VM Router Not impressed with SuperHub

japitts · 24-04-2012, 23:51

As per http://tinyurl.com/d4wtrsl

Last night I moved from 10Mb to 30Mb and migrated across to the SuperHub as part of it. It's not worked from the start, and I've played with more settings and restarts than I care to mention... what I can reliably replicate is..

Superhub in modem mode, going through my old Netgear WGR614 router, I can get typically 8-10Mb, but WiFi is reliable.

Superhub in wireless mode, I can scrape maybe 10-15Mb. Has been as low as 0.5Mb

Superhub in wired mode, I can make 20Mb.

Only an hour ago, I lost wireless completely, and could not get any internet access except by using modem mode and going through Netgear.

Digging around in the "connection" section of the Superhub maintenance screens, the "max traffic rates" do seem to be set to 30down/2 up... so it's either the Superhub hardware, or the DOCSIS3 (??) network that's causing this....

I've certainly never achieved more than 20Mb, which I might be inclined to blame on congestion in the area (BS37).. but I am within an inch of calling VM to "upgrade" back to 10Mb and the old Motorola surfboard modem. At least that setup worked.

Apart from the Superhub being rubbish (hence the whole point of using modem mode)... is there anything I'm missing here?! Any pointers would be appreciated.

Thanks

General Maximus · 25-04-2012, 09:02

your max traffic rate is the speed you are provisioned for by VM i.e. 30mbit down and at the modem, 2mbit up till they bang everyone up to 3mbit when the upgrades are completed. The shub can obviously handle more than that because peeps also use it for 60mbit and 100mbit.

The shub is useless for wireless so I wouldn't even bother. If you are using your "old netgear router" I am assuming it is 802.11g which means the most you are going to get out of it wirelessly is 22mbit max. The best thing you can do to see where the problem lies is to do a couple of speed tests from speedtest.net using the londer server (in the morning) with the shub in router mode to start off with and pc directly connected, and them in modem mode with your netgear router attached. Then we'll know which one is causing the problem.

When you put the shub in modem mode can you go into the connections tab paste your power levels etc into here.

kwikbreaks · 25-04-2012, 09:16

Hmmm...
If you had a modem before and not a Superhub moving from 10Mbps to 30Mbps has put you from the legacy network to DOCSIS3 and you may be unlucky enough to have a torrent freak nearby causing congestion. Now you are on DOCSIS3 there is no going back unless you can get back to your old modem as downgrading to 10Mbps with a Superhub will leave you on DOCSIS3.

Set up a TBB monitor and post it up after a day's worth

japitts · 25-04-2012, 10:51

Thanks for the reply. I probably should have mentioned in my original post - to get the speed figures I've been using a combination of speedtest.net, and "multiple concurrent downloads".

That SH wireless is so bad doesn't surprise me - that I'm not even getting anywhere near maximum speed when in modem mode, is what's been causing some serious head-scratching. 20Mb or so wired, 7-9Mb tops wirelessly, that's worse than the old Motorola Surfboard. It's almost "as if" the SH is capping the speed when anything other than the Ethernet cable is used...

I probably will change my Netgear for an 802.11g box, but right now it's not as if that's anywhere close to being maxed out. Will probably have to do battle with VM Tech Support soon...

kwikbreaks · 25-04-2012, 11:36

Concentrate on checking your connection speed using ethernet as 802.11g WiFi tops out at a maximum of 22-23Mbps so can't be used to decide if a 30Mbps connection is up to par. It was your 20Mbps wired results that made me question whether or not congestion may be playing a hand.

Speedtest.net Birmingham, Maidenhead, and London servers are all up to the task of checking if a 30Mbps connection is giving the headline speed. The most reliable of course is multiple downloads rather than speedtests but imo Speedtest.net is plenty good enough to test 30Mbps.

Sephiroth · 25-04-2012, 12:25

To amplify Kwikkie's excellent advice: Speedtest.net is now becoming somewhat unrliable. I now use speedtest.bbmax.co.uk which at least doesn't stutter.

But simultaneous downloads is the real test. Subject to seeing your modem stats, at first sight it seems possible that what coud support 10 meg in traffic terms can't support 30 meg. And, of course, VM won't return you to the legacy network if you downgrade, but you are more likely in a congested area to get 10 meg than 30 meg.

The SH in modem mode won't slow you down.

kwikbreaks · 25-04-2012, 14:16

Seph - I don't believe that the BBMax speedtest downloads sufficient data for it to be a good test.

These are the numbers from my Tomato router stats for a run of it (there was a small amount of other traffic while it ran)..

RX 0.40 kbit/s
(0.05 KB/s) Avg 128.75 kbit/s
(15.72 KB/s) Peak 10056.44 kbit/s
(1227.59 KB/s) Total 9,429.83 KB

TX 0.37 kbit/s
(0.04 KB/s) Avg 29.21 kbit/s
(3.57 KB/s) Peak 1082.79 kbit/s
(132.18 KB/s) Total 2,139.40 KB

The test results - http://www.speedtest.bbmax.co.uk/res...437&v=15683459

So it used less than a 10MB test file down and ~ 2MB up. Probably OK for my 10Mbps but surely insufficient for faster connections - from memory when I was on 50Mbps it didn't use a bigger file but of course I could be wrong.

alexcopeland · 25-04-2012, 15:27

I'd advise you monitor your wireless connection from your laptop. If you're in a busy wireless environment crowded by wireless networks from your neighbours the super hub signal will sometimes struggle to get through the air waves. I'd advise you reduce the 802.11 mode to 145 Mbps and change the wireless channel to a less congested one after running a program like inSIDDer.

japitts · 25-04-2012, 23:35

Well my connection levels, taken earlier tonight...

