HubOne 5Ghz Issues
FIXED
@Mikewalsh wrote:
Hi all
I have been eyeing up the netgear d7000 as a replacement for my Tplink. It is £40 cheaper than the d7800 on Amazon (no fancy mu-mimo tho) Set up looks about the same and it uses beam forming wifi (whatever that is).
Regards
Mike
A very geeky friend of mine has the D7000 and loves it!
Beamforming is basically just a way of using constructive interference from a phased array to transmit a greater amplitude (essentially a stronger signal) in the direction of the receiver.
@ejs wrote:
To see what's happening, what you really need is something like wireshark running on Linux, which can then capture wireless traffic in monitor mode. Microsoft Network Monitor might be able to do the same thing on Windows, but I've not tried it.
Wireshark (as does MS Network Monitor and it's excellent big brother MS Message Analyzer) only starts at later 2 by which time the problem has been and gone. I.e, frame / packet capturing I doubt will help (other than showing them not moving very fast!) I'm not aware of any protocols that cause an issue in one carrier frequency to affect another carrier frequency, but I suppose it could be written into the code itself independently of the protocol - though quite why one would do that, I know not!
@stewakeman wrote:
The marketing blurb is "It's faster, works with less interference" - but yours doesn't, so just use 2.4Ghz. I'm not buying that. At point blank range there is literally no usage scenario where 2.4Ghz should be so much faster than 5Ghz.
Laughable, isn't it? 5 Ghz and 2.4 Ghz are simply carriers - nothing more, nothing less. The way they are carved into discrete "channels" and then the aggregation of those channels is what gives bandwidth. (Kind of.) As it happens, the higher the frequency, the more room to move there is which results in greater channel bandwidth (something that is really defined by the 802.11 protocol in question more than the carrier frequency). But, 2.4 or 5 Ghz per se doesn't have a speed (well, it does: it's the speed of light in a vacuum, but you know what I mean)....it has features that facilitate protocols to support greater bandwidths. Maybe that's what the marketeers mean? 
One question though: how do you know that the 5 Ghz channels you are using are "uncontested". (I do believe you, just wondering how you checked?) Have you tried 56 instead? Apologies if you said you'd tried that already earlier in the post.
So to summarise,
- Likely that 5 Ghz interference has been ruled out. (Sure, not 100% conclusively because interference need not come from computers but other tests make that even less likely.) Validated by trying other 5 Ghz channels.
- Every device (Windows, android, whatever) using 5 Ghz maxes out on a speedtest at 30 Mbps.
- The same devices on the 2.4 Ghz carrier max out at the correct speed (70+).
- The same devices on the 5 Ghz carrier max out at the correct speed (70+) on a different router. (I think you said you tried that?)
The conclusion: the problem is isolated to the 5 Ghz carrier on this particular router. The next question that springs to mind: is it all routers of the same model and firmware?
I'd love to help you here but I only get about 17 Mpbs download myself, so I can't check. I did have a HH5 A when in BT Infinity 2 and I didn't have that problem myself. If you are feeling adventurous, you could try a local network speed test, maybe such as a WIndows port of TTCP which does a speed test between a source and target servers. You would make the source a copper-connected box and the destination your poor-performing laptop. I used to use that to test the bandwidth of aggregated ports on Cisco 7000s. 
It'll rule out the DSL side of things.
If I ever get my speed beyond 17 Mbps (it should be 80), I'll run a test at my end.
So I bit the bullet and purchased a Netgear D6220 which actually only has 900Mb AC WiFi, but I figured that far exceeds my usage requirements. Working for 48hrs no issues yet. Speed test from my Nexus 9 tablet in the bedroom:
I seem to be getting slightly better overall sync speeds compared to Hub One but it's marginal, it's definitely more consistent though and when I did an in front of router test I got almost 75 down and 20 up. I'm crossing my fingers that it stays like this and it'll have been money well spent.
So, just as a further update. A week on, we're still nicely connected, no drop outs or anything. Got this:
It's a pity that moving away from supplied equipment can result in such a good experience, but I'd have done it sooner had I known how much more consistent a slightly higher modem router would be.
