Should I replace the BT HH3?

Associate
Joined
3 Aug 2004
Posts
252
Hi,

I've just had Infinity 2 installed, (4 days ago) so it's still within the 10 day training period. I have consistently been getting 76Mbps down and around 15Mbps up using Speedtest, but when I tried last night, it's dropped to 55Mbps.

Don't get me wrong, these speeds are incredible and a lot better than some of the congested speeds I used to get with ADSL2, so it's not a speed complaint, but I've read that a lot of people change the BT supplied HomeHub 3 to a different router because of speed drops after the initial installation and it seems to help get them back to their original speeds.

After a bit of research, I'm seriously considering getting the Asus RT-AC66U because of the wireless range and having USB ports that I connect a shared printer or external drive to.

Up until the Infinity installation, I was using a Netgear DGND3300 dual-band ADSL2 router. I've currently got the HH3 connected via its GB port to a GB switch and that goes off to 2 PCs and a number of Devolo 200Mbps powerline adapters, (including the one with built in wireless to extend the range). These adapters then supply another 2 PCs, games consoles, media players etc. around the house, (not simultaneously...obviously;))

The wireless is used for iPads to connect.

Based on everything, would the RT-AC66U be a good choice or should I just stick with the HH3? I changed the original Thomson TG585v7 that was supplied by the ADSL2 ISP to the Netgear to help with lockups, so I'm quite happy to change the HH3 if it'll help with running everything.

Any advice would be greatly appreciated.

Thanks! :)
 
I've read that a lot of people change the BT supplied HomeHub 3 to a different router because of speed drops after the initial installation and it seems to help get them back to their original speeds.

This isn't the reason. It's usually because people want gigabit LAN, better WiFi, multiple SSIDs, more firmware options, etc.
 
Speeds drop a lot due to bandwidth management to be honest.

It won't be the HH3 unless you have a ton of wifi devices (it struggles with wifi, we have a seperate router just for wifi).
 
The Asus is a great router - I'm using one but there are a few issues.

I found the out the box firmware didn't handle lots of connections too well and needed rebooting every other day if people were doing torrents, refreshing server lists in some games, etc. flashing to the latest official FW solves that but introduces some wireless issues.

Flashing it with the latest merlin FW makes it a very stable, high performance router but there are still some niggles with wireless on almost every version of the FW (the odd device will see degraded wireless connection quality over time until BOTH the device and router are rebooted).

TBH tho I wonder if BT put connections on a "high priority" setting or something when its first installed then drop it back into the "normal" pool a few days later - pretty much every time I've had a connection done the engineers showed me it running at the full 8Mbit, 20Mbit, etc. when my FTTC was done it was holding 76MBit/s solid speedtesting while he was there (which was during peak time) - then a few days later its bouncing around 50-60Mbit during peak times and rarely maxes out out side peak times any more.

EDIT: Also a lot of speed testing sites don't give the most accurate speeds on connections faster than 20-30Mbit anyhow.
 
Last edited:
Thanks for all of your replies.

This isn't the reason. It's usually because people want gigabit LAN, better WiFi, multiple SSIDs, more firmware options, etc.

That makes sense. I didn't think about the need for extra GB ports because my HH3's GB port is connected to a GB switch. Also, better WiFi is one of the reasons I'm looking at a different router too.

Speeds drop a lot due to bandwidth management to be honest.

It won't be the HH3 unless you have a ton of wifi devices (it struggles with wifi, we have a seperate router just for wifi).

Don't talk to me about wireless! :p I wasted 2 hours last night trying to get iPads connecting to it without success. We've got an Android phone and an iPhone5 connected without any problems, but the two iPads consistently say 'Unable to join network'. One is running IOS5 and the other IOS6, (like the iPhone5), so it's not that and I tried a whole load of suggestions including adding the wireless network manually and using different levels of encryption. The only way I could get them both to connect was to turn off encryption altogether. As soon as I tried WPA, WPA2 etc. it wouldn't connect. Another reason to change the router I suppose. ;)

The Asus is a great router - I'm using one but there are a few issues.

I found the out the box firmware didn't handle lots of connections too well and needed rebooting every other day if people were doing torrents, refreshing server lists in some games, etc. flashing to the latest official FW solves that but introduces some wireless issues.

EDIT: Also a lot of speed testing sites don't give the most accurate speeds on connections faster than 20-30Mbit anyhow.

