing over time (Circa 10 mS -> 17mS).

When I was migrated back to BTW, there was no cancellation, only a Zen order for Migration, no cancellations. It does sound as though you are having issues with billing / contracts, who are you dealing with? Aside from sometimes all going quiet, I have had no real problems dealing with Andy and Brandon. It will be interesting if anything is said re: my disconnections or the latency climbing, but I am actually using my BTW line, and the Zen GEA one is just used by Zen at the moment, apart from some random speed tests, and occasional 70GB downloads just to validate it looks good to me.

As a matter of interest, I got a letter from ZZOOMM who will be cabling up this area soon, offering symmetrical packages right up to 2Gbits up/down, but you have to get a business package to get fixed ip, and it isn't super cheap (£129 for that business package or £99 on the home package with first 6 months of 24 month contract free) - Will be interesting to monitor that......

Hi Steve hope all is good mate, not convinced they will ever get to the bottom of this and its been going on for so long now that those who were originally interesting in the problem here are getting fewer and fewer.

Indeed! Agree with all those comments.... Pity that people are getting hung up on peripheral issues to the main thrust of this, but there you go. Steve

Congrats all on the second thread (we have had a thread, yes, but what about second thread?)

I've not been about for a while as this forum started to take up all my spare time 😢

Just a quick message to say that the connection has continued to be good. I haven't noticed any blips or slowdowns.

I should probably email Brandon with the news. He asked me to monitor the connection for a while... Well it's been quite a while now... Oops 😊

Interestingly my samknows box always complains of disconnections and sometimes a bit of packet loss, but if you look at the TBB BQM for those times, everything is fine. Dodgy ethernet cable going to the samknows box maybe?

I'm not really familiar with Samknows boxes, do they ping out from your property where TBB Quality Monitors ping in towards your router? could that be the difference?

It's OK, seems that everyone losing interest in favour of complaining of the service, and worsening of ping times. Both of my lines seem pretty good at the moment, but the GEA one does have latency that's variable, and there are disconnects and some packet loss (seemingly due to the disconnections). The whole thing is really fizzled out as it 'mended itself' - which is a real concern I think, as that which suddenly becomes good can equally suddenly become bad surely?

Regarding the SamKnows box, I really don't like it that it is behind the actual PPPOE connection (LAN-side of the FritzBox), so therefore not 'really' measuring the connection, but more what the FritzBox presents as the facts, but that's something for others to discuss, where is the actual 'boundary' of the connection? Don't forget that Zen say 'try a computer using PPPOE, and tell us what happens' as being the criteria when a connection is bad, NOT what you are getting out of the FritzBox! According to what I saw when it was bad, then there would have been a very different picture dependent on what router I used.

For interest - Using BTW line:

steve@Steves-Mac-mini ~ % traceroute 8.8.8.8
traceroute to 8.8.8.8 (8.8.8.8), 64 hops max, 52 byte packets
1 fritz.box (192.168.1.254) 0.914 ms 0.372 ms 0.219 ms
2 vt1.cor2.lond2.ptn.zen.net.uk (51.148.72.24) 8.973 ms 9.005 ms 9.360 ms
3 lag-9.p2.ixn-lon.zen.net.uk (51.148.73.208) 9.418 ms 9.236 ms 9.221 ms
4 lag-2.p2.thn-lon.zen.net.uk (51.148.73.138) 9.289 ms
lag-2.p1.thn-lon.zen.net.uk (51.148.73.132) 9.404 ms 9.346 ms
5 lag-2.br1.thn-lon.zen.net.uk (51.148.73.167) 9.517 ms 9.519 ms
lag-1.br1.thn-lon.zen.net.uk (51.148.73.153) 9.128 ms
6 72.14.223.28 (72.14.223.28) 9.695 ms 9.742 ms 9.909 ms
7 * * *
8 dns.google (8.8.8.8) 9.373 ms 9.214 ms 9.721 ms
steve@Steves-Mac-mini ~ %

