AC1900 First Look: NETGEAR R7000 & ASUS RT-AC68U

[BatJoe]

Netgear R7000 doesn't seem to be a choice for me: I've been down the path of using a superior hardware with pathetic software support: I feel one is better off with even a slightly inferior hardware that has better support. An example: both companies give personal cloud feature: Netgear's ReadyShare app that costs couple of bucks hasn't been updated in play store for more than 2 years, with 2 star average rating, while Asus aiCloud *free* app is regularly updated with 4 star average rating. Similarly, my 2 year old ageing Asus router is still getting new firmwares with new features every now and then, Netgear isn't known to be that proficient here as well.

Asus has to get its act together to pull through this one!

Normally, I would agree and say ASUS would be the way to go, but lately their router support has been ridiculous.

They may release more updates, but they seem to break more things than they fix or they break things that were fixed before! And it is taking them a long time to get these fixes sorted and released to the public.

I'd rather have less updates to do and have a product that just works well. I will be replacing my RT-N66U with the Netgear R7000 and am hoping for smoother waters. I won't be buying another ASUS router for a while. Luckily their motherboard support has remained the best in the industry and I will continue to use them there.

 

[Razor512]

You overestimate R7000 in regards to OpenVPN client... we are talking about real world test with VPN provider like Private Internet Access or alike...

Not sure how I am overestimating it, I simply set up the openVPN server, then connected to ti and ran a throughput test.

While it does not support android clients due to lack of things like OSI layer 3 and other features, when the client supports the server, it does what a VPN should do, and it does it properly and at a speed that is higher than most internet connections.

My testing involved 1 computer on the LAN side of the R7000, and 1 on the WAN side The computer on the WAN side then connects as a client to the VPN server in the R7000 thus allowing it to access the the system on the R7000's LAN

In that setup the speeds will hover around 39-41mbit/s. So thats pretty much the throughput limit when traffic is going through the VPN server

it is not really over estimating, it is just the throughput I was able to measure.

If used as a way to securely access the internet, eg securing yourself while on a public hotspot, then your overall throughput will be limited to around 20mbit/s since you are making the VPN server receive at equal rates.

 

[Sky1111]

Not sure how I am overestimating it, I simply set up the openVPN server, then connected to ti and ran a throughput test.

While it does not support android clients due to lack of things like OSI layer 3 and other features, when the client supports the server, it does what a VPN should do, and it does it properly and at a speed that is higher than most internet connections.

My testing involved 1 computer on the LAN side of the R7000, and 1 on the WAN side The computer on the WAN side then connects as a client to the VPN server in the R7000 thus allowing it to access the the system on the R7000's LAN

In that setup the speeds will hover around 39-41mbit/s. So thats pretty much the throughput limit when traffic is going through the VPN server

it is not really over estimating, it is just the throughput I was able to measure.

If used as a way to securely access the internet, eg securing yourself while on a public hotspot, then your overall throughput will be limited to around 20mbit/s since you are making the VPN server receive at equal rates.

My apologies - we are talking about 2 different things. There is VPN Server and there is VPN Client. You might be getting a great performance on one, but that does not mean automatically that Client works great too - and for one, VPN Client functionality is not even available in stock firmware..

 

[RogerSC]

The CNET review for the R7000 mentions that both routers, R7000 and RT-AC68U use the same Broadcom chipset (they mention the Broadcom BCM4709) which is not correct and casts some doubt on their reviews of high-end routers for me. If CNET had read Mr. Easy's review, they'd know better *smile*. I think that there's a small edge to the R7000 in the comparison with the RT-AC68U on the hardware side, but better firmware (on either side *smile*) will determine who ends up on top.

I've gotten an R7000 now, and am not impressed with the firmware. On the other hand, it is fast, has great range (slightly better than the RT-N66U at my place *smile*). Seems like a nice piece of hardware waiting for better firmware. Hoping that things will improve with future releases. Don't know yet if I'm going to keep it. Had one weird thing happen already with the firmware that was disturbing, but could be an early firmware burp.

If Asus can pull it together on the firmware side, they will do quite well competitively against a slightly more powerful competitor. RMerlin can help them there.

 

[GypsyDRB]

I've had teh R7000 for 2-3 days now. I'm getting fairly frequent disconnects on my connection. Router pops back up in under a minute.

I'm a TWC subscriber with a surfboard modem in front of the router (MAC address hardcoded into the router). Behind the router is an 8 port switch which is the only connection to the router (all of my physical connections go through the switch to the router). I have 14+ wireless connections that are a mix of 2.4 and 5. Params on the router I have changed are 20/40 co-existence, gone to short preamble, and I enabled traffic meter (which i have since turned off as of this post)...and I'm pretty sure that's it.

