I have both the R7000 and AC68U... which to keep

[PrivateJoker]

Holy sh@@@@!!!

All my traffic is going thru VPN with a gateway in another country, so ANY ip location service I used before, showed me as I am there (where VPN server is).

However, when I tried that HTML5 method, Firefox did ask me I want to share my location, I clicked yes - and that thing showed my actual location as in street address!!!!!! The laptop does not have GPS sensor, so how on earth it detected my precise location?

I tested my VPN for DNS leaks, nothing came up suspicious... Any ideas?

Sorry for not posting the link on the HTML5 geolocation spec, but in order it will query: IP, wifi AP MAC to look in MAC databases of tagged AP's (android reports this info to google constantly, ios reports it to apple constantly, if your street is on google streetview all AP MACs have been hard tagged with lat/long), and only then after that it will query your GPS (if any).

Adding "_nomap" to your SSIDs will prevent Google streetview cars (theoretically) from tagging them, I'm not sure if it's a universally accepted instruction. My APs have been tagged "_nomap" for years, and my home is blurred out on google streetview, but HTML5 can still pinpoint me right on the money. Certain websites will also do things like spook me with HTML5 ads that say things like "meet hot cougars in east Dallas!"

The one thing I haven't intentionally tried is on a PC hardwired to my router if HTML5 will peak at my wifi even if I'm not using it.

The AP MAC tagging does feel slightly invasive, but if the GPS in our mobile devices did a cold boot every time it needed our location it would take 60-90sec +. And I think cellular assist isn't anywhere near the accuracy of AP MAC scanning. In theory, in many areas one could practically get turn by turn directions with little to no GPS fix. It also allows for GPS type capabilities indoors, etc.

 

[Sky1111]

I did some more testing and this is what I found out:

- Hardwired clients - confirmed on 2 different machines - are pinpointed to the remote location (the one I expected to be)

- tried on a different laptop (same wifi but I think 5GHz band) - and again it shows remote location

Puzzled I am... the first laptop which pinpointed to home address does not not cell modem or GPS...

 

[hungarianhc]

Hi All,

I've been playing with both of these routers for a few days now, and I thought I'd just report back con a few more notes...

- It still surprises me just how freaking huge the nighthawk is. I think it's unreasonably large...

- I have noticed the 5ghz signal drop out from time to time on the AC68U. I have done some searching on a few other forums, and it appears that others have experienced this too. There was an Amazon review that mentioned this as well.

- Ping times seem to be ever so slightly lower with the R7000. On 802.11N, right next to the router, the closest I could ever get the AC68U down to was 17ms. I can consistently get 12ms with the R7000.

- Signal strength seems somewhat comparable, but I want to give them both time before I make that final judgement.

- I haven't done any USB testing. I probably won't.

I just swapped out the AC68U to the Nighthawk as my primary router. Going to stick with that for the next few days, and I'll see how things go. Assuming my wife doesn't complain about the 5ghz signal (her macbook air is 802.11ac - today right next to the router it was showing only 7mbps connection w/ the AC68U until I restarted the router) on the Nighthawk, I will probably stick with that one. Time will tell though.

As a real fix? By telling your ISP what your IP is, and telling them to get their ARIN database record updated. They're the only ones that can really fix it.

If for some reason this is really bothering you, you could always have your router clone the MAC of the other router, so it will inherit the same IP.

Thanks for this tip! Yes I'm weird - yes it bothered me. Yes I am now using my AC68U mac address w/ my R7000, and as you predicted, speedtest.net now puts me in the right place .

 

[jlake]

Hi All,

I've been playing with both of these routers for a few days now, and I thought I'd just report back con a few more notes...

- It still surprises me just how freaking huge the nighthawk is. I think it's unreasonably large...

- I have noticed the 5ghz signal drop out from time to time on the AC68U. I have done some searching on a few other forums, and it appears that others have experienced this too. There was an Amazon review that mentioned this as well.

- Ping times seem to be ever so slightly lower with the R7000. On 802.11N, right next to the router, the closest I could ever get the AC68U down to was 17ms. I can consistently get 12ms with the R7000.

- Signal strength seems somewhat comparable, but I want to give them both time before I make that final judgement.

- I haven't done any USB testing. I probably won't.

I just swapped out the AC68U to the Nighthawk as my primary router. Going to stick with that for the next few days, and I'll see how things go. Assuming my wife doesn't complain about the 5ghz signal (her macbook air is 802.11ac - today right next to the router it was showing only 7mbps connection w/ the AC68U until I restarted the router) on the Nighthawk, I will probably stick with that one. Time will tell though.



Thanks for this tip! Yes I'm weird - yes it bothered me. Yes I am now using my AC68U mac address w/ my R7000, and as you predicted, speedtest.net now puts me in the right place .