Using Zen GEA Line:

steve@Mini-2011 ~ % traceroute 8.8.8.8
traceroute to 8.8.8.8 (8.8.8.8), 64 hops max, 52 byte packets
1 fritz.box (192.168.178.1) 1.164 ms 0.647 ms 0.459 ms
2 lo0-0.bng4.thn-lon.zen.net.uk (51.148.77.132) 16.402 ms 14.426 ms 15.581 ms
3 lag-14.p2.thn-lon.zen.net.uk (51.148.73.96) 18.625 ms
lag-14.p1.thn-lon.zen.net.uk (51.148.73.94) 11.833 ms 16.582 ms
4 lag-2.br2.thn-lon.zen.net.uk (51.148.73.155) 15.408 ms
lag-1.br2.ixn-lon.zen.net.uk (51.148.73.169) 11.184 ms
lag-1.br1.thn-lon.zen.net.uk (51.148.73.153) 12.452 ms
5 72.14.217.190 (72.14.217.190) 15.977 ms 11.527 ms
72.14.223.28 (72.14.223.28) 12.967 ms
6 74.125.242.97 (74.125.242.97) 12.263 ms 12.369 ms *
7 dns.google (8.8.8.8) 12.043 ms 12.230 ms
216.239.63.219 (216.239.63.219) 13.097 ms

Those tests a few seconds apart, seems BTW is better all the way?

Edited by SteveBushell999 (Fri 19-Aug-22 15:19:13)

In reply to a post by SteveBushell999:

It's OK, seems that everyone losing interest in favour of complaining of the service, and worsening of ping times. Both of my lines seem pretty good at the moment, but the GEA one does have latency that's variable, and there are disconnects and some packet loss (seemingly due to the disconnections). The whole thing is really fizzled out as it 'mended itself' - which is a real concern I think, as that which suddenly becomes good can equally suddenly become bad surely?

As you have a GEA and non-GEA connection, an interesting test you could do on your mac from the command line is to get a large file on each connection and see what the single-threaded performance is like.

I've been comparing my GEA line to another user's BTW line and I have much worse single threaded performance, but only to certain servers...(see some of the results from this post onwards... https://forum.kitz.co.uk/index.php/topic,27091.msg45... Speed tests may deploy variable numbers of threads, so depending on the extent of such issues you might see the impact on a speedtest, or might not.

You could try the following command lines from your Mac without having to install anything:

curl --output /dev/null http://ipv4.download.thinkbroadband.com:80/512MB.zip

and:

curl --output /dev/null http://slackware.uk/slackware/slackware-iso/slackwar...

These will be single threaded downloads. I see the thinkbroadband download capped around 35-40MB/sec, whereas the slackware one can achieve full line rate up to 100MB/sec. I have to have many multiple threads to the thinkbroadband site before I approach line rate. Whereas another user with a BTW Zen 900 connection finds his single thread to Thinkbroadband's download achieves line rate, which starts to appear to show some difference between the two.

To get multi-threaded downloads from command line in OSX you could use axel, but you have to install that via installing homebrew first. There is a tutorial here: https://macappstore.org/axel/

Command line for single thread TBB download with Axel:

rm 512MB.zip ; time axel -U Chrome -n 1 -v -a -o 512MB.zip http://ipv4.download.thinkbroadband.com/512MB.zip

To increase the number of threads, change -n 1 to -n 40, for instance, for 40 threads instead of one. I have to use 40 threads for the TBB download to approach line rate on my 900Mbps connection, which seems ludicrous.

The other download via Axel can be grabbed with this command line:

rm bigfile ; time axel -n 1 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...

Note axel barfs if the file is alread there and you try to overwrite it, hence why I have the rm first to remove the downloaded file.

Anyway, if you get a moment, it will be interesting to see if you see a difference between your GEA and BTW connections via this method. Please also note which gateway you are on for each (the second address on the traceroute, eg vt1.cor2.lond2.ptn.zen.net.uk or lo0-0.bng4.thn-lon.zen.net.uk from your examples).

Edited by jimbof (Sun 21-Aug-22 13:43:15)

Hi, and thank you Jimbof, now things get REALLY interesting. I had realised the multithread thing a while ago, and experimented on Fast.com (you can change the thread count with that speedtester) - The only conclusion I came to was the more is better, but that was when it was all broken.