I am on the latest firmware (V1.0.1.22_1.0.15)

Thanks for any help.

 

[stevech]

Are the connectivity problems WiFi - only, or both wired/WiFi?
Disable 40MHz mode for now.
Set preamble length to factory default.
Maybe reset router to defaults, enter just WiFi SSID/passwords. Disable 5GHz for now.
I use TWC and do not need to enter a spoof-MAC into the router to use it with the TWC modem.

 

[twitchyzero]

Tim when can we expect a full-fledged review for each router?

 

[Clive.B]

My apologies - we are talking about 2 different things. There is VPN Server and there is VPN Client. You might be getting a great performance on one, but that does not mean automatically that Client works great too - and for one, VPN Client functionality is not even available in stock firmware..

Was about to say the same thing, the OpenVPN client is what I would love to have tested.

As I think before mentioned no direct openvpn client support on the R7000s yet ? I think only 3rd party has it and its buggy....

Also does anyone know if the R7000 has 2x 1ghz cpus or 2x500mhz cpus ?

I do not believe 1ghz alone would solve a bottle neck of OpenVPN client performance under a decent vpn provider but perhaps am wrong.

 

[PrivateJoker]

Also does anyone know if the R7000 has 2x 1ghz cpus or 2x500mhz cpus ?

I do not believe 1ghz alone would solve a bottle neck of OpenVPN client performance under a decent vpn provider but perhaps am wrong.

BCM4709 is a 1ghz dual core SoC (2 x CPUs ≠ 2 x cores, to be geeky about it. )

http://www.broadcom.com/products/Wireless-LAN/802.11-Wireless-LAN-Solutions/BCM4707-4708-4709

 

[GypsyDRB]

Are the connectivity problems WiFi - only, or both wired/WiFi?
Disable 40MHz mode for now.
Set preamble length to factory default.
Maybe reset router to defaults, enter just WiFi SSID/passwords. Disable 5GHz for now.
I use TWC and do not need to enter a spoof-MAC into the router to use it with the TWC modem.

Thanks for the response.

Both wired and wifi.

I seem to recall at initial setup 3 options for preamble length (default setting?)...but maybe I'm wrong. At any rate I only have two options for each band long or short so I changed them both to long.

Ref the coexistence setting...I had it unchecked so no coexistence was happening, but have gone back and checked the box...should I leave that one unchecked?

on the 5ghz I changed from 153 to auto and dropped the speed from 1300 to 600 (don't have the clients). Will monitor further. In the morning I'll try disabling 5ghz and also see if I can get a connection without spoofing the previous netgear router MAC.

****edit****

definitely have an issue with the MAC address. If I go with anything other than what my "old" centria's MAC address is, the router won't connect to the modem and provide internet access. Checked the surfboard logs and saw this:

Oct 17 2013 05:59:08 6-Notice I401.0 TLV-11 - unrecognized OID;CM-MAC=7c:bf:b1:d4:05:6f;CMTS-MAC=00:01:5c:33:93:96;CM-QOS=1.1;CM-VER=3.0;

 

[thiggins]

Tim when can we expect a full-fledged review for each router?

First priority is the EA6900 review. What information are you looking for? I covered the key points for the ASUS and NETGEAR in the first look.

Only thing not covered was TurboQAM performance. I would prefer to ru n a full throughput vs. attenuation profile for at least one of the routers. But ASUS can't tell me when they will have a PCE-AC68 driver that has TurboQAM enabled.

 

[RogerSC]

First priority is the EA6900 review. What information are you looking for? I covered the key points for the ASUS and NETGEAR in the first look.

Only thing not covered was TurboQAM performance. I would prefer to ru n a full throughput vs. attenuation profile for at least one of the routers. But ASUS can't tell me when they will have a PCE-AC68 driver that has TurboQAM enabled.

I have to say that I thought that your dual review of the Asus and Netgear was very well done, in enough depth. I'm not expecting the TurboQAM to be useful for me, in any case, so that isn't something that I'm waiting for.

And I really appreciated that you put them head to head with each other, and compared them on each feature. It would be nice to add a review for the Linksys EA6900, but you can't do that until you have one *smile*.

Anyways, thanks very much for the AC1900 router reviews, they are VERY useful and timely.

 

[blowers]

Hell thiggins,

Just a quick question.

When you tested the range and throughput of the ASUS RT-AC68U router, what power transmission rate did you set in the "Wireless > Professional" settings.

 

[BatJoe]

Hell thiggins,

Just a quick question.

When you tested the range and throughput of the ASUS RT-AC68U router, what power transmission rate did you set in the "Wireless > Professional" settings.

I assume he would have used the default which ASUS usually set to 80.

I found changing this didn't really make a difference in range.