My speed was back up again last night, so either it was down to the training period or somebody was using the hub through BT Fon.

Thanks for the info, I don't really want a router that needs rebooting that frequently because of lock-ups. I used to have a Draytek 2800G and then the Netgear DGND3300 and they were 100% stable and never needed rebooting, (for me at least).

I won't need it for torrents, it'll mainly be used for general and Steam downloads, online gaming and TV on demand through Sky and Youview. Everything is cabled either directly or through powerline adapters, it's only the iPads and phones that use wireless and I need a fairly good Wifi signal to get around the house. Do you think with that kind of setup / use that I'll have to keep rebooting the Asus, (if I went for the RT-AC66U), or should I consider a different make and model of router?

Thanks again to all of you for your help and advice! :)
 
You have FTTC, which means that varying speeds are pretty much down to two things (assuming your HH3 isn't faulty and you're testing via Gigabit Ethernet): 1) the actual line speed, which is negotiated between the modem (not the HH3) and the cabinet, and 2) contention with every other user on the Internet. So changing the HH3 isn't going to help with either of those.

I just had BT Infinity 3 installed a week ago (FTTP, so I have actual fibre coming into my house), and using speedtest.net I was getting over 300 Mbps consistently with the HH3. So I know it can definitely handle the speed. However, the wifi was rubbish, low signal in some parts of the house, regular drop outs* (where the device loses the wifi signal completely and eventually reconnects, was happening at least every couple of hours), and to top it all off, because it has BT's proprietary firmware, you can never be sure of what is open to the world and what is happening inside it (e.g. the BTWifi stuff).

So last night I received and installed an Asus RT-N66U, and so far I'm really happy with it. It's a very high-quality piece of kit, and the difference in the wifi quality is night and day. Super strong signal everywhere in the house, rock solid so far (not a single drop out), and 5 GHz wifi is oh so much faster. With the HH3 I was getting consistent 40 Mbps on my iPhone 5 right next to the router (using the Speedtest.net app), and with the Asus on 5 GHz wifi I get consistent 70 Mbps. I did some iperf tests from my 3+ year-old laptop (which has 5 GHz wifi but probably single-stream) and got over 200 Mbps next to the Asus router and just over 100 Mbps upstairs (more or less directly above the router). So with a modern multi-stream wifi card in the laptop, I'm sure I could get 400+ Mbps.

The one fly in the ointment is that speedtest.net through the Asus maxed out at around 200+ Mbps, although the ping dropped from average 9ms through the HH3 to 7ms through the Asus. So tonight I will be doing some A/B testing between the HH3 and the Asus to check whether that was just an anomaly last night, or whether the Asus is actually not handling the speed as well as the HH3. I will update the post once I've done that.
 
Speeds drop a lot due to bandwidth management to be honest.

What traffic management?

The only managed traffic is P2P.

The one fly in the ointment is that speedtest.net through the Asus maxed out at around 200+ Mbps, although the ping dropped from average 9ms through the HH3 to 7ms through the Asus. So tonight I will be doing some A/B testing between the HH3 and the Asus to check whether that was just an anomaly last night, or whether the Asus is actually not handling the speed as well as the HH3. I will update the post once I've done that.

Unlikely. http://www.smallnetbuilder.com/wire...n900-gigabit-router-reviewed?showall=&start=2
 
Oh, forgot to add that I went for the RT-N66U rather than the RT-AC66U because there's a big price difference, the firmware is reportedly quite buggy, and basically it is a draft standard that is still probably 2-3 years from actually showing up in client devices.

As an aside, last night I ran INSSIDER from my laptop, and the only 5 GHz network visible was mine, compared to a dozen or so 2.4 GHz networks. So if 5 GHz 802.11n isn't that popular given that it has been around for several years, then I predict that 802.11ac is unlikely to be widely available for at the very least 2 or 3 years. Might as well save the money and buy a new awesome router in 3 years or so.
 
All this to-and-fro about traffic management (not referring just to this post, it just seems to be one of the most common themes on this forum) is down to semantics.

Ultimately, the network does not have infinite capacity, and there are bottlenecks everywhere. "Traffic management", "traffic shaping" etc. aren't about artificially restricting someone just for the sake of it, just to pi$$ people off; it's about transmitting all of the traffic for all of the users in the fairest and most useful way to everyone (restricting individuals from engaging in certain behaviours for the benefit of the group). If a torrent download takes slightly longer, nobody really cares; but if a Skype call breaks up or web browsing is slow or YouTube videos require a lot of buffering... THEN a LOT of people care.

