5) Re-ran speed test - speed had doubled to the speed mentioned above
6) Also worth noting I loaded up Steam and tried to uninstall and re-download some of my games to test the real download speed. They matched approximately what speedtest had predicted (in total).
As far as I'm aware there's nothing else I can toggle to improve the speed. I know speed tests aren't always realistic, but this is abysmally slow. My old 5MB ADSL with plus.net was faster than this.
Here are my stats:
Tracing route to www-cctld.l.google.com [18.104.22.168] over a maximum of 30 hops:
1 11 ms 9 ms 9 ms 10.89.176.1 2 12 ms 9 ms 10 ms croy-core-1a-ge22.network.virginmedia.net [81.96 .224.197] 3 46 ms 11 ms 9 ms croy-core-2a-ae2-0.network.virginmedia.net [195. 182.178.82] 4 24 ms 11 ms 19 ms brnt-bb-1a-ae9-0.network.virginmedia.net [81.96. 226.1] 5 17 ms 17 ms 15 ms manc-bb-1b-as5-0.network.virginmedia.net [62.252 .192.90] 6 44 ms 24 ms 23 ms tele-ic-3-ae0-0.network.virginmedia.net [212.43. 163.70] 7 77 ms 75 ms 69 ms 138-14-250-212.static.virginmedia.com [212.250.1 4.138] 8 24 ms 23 ms 75 ms 22.214.171.124 9 24 ms 25 ms 33 ms 126.96.36.199 10 30 ms 28 ms 31 ms 188.8.131.52 11 39 ms 31 ms 30 ms 184.108.40.206 12 * * * Request timed out. 13 50 ms 31 ms 30 ms wi-in-f94.1e100.net [220.127.116.11]
By the looks of it, a server outside of the Virgin network is conking out?
Your downstream power levels are right on the edge of the acceptable range (Slightly over)
Though this should not cause that slow speed. Are you testing over a wired connection to the SH and also can you try it in modem mode connected to eth port 1 (Next to the coax) and see what results you get.
You will have to do more then one speed test to get some sort of result as the servers can be busy and give a false reading.
Vivid 200Mb (SH2AC & Hub 3.0) 2XTivo TV XL + Sky Movies. Phone M. 100Mb Business (Hitron Hub).
Apologies, I did say I was hooked directly into the router, but perhaps I should have specifically said wired.
Thank you for the advice. I have tried modem mode and it made no difference at all. I have used a number of speed test sites at different times during the day and this morning as well (after turning on the router which was off overnight), with no effect on the poor speed. It is consistently poor, time of day doesn't seem to have an effect.
Might I recommend that when you ask someone to try modem mode, you mention that it actually changes the IP address of the router to a different subnet? I had to use ipconfig to refresh the connection and there's no information on that in the router screen. So essentially you are stuck without internet and no way of renabling it unless you have another device to google the answer.
Also as another test, I used the wireless with my phone right next to the hub. It gives similar speed results. If I turn off WiFi and use my TMobile account, it's actually faster than the wireless and wired.
I have just checked your connection and network segment. Everything appears to be OK at present and there do not appear to have been any network problems recently. If you are still experiencing difficulties then please re-post with the information described here taken at the time, together with traceroutes to bbc.co.uk (no WWW) and any other sites which are causing you problems.
We have seen Kaspersky and Avast! interfering with speeds so, if you are using either, please disable it when speed testing. Please also check that the power supply is 12Vac, 1.5A. A CAT 6 ethernet cable is also recommended.
If you are using speedtest.net to check your speeds then I suggest that you use the Paris server option as it is reputed to be more accurate/reliable than some of the other servers. You may also find that less heavily loaded sites (e.g. speedtest.magnet.ie or top10.com) give more accurate/reliable results. Speedtests should always be run from a single, ethernet connected PC, with no devices connected wirelessly.
Concurrently downloading multiple large files from multiple servers and adding the throughput will usually give a more accurate indication of actual download speeds than speed testing sites.
Help & Support Forum Team
If someone's helped you out say thanks by clicking on the Kudos Star. If someone's solved your problem, why not mark their message as a Helpful Answer? You can also help other users find useful posts by Tagging them with keywords.
If you are new to the forum, please remember to search the forum before posting your question. Please don't send private messages directly to the forum team unless they ask you to, post your issue to the forum first.
It's a Super Hub, it's using the default power supply that comes with it.
See first post for Modem connection details.
Measurements of speed for both download and upload are reasonably consistent regardless of when I measure them (Saturday various times during the day, Sunday at various times during the day). I'll redo the speedtests below tomorrow morning (the ones below were taken at 23:00 after I read your post).
As I said in the original post, I ran Steam and saw how long it took to download a few games at once. The speeds are consistent with the web page speed tests and are much slower than my old ADSL plus.net connection (which was way less than the 30MB offered at Virgin).