Startup Procedure
Procedure Status Comment
Acquire Downstream Channel 315000000 Hz Locked
Connectivity State OK Operational
Boot State OK Operational
Configuration File OK
Security Enabled BPI+

Downstream Channels
Lock Status Modulation Channel ID Max Raw Bit Rate Frequency Power SNR Docsis/EuroDocsis locked
Locked QAM256 212 55616000 Kbits/sec 315000000 Hz -3.3 dBmV 32.7 dB Hybrid
Locked QAM256 209 55616000 Kbits/sec 291000000 Hz -2.8 dBmV 34.2 dB Hybrid
Locked QAM256 210 55616000 Kbits/sec 299000000 Hz -2.9 dBmV 33.5 dB Hybrid
Locked QAM256 211 55616000 Kbits/sec 307000000 Hz -3.0 dBmV 33.1 dB Hybrid
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV 0.0 dB Unknown
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV 0.0 dB Unknown
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV 0.0 dB Unknown
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV 0.0 dB Unknown

Upstream Channels
Lock Status Modulation Channel ID Max Raw Bit Rate Frequency Power
Locked ATDMA 1 20480 Kbits/sec 35800000 Hz 49.3 dBmV
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV
Unlocked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV

Primary Downstream Service Flow
Downstream(0)
SFID 22229
Max Traffic Rate 33000000 bps
Max Traffic Burst 3044 bytes
Mix Traffic Rate 0 bps

Primary Upstream Service Flow
Upstream(0)
SFID 22228
Max Traffic Rate 2058000 bps
Max Traffic Burst 8160 bytes
Mix Traffic Rate 0 bps
Max Concatenated Burst 8160 bytes
Scheduling Type Best Effort

From more experimenting tonight, I'm pretty happy fault doesn't lie with DOCSIS3/VM, and it's probably my router.. however one inconsistency remains, that I can't figure out, and which was causing me to really get confused before...

SH in wireless mode - all proven, useless and unreliable at anything more than 10Mb.
SH in modem mode - Ethernet to the laptop, anything between 25-29Mb depending on which testing method is used. VM connection proven good.
SH in modem mode - Ethernet into Netgear WGR614, wireless - I can scrape approx 8Mb. Yet the old Motorola Surfboard would consistently get high-9s via the same router. Which proves the Netgear is capable of delivering more than 10Mb.

I'm looking at investing in a new Netgear WNR1000 as the current equivalent, but with 802.11n speeds... although I just cannot understand why I'm getting slower speeds from the SH than the Surfboard did. That's what set this whole "poking and prodding" off, and I still can't work it out.

qasdfdsaq · 26-04-2012, 04:08

Quote:

Originally Posted by alexcopeland

change the wireless channel to a less congested one after running a program like inSIDDer.

Common misconception.

inSSIDer cannot measure how congested a wireless channel is.

---------- Post added at 04:08 ---------- Previous post was at 04:07 ----------

Quote:

Originally Posted by japitts

SH in modem mode - Ethernet into Netgear WGR614, wireless - I can scrape approx 8Mb. Yet the old Motorola Surfboard would consistently get high-9s via the same router. Which proves the Netgear is capable of delivering more than 10Mb.

What happens when you do wired into the WGR614?

japitts · 26-04-2012, 08:02

Quote:

Originally Posted by qasdfdsaq

Common misconception.

inSSIDer cannot measure how congested a wireless channel is.

---------- Post added at 04:08 ---------- Previous post was at 04:07 ----------

What happens when you do wired into the WGR614?

Unfortunately I've no way of testing this, don't think there's any other Ethernet cables knocking around at the mo :-(

kwikbreaks · 26-04-2012, 08:30

Quote:

Originally Posted by japitts

Unfortunately I've no way of testing this, don't think there's any other Ethernet cables knocking around at the mo :-(

.. and wow they're so expensive - some on eBay cost nearly £1.50 delivered.

japitts · 26-04-2012, 17:39

Quote:

Originally Posted by qasdfdsaq

Common misconception.

inSSIDer cannot measure how congested a wireless channel is.

---------- Post added at 04:08 ---------- Previous post was at 04:07 ----------

What happens when you do wired into the WGR614?

Tried this now - 25 or so Mb, which considering peak times is pretty respectable. So it seems the problem is indeed 802.11g constraints, although I'm still at a loss as to why this method only gets 8 or so Mb when it got much closer to 10Mb before. Most bizarre.

Guess it's time for a new router then.... Thanks for the suggestions.. has anyone got any good or bad feedback about these two routers...

http://tinyurl.com/d7m77pl or http://tinyurl.com/d7m77pl - the 1000 seems to fit the bill except for a few reviews I've seen of poor wireless performance.

TIA.

qasdfdsaq · 27-04-2012, 11:16

Same link... the first device is a "half rate" 11n router so obviously will perform poorly. By default it'd run in 65mbps mode and give you about 30mbps useable speed. You'll want a proper "N300" router which can be had for less than £10 second hand anyway.