For some reason people assume that they are entitled to 100% of their line speed 100% of the time from 100% of the Internet; in the real world however, resources are always limited (even water and air, it turns out). And it just happens that ISPs know this, and have put in place certain technologies to even things out a bit.

Of course, everything within reason. Some ISPs just don't want to spend money upgrading their network to what it should be based on the number of users they have, and are throttling everything. But then the discussion should be about under-investment in the network, and ultimately ripping off their customers, not specifically about "traffic management".
 
So last night I received and installed an Asus RT-N66U, and so far I'm really happy with it. It's a very high-quality piece of kit, and the difference in the wifi quality is night and day. Super strong signal everywhere in the house, rock solid so far (not a single drop out), and 5 GHz wifi is oh so much faster.

That sounds ideal, it'll be nice to have a decent wifi signal back.

Thanks very much for all of your suggestions and help. The result....

Ordered! :D

In the end, I did go the AC66U. Probably a little bit stupid for the price considering the lack of AC devices and the fact that it's draft and may change, but if it doesn't, I'll have a bit of future-proofing.

I very nearly went for the N66U, (and considered the N65U because of the USB 3.0 ports on it), but after reading quite a lot of favourable reviews of the AC66U and that the buggy firmware out of the box has generally been fixed by a better official firmware, I thought I'd go for it. :)
 

Well, I did my A/B testing this morning at 7:00, and unfortunately, the Asus RT-N66U maxed out at 210 Mbps whereas the HH3 got to 305 Mbps! This is not what I expected at all.

These numbers are from visiting speedtest.net on a desktop Core i7 875K with 8 GB RAM, an SSD, and connected to each of the routers via Gigabit Ethernet (through a 3Com switch). Before testing, I also made sure there was nothing on my network that was either downloading or uploading (thus tying up part of the bandwidth). The results are consistent on both routers. I tested within 5 minutes (more or less) of each other.

The only slight difference in configuration is that the HH3 is configured with the default BT PPoE username ([email protected]), whereas the RT-N66U refuses to connect with that username (and no password), so it is configured with my @btinernet.com username and password.

Given there are so few people with FTTP, and even fewer with an RT-N66U, I don't fancy my chances of getting to the bottom of this. But if anyone has any ideas, they'd be welcome.
 
The only slight difference in configuration is that the HH3 is configured with the default BT PPoE username ([email protected]), whereas the RT-N66U refuses to connect with that username (and no password), so it is configured with my @btinernet.com username and password.

Given there are so few people with FTTP, and even fewer with an RT-N66U, I don't fancy my chances of getting to the bottom of this. But if anyone has any ideas, they'd be welcome.

I can't remember where I saw this because I've been reading so much about the Asus routers and the HH3 over the last couple of days, but apparantly, the way around the router not accepting a blank password along with the '[email protected]' username is to put a space in. It seems that most routers including the Asus will accept this and the ones that don't seem to work with just 'BT'. As that login doesn't need a password, I think it's just the router that's expecting one.

It might be worth trying the proper username along with a space for the password to see if you can get it to connect....
 
Thanks for that! Coincidentally I just came across that five minutes ago, so I'm definitely trying that tonight. Suggestions I found are to either try a single space or "password" (which as you said probably implies anything will work, just not completely blank). The funny thing is that the RT-N66U accepts a blank password, it just won't connect.

My other thing to check is MTU size on the HH3, to see whether it is different from 1492.
 
The tip about using "password" for the [email protected] username works. However, there is no speed difference. The HH3 still gives me ~310 Mbps, while the RT-N66U only gives me ~210 Mbps.

I tested plugging my 3-year-old Core 2 Duo laptop directly into the modem and configured a PPPoE connection in Windows 8, and it connects fine, but I only achieve ~60 Mbps while the laptop's CPU rockets to close to 100%. So it appears that the PPPoE connection is very CPU intensive, and perhaps that is what the Asus isn't handling very well. I suspect the HH3 has a flag on its PPPoE connection that disables something (perhaps encryption) that makes it a lot less CPU intensive, allowing it to achieve the full 300 Mbps. Now I just need to know what PPPoE tweak is needed on the RT-N66U (there is a field for "additional parameters).
 
Back
Top Bottom