But I will also run the GamersHell speed test: 2 files downloading concurrently: #2 The Sims 3 v1.31.118 Patch for PC - 74 kB/sec #3 CABAL Online Episode VII: Saint's Requiem Client - 70 kB/sec From Google - 1 kilobyte = 0.0078125 megabits So 74 kB/sec + 70 kB/sec = 144 kB/sec = 1.125 Mbps (I believe - which makes it quite similar to the speed tests) On my old ADSL connection, I used to be able to have a few games downloading from Steam at around 540 kB/s in total (I believe I had a 5 meg connection)
Tracing route to bbc.co.uk [18.104.22.168] over a maximum of 30 hops:
1 12 ms 25 ms 12 ms 10.89.176.1 2 9 ms 9 ms 27 ms croy-core-1b-ge21.network.virginmedia.net [81.96 .224.225] 3 20 ms 13 ms 10 ms croy-core-2b-ae2-0.network.virginmedia.net [195. 182.178.86] 4 11 ms 11 ms 11 ms popl-bb-1b-ae10-0.network.virginmedia.net [81.96 .226.5] 5 12 ms 31 ms 11 ms glfd-bb-1a-ae4-0.network.virginmedia.net [213.10 5.159.61] 6 12 ms 69 ms 11 ms glfd-bb-1b-ae0-0.network.virginmedia.net [213.10 5.172.6] 7 12 ms 35 ms 13 ms glfd-tmr-1-ae5-0.network.virginmedia.net [213.10 5.159.46] 8 13 ms 13 ms 14 ms tcl5-ic-1-as0-0.network.virginmedia.net [62.253. 185.78] 9 14 ms 21 ms 58 ms 22.214.171.124 10 * * * Request timed out. 11 30 ms 13 ms 13 ms ae1.er01.rbsov.bbc.co.uk [126.96.36.199] 12 22 ms 21 ms 19 ms 188.8.131.52 13 14 ms 13 ms 29 ms 184.108.40.206
Ok I think I've found the problem, it must be something to do with the cabling. The fibre optic cable on my upper floor wasn't functional so the engineer suggested I just use the Ethernet Over Power adapters I had instead.
Using my phone with the speedtest app I re-did the WiFi tests and got:
Unfortunately that's right next to the router. If I try the WiFi test upstairs, it's only 2Mbps because the signal degrades so quickly.
My TV used to stream movies using Ethernet Over Power so it's not like it's caused problems in the past. And what I don't understand is why the Upload speed doesn't appear to be affected by the Ethernet Over Power, but the download speed does. Can you think of a reason for this?
Also could I suggest that you add an additional test to the Speed Problems sticky? It would be a good idea, if possible, for the user to use their phone to test the speed over WiFi as this would help eliminate cabling issues very quickly.
Also it would be really good if you could programmatically supply the information you normally request and get an "OK" status from a webpage. Presumably you have some internal tools that allow you to check the server loads. And while you don't want a user to be able to see the load on the network hub as they could demand to be moved elsewhere unnecessarily, a simple green OK light would be enough. It would hopefully save you guys some time and cut down on the amount of people posting .
I'm still not sure what to do about the cabling. I'll try some different Ethernet cables and also see if I can find a really long one. Either that or it would be good if an engineer could fix the fibre optic cable on my upper floor.
Ok I think I've found the problem, it must be something to do with the cabling. .....And what I don't understand is why the Upload speed doesn't appear to be affected by the Ethernet Over Power, but the download speed does. Can you think of a reason for this?
You don't have FibreOptic anyway(unless you specifically deployed this on your Lan internally yourself in which case you probably wouldn't be asking the question), but presumably you're referring to seperate coax (upstairs) feed via a splitter in external omnibox or via internal wall box that feeds your upstairs previous located CM/Hub?
If so the degradation could potentially be caused by faulty dielectic/shielding damage on that cable run which will have more attenuation/effect on higher downstream frequencies used than far lower (multiple of 10 approx less) utilised on upstream.
So do you have two coax cable (and not just Lan Cat5 utp cable ethernet) feeds to upstairs and downstairs as I think you implied with your ambiguous terminology?
If however you're only experiencing this d/s degradation on Homeplug's as opposed to direct (ethernet/cat5x) cabled connections then some leakage/similar frequency impairment must exist on homeplug units/mains cabling?
Regards Tony "Life is a Binary Inspired Turing Computed Hologram"(don't PM or @Mention me - in case ignoring you offends)
Ah I'm really sorry, I just assumed that's what a fibre optic cable looked like, bad me . There's a Coax cable downstairs that comes into the house. Then a splitter. Then the Coax cabling disappears into the stairs and the engineer reckoned there was another splitter somewhere in the stair boards which was malfunctioning, because there's another Coax cable on the other side of the stairs that doesn't work and one that goes upstairs, which also doesn't work. The engineer suggested I use the HomePlug as he didn't have his toolkit with him and it would be just as good.
Unfortunately I didn't own the house when the cabling was put in, so trying to work out where it all goes is a bit difficult.
So I believe it is degradation either in the HomePlug units or in the electric wiring of the house. I will see if I can find a long ethernet cable to prove it to myself. I didn't realise the downstream frequencies would be affected more greatly than the upstream ones.