 

[blowers]

It would be interesting to know if Netgear allow the transmission power to be manually adjusted like the Asus does.

I know that Netgear have been complaining about Asus allowing the power to be ramped up beyond what is considered to be neighbour-friendly levels.

I need greater range to reach an apartment 3 levels above mine. I already have a Asus RT-A66U that does the job reasonable well, even better than my last router, a Netgear WNDR4500. The tenant claims that they are getting a faster connection throughput from the Asus, but if I could get even better range and throughput from one of the two compared R7000 and RT-A68U, I would buy one.

 

[GypsyDRB]

Thanks for the response.

Both wired and wifi.

I seem to recall at initial setup 3 options for preamble length (default setting?)...but maybe I'm wrong. At any rate I only have two options for each band long or short so I changed them both to long.

Ref the coexistence setting...I had it unchecked so no coexistence was happening, but have gone back and checked the box...should I leave that one unchecked?

on the 5ghz I changed from 153 to auto and dropped the speed from 1300 to 600 (don't have the clients). Will monitor further. In the morning I'll try disabling 5ghz and also see if I can get a connection without spoofing the previous netgear router MAC.

****edit****

definitely have an issue with the MAC address. If I go with anything other than what my "old" centria's MAC address is, the router won't connect to the modem and provide internet access. Checked the surfboard logs and saw this:

Oct 17 2013 05:59:08 6-Notice I401.0 TLV-11 - unrecognized OID;CM-MAC=7c:bf:b1:d4:05:6f;CMTS-MAC=00:01:5c:33:93:96;CM-QOS=1.1;CM-VER=3.0;

also did the reset to defaults last night. firmware stayed current though. originally I had set up my SSIDs on the factory firmware out of the box and then upgraded. It seemed like things were stable until this morning when I realized my USB3.0 wasn't recognized (unplugged and plugged back in). I've already had one connection drop again (wired).

Will continue to monitor and try the other steps (i.e. shutting down 5GHZ band) depending on how the day goes.

 

[BatJoe]

It would be interesting to know if Netgear allow the transmission power to be manually adjusted like the Asus does.

I know that Netgear have been complaining about Asus allowing the power to be ramped up beyond what is considered to be neighbour-friendly levels.

I need greater range to reach an apartment 3 levels above mine. I already have a Asus RT-A66U that does the job reasonable well, even better than my last router, a Netgear WNDR4500. The tenant claims that they are getting a faster connection throughput from the Asus, but if I could get even better range and throughput from one of the two compared R7000 and RT-A68U, I would buy one.

On my Netgear R7000 there is no option to let you change the transmission value.

You may want to invest in an extender - these routers can only do so much.

 

[RogerSC]

It would be interesting to know if Netgear allow the transmission power to be manually adjusted like the Asus does.

I know that Netgear have been complaining about Asus allowing the power to be ramped up beyond what is considered to be neighbour-friendly levels.

I need greater range to reach an apartment 3 levels above mine. I already have a Asus RT-A66U that does the job reasonable well, even better than my last router, a Netgear WNDR4500. The tenant claims that they are getting a faster connection throughput from the Asus, but if I could get even better range and throughput from one of the two compared R7000 and RT-A68U, I would buy one.

First off, you're not going to get significantly more wireless range with an R7000 or RT-AC68U. You might get a little, I see a little more range at my place with the R7000 than the RT-N66U, but not enough to make the kind of difference you're looking for.

I think that the next step up for an apartment 3 levels up is either an Access Point (AP) in the apartment, or possibly a "network extender" in that apartment in a place where the signal from your place is strongest. The problem with a network extender is that it halves the speed, but does provide more signal in a weak signal area.

If you can get to that apartment either via powerline networking or MoCA, that would give you the ability to put in an AP. This would give you more speed than a network extender, assuming that it would work. Don't know how the power or cable TV is wired in your building, but this doesn't seem likely. Running an ethernet cable between your place and the apartment seems more likely to work for you. Then you could use an AP.

One note, having a more powerful router (or turning up the power on a router) isn't the full answer, wireless is two-way. The signals from the wireless clients also have to be strong enough to reach your router...that's why I suggested an AP or network extender, both provide that two-way connection. Also, turning up the power in a router can cause distortion problems, and cause interference for your neighbors, and doesn't answer the problem with hearing the wireless clients.

 

[thiggins]

Hell thiggins,

Just a quick question.

When you tested the range and throughput of the ASUS RT-AC68U router, what power transmission rate did you set in the "Wireless > Professional" settings.

The default. When testing I change only the SSID, Channels and bandwidth settings as stated in the review.

 

[blowers]

Have you considered rerunning the tests with the transmission power set at 200mW on the RT-AC68U in order to ascertain whether or not it makes any difference to the results of the various tests ?