What was the make/model of your router before you bought the AC68U/r7000?

Are you using the same SSID for your tests? You should really remove your wireless profiles when you change routers. Or better yet, you should give the new router(s) a completely new and unique/different SSID (something that you've never used before) when testing. RMerlin has often posted that giving the router a new SSID is actually better than removing/deleting old wireless profiles, and I can confirm that to be true from personal experience. Sometimes removing a previous wireless profile just won't cut the mustard as some low level settings are not deleted.

I'm not saying that is the cause of your 7Mbps link speed on mac, but it is very possible.

 

[RMerlin]

However, when I tried that HTML5 method, Firefox did ask me I want to share my location, I clicked yes - and that thing showed my actual location as in street address!!!!!! The laptop does not have GPS sensor, so how on earth it detected my precise location?

I tested my VPN for DNS leaks, nothing came up suspicious... Any ideas?

If the device has a wireless card, then it used the list of all visible access points to track down your location, by comparing with what Google had logged in their database. Google does the same thing under Android.

 

[hungarianhc]

What was the make/model of your router before you bought the AC68U/r7000?

Are you using the same SSID for your tests? You should really remove your wireless profiles when you change routers. Or better yet, you should give the new router(s) a completely new and unique/different SSID (something that you've never used before) when testing. RMerlin has often posted that giving the router a new SSID is actually better than removing/deleting old wireless profiles, and I can confirm that to be true from personal experience. Sometimes removing a previous wireless profile just won't cut the mustard as some low level settings are not deleted.

I'm not saying that is the cause of your 7Mbps link speed on mac, but it is very possible.

completely different SSID on all 3 - my previous router was an Amped Wireless R20000G. Now I have the AC68U and the R7000. All 3 have had different SSIDs.

 

[twitchyzero]

What are you doing that requires a faster CPU? Cause the CPU speed will have zero impact on wired/wireless performance. It will have an impact om USB-based disk sharing and VPN mostly.

faster CPU doesn't improve # of simultaneous connections? That's my biggest requirement is to handle loads of heavy traffic. Still eagerly waiting for the full, detailed review of each router from Tim. I recall reading R7K has the faster memory too...dunno if that has any advantages.

OP i'm in a similar position and it's hard to tell which one is ultimately better without more folks playing with em.

From reading around Asus has a better track record for FW updates...how's the CFW scene for Netgear products? I'm guessing ASUS will usually have a wider array of CFW support.

If ASUS can prove matched USB3 performance after their upcoming FW then I may just side with the ASUS to caution on the side of safety.

 

[whsbuss]

faster CPU doesn't improve # of simultaneous connections? That's my biggest requirement is to handle loads of heavy traffic. Still eagerly waiting for the full, detailed review of each router from Tim. I recall reading R7K has the faster memory too...dunno if that has any advantages.

OP i'm in a similar position and it's hard to tell which one is ultimately better without more folks playing with em.

From reading around Asus has a better track record for FW updates...how's the CFW scene for Netgear products? I'm guessing ASUS will usually have a wider array of CFW support.

If ASUS can prove matched USB3 performance after their upcoming FW then I may just side with the ASUS to caution on the side of safety.

Yep me too. I know Asus pulled their f/w to tweak some of the issues reported.

 

[thiggins]

No, faster CPU does not = more connections.

Is a vendor really better when they ship products that NEED multiple firmware spins to fix problems that shouldn't have been there in the first place?

 

[Clive.B]

But wouldn't a faster cpu in a router be able to push more bandwidth done to each ethernet port ?

Also a silly question does these new routers broadcom cpus support AES instructions ? This would help with openvpn overhead so ive heard....

 

[RMerlin]

But wouldn't a faster cpu in a router be able to push more bandwidth done to each ethernet port ?

Also a silly question does these new routers broadcom cpus support AES instructions ? This would help with openvpn overhead so ive heard....

LAN to LAN transfer are entirely handled by the switch, with no CPU activity being required. The only place the CPU will help is in WAN to LAN. And if you don't use any of the features that forces Broadcom's HW acceleration to be disabled, then you would have no problem hitting well over 700 Mbits that way.

I don't think the BCM4708 has hardware-based AES support (at least not through an open API), based on the CPU features that are enumerated:

Processor	: ARMv7 Processor rev 0 (v7l)
processor	: 0
BogoMIPS	: 1599.07

processor	: 1
BogoMIPS	: 1599.07

Features	: swp half thumb fastmult edsp 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x3
CPU part	: 0xc09
CPU revision	: 0

Hardware	: Northstar Prototype
Revision	: 0000
Serial		: 0000000000000000

 

[thiggins]

If you are referring to Local (non Internet) traffic, that does not depend on CPU, just the switch.