Right, firstly BTW

steve@Steves-Mac-mini ~ % curl --output /dev/null http://ipv4.download.thinkbroadband.com:80/512MB.zip
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 512M 100 512M 0 0 54.4M 0 0:00:09 0:00:09 --:--:-- 57.8M
steve@Steves-Mac-mini ~ % curl --output /dev/null http://slackware.uk/slackware/slackware-iso/slackwar...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 3605M 100 3605M 0 0 35.9M 0 0:01:40 0:01:40 --:--:-- 48.4M
steve@Steves-Mac-mini ~ % curl --output /dev/null http://slackware.uk/slackware/slackware-iso/slackwar...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 3605M 100 3605M 0 0 35.8M 0 0:01:40 0:01:40 --:--:-- 50.1M
steve@Steves-Mac-mini ~ % rm bigfile ; time axel -n 1 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...
rm: bigfile: No such file or directory
Initializing download: http://slackware.uk/slackware/slackware-iso/slackwar...
File size: 3.5209 Gigabyte(s) (3780542464 bytes)
Opening output file bigfile
Starting download

[100%] [..........................................................................................................................................] [ 19.0MB/s] [00:00]

Downloaded 3.5209 Gigabyte(s) in 3:10 minute(s). (19412.85 KB/s)
axel -n 1 -v -a -o bigfile 4.32s user 17.92s system 11% cpu 3:10.58 total
steve@Steves-Mac-mini ~ % rm bigfile ; time axel -n 40 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...
Initializing download: http://slackware.uk/slackware/slackware-iso/slackwar...
File size: 3.5209 Gigabyte(s) (3780542464 bytes)
Opening output file bigfile
Starting download

[100%] [.................................................................................................................................................] [ 53.9MB/s] [00:00]

Downloaded 3.5209 Gigabyte(s) in 1:06 minute(s). (55232.45 KB/s)
axel -n 40 -v -a -o bigfile 7.36s user 25.87s system 49% cpu 1:07.43 total
steve@Steves-Mac-mini ~ % rm bigfile ; time axel -n 2 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...
Initializing download: http://slackware.uk/slackware/slackware-iso/slackwar...
File size: 3.5209 Gigabyte(s) (3780542464 bytes)
Opening output file bigfile
Starting download

Threads finishing etc....
[100%] [.................................................................................................................................................] [ 31.4MB/s] [00:00]

Downloaded 3.5209 Gigabyte(s) in 1:54 minute(s). (32127.28 KB/s)
axel -n 2 -v -a -o bigfile 4.20s user 18.34s system 19% cpu 1:54.99 total
steve@Steves-Mac-mini ~ % traceroute 8.8.8.8
traceroute to 8.8.8.8 (8.8.8.8), 64 hops max, 52 byte packets
1 fritz.box (192.168.1.254) 1.400 ms 0.329 ms 0.285 ms
2 vt1.cor2.lond2.ptn.zen.net.uk (51.148.72.24) 9.044 ms 11.268 ms 8.875 ms
3 lag-9.p2.ixn-lon.zen.net.uk (51.148.73.208) 9.095 ms 9.392 ms 9.613 ms
4 lag-2.p2.thn-lon.zen.net.uk (51.148.73.138) 9.493 ms
lag-2.p1.thn-lon.zen.net.uk (51.148.73.132) 9.331 ms 9.445 ms
5 lag-2.br1.thn-lon.zen.net.uk (51.148.73.167) 9.510 ms 9.272 ms
lag-1.br1.thn-lon.zen.net.uk (51.148.73.153) 9.239 ms
6 72.14.223.28 (72.14.223.28) 9.644 ms 9.679 ms 9.587 ms
7 * * *
8 dns.google (8.8.8.8) 9.783 ms 8.963 ms 13.665 ms
steve@Steves-Mac-mini ~ %

NOW, for GEA migrated line

steve@Mini-2011 ~ % curl --output /dev/null http://ipv4.download.thinkbroadband.com:80/512MB.zip
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 512M 100 512M 0 0 59.9M 0 0:00:08 0:00:08 --:--:-- 60.7M
steve@Mini-2011 ~ % curl --output /dev/null http://slackware.uk/slackware/slackware-iso/slackwar...

% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
steve@Mini-2011 ~ % curl --output /dev/null http://slackware.uk/slackware/slackware-iso/slackwar...

% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 3605M 100 3605M 0 0 60.6M 0 0:00:59 0:00:59 --:--:-- 60.7M
steve@Mini-2011 ~ % rm bigfile ; time axel -n 1 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...
rm: bigfile: No such file or directory
Initializing download: http://slackware.uk/slackware/slackware-iso/slackwar...
File size: 3.5209 Gigabyte(s) (3780542464 bytes)
Opening output file bigfile
Starting download

[100%] [....................................................................................................................................................] [ 58.7MB/s] [00:00]

Downloaded 3.5209 Gigabyte(s) in 1:01 minute(s). (60136.08 KB/s)
axel -n 1 -v -a -o bigfile 7.10s user 30.68s system 61% cpu 1:01.48 total
steve@Mini-2011 ~ % rm bigfile ; time axel -n 40 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...
Initializing download: http://slackware.uk/slackware/slackware-iso/slackwar...
File size: 3.5209 Gigabyte(s) (3780542464 bytes)
Opening output file bigfile
Starting download

Threads finishing etc....
[100%] [....................................................................................................................................................] [ 50.5MB/s] [00:00]

Downloaded 3.5209 Gigabyte(s) in 1:11 minute(s). (51762.57 KB/s)
axel -n 40 -v -a -o bigfile 6.79s user 34.00s system 55% cpu 1:13.74 total
steve@Mini-2011 ~ % rm bigfile ; time axel -n 2 -v -a -o bigfile http://slackware.uk/slackware/slackware-iso/slackwar...
Initializing download: http://slackware.uk/slackware/slackware-iso/slackwar...
File size: 3.5209 Gigabyte(s) (3780542464 bytes)
Opening output file bigfile
Starting download

Connection 0 finished
Connection 1 finished
Connection 1 finished
Connection 0 finished
[100%] [....................................................................................................................................................] [ 49.7MB/s] [00:00]

Downloaded 3.5209 Gigabyte(s) in 1:12 minute(s). (50864.11 KB/s)
axel -n 2 -v -a -o bigfile 6.73s user 29.95s system 50% cpu 1:13.23 total
steve@Mini-2011 ~ % traceroute 8.8.8.8
traceroute to 8.8.8.8 (8.8.8.8), 64 hops max, 52 byte packets
1 fritz.box (192.168.178.1) 1.420 ms 0.591 ms 0.449 ms
2 lo0-0.bng4.thn-lon.zen.net.uk (51.148.77.132) 13.838 ms 14.579 ms 15.640 ms
3 lag-14.p2.thn-lon.zen.net.uk (51.148.73.96) 12.920 ms
lag-14.p1.thn-lon.zen.net.uk (51.148.73.94) 12.315 ms 12.554 ms
4 lag-2.br1.thn-lon.zen.net.uk (51.148.73.167) 12.958 ms 12.358 ms
lag-1.br1.thn-lon.zen.net.uk (51.148.73.153) 12.906 ms
5 72.14.223.28 (72.14.223.28) 12.365 ms
72.14.217.190 (72.14.217.190) 12.838 ms 11.754 ms
6 108.170.246.129 (108.170.246.129) 12.785 ms
74.125.242.97 (74.125.242.97) 14.753 ms *
7 108.170.234.231 (108.170.234.231) 12.503 ms
dns.google (8.8.8.8) 11.412 ms
142.250.215.127 (142.250.215.127) 11.709 ms
steve@Mini-2011 ~ %

Conclusions? Both connecting via London, single thread very poor on BTW, but better gets good at 40 threads. As for GEA: - Well, seems I have more raw bandwidth available, and that it is better with much less threads - what's your view?

Steve

Edited by SteveBushell999 (Sun 21-Aug-22 15:33:00)