Very confused-- please help

[djayjp]

I've literally spent *hours* carefully combing through the data on SNB. I have now cancelled two separate router orders (which is quite unlike me) upon coming across new information. The more data I came across today, the more confused I became, as many articles (and recommendations by its author) seem to contradict each other, to the point that it has me questioning the legitimacy of such. All I really want is a long range 5Ghz downlink oriented router (with at least 20Mbps at, preferably, location F)-- any other considerations (such as "class") are purely secondary. I am not an expert.

At first, I thought the Archer C7 (AC1750) would be the best as the router charts show it providing the highest throughput at location F of any router in the charts at ~115Mbps (using the method just prior to the current test method). It supposedly roughly equals the RT-AC66U in this respect:

http://www.smallnetbuilder.com/wire...ess-dual-band-gigabit-router-reviewed?start=3


Then, however, I came across a post by the site's author, who linked to a particular comparison chart (such comparisons are seemingly unavailable to users). It recommends going with the RT-N66U, as it bests its aforementioned AC cousin with 85Mbps at 45dB of attenuation vs. 0Mbps at 39dB (however, in the RT-N66U review, it is explicitly stated that the router only operates at less than 2Mbps, yes 2Mbps, at location F):

http://www.smallnetbuilder.com/tool...rk-knight/1140-asus-rtac66u/1226-asus-rtac56u


So, I figured the RT-N66U is the best of all the routers mentioned for 5Ghz downlink range. Then, I came across this post which, again, seems to contradict all the above data, showing up to a ~400% improvement by a number of AC1900 class routers compared to the RT-N66U at location D:

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance


So, which is correct, the router charts, the forum post by thiggins (the second link), or the results of the last test which pitches the, presumably, best N class router against newer AC ones...?

What is the best router for this purpose?


*Edit: Then there's this recent comparison:

http://www.smallnetbuilder.com/wire...32896-will-a-new-ac-router-bring-you-wifi-joy

Soooo confused....

*Edit2: Then I also came across this, which states that the RT-AC66U produces 52Mbps at location F!!!:

http://www.smallnetbuilder.com/wire...1750-gigabit-router-reviewed?showall=&start=3

 

[TonyH]

I've literally spent *hours* carefully combing through the data on SNB. I have now cancelled two separate router orders (which is quite unlike me) upon coming across new information. The more data I came across today, the more confused I became, as many articles (and recommendations by its author) seem to contradict each other, to the point that it has me questioning the legitimacy of such. All I really want is a long range 5Ghz downlink oriented router (with at least 20Mbps at, preferably, location F)-- any other considerations (such as "class") are purely secondary. I am not an expert.

At first, I thought the Archer C7 (AC1750) would be the best as the router charts show it providing the highest throughput at location F of any router in the charts at ~115Mbps (using the method just prior to the current test method). It supposedly roughly equals the RT-AC66U in this respect:

http://www.smallnetbuilder.com/wire...ess-dual-band-gigabit-router-reviewed?start=3


Then, however, I came across a post by the site's author, who linked to a particular comparison chart (such comparisons are seemingly unavailable to users). It recommends going with the RT-N66U, as it bests its aforementioned AC cousin with 85Mbps at 45dB of attenuation vs. 0Mbps at 39dB (however, in the RT-N66U review, it is explicitly stated that the router only operates at less than 2Mbps, yes 2Mbps, at location F):

http://www.smallnetbuilder.com/tool...rk-knight/1140-asus-rtac66u/1226-asus-rtac56u


So, I figured the RT-N66U is the best of all the routers mentioned for 5Ghz downlink range. Then, I came across this post which, again, seems to contradict all the above data, showing up to a ~400% improvement by a number of AC1900 class routers compared to the RT-N66U at location D:

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance


So, which is correct, the router charts, the forum post by thiggins (the second link), or the results of the last test which pitches the, presumably, best N class router against newer AC ones...?

What is the best router for this purpose?


*Edit: Then there's this recent comparison:

http://www.smallnetbuilder.com/wire...32896-will-a-new-ac-router-bring-you-wifi-joy

Soooo confused....

*Edit2: Then I also came across this, which states that the RT-AC66U produces 52Mbps at location F!!!:

http://www.smallnetbuilder.com/wire...1750-gigabit-router-reviewed?showall=&start=3

May I ask what router do you have now? What were two routers you cancelled after ordering? Router ranker is one factor for your decision making. Just a reference according to set test procedure. In real world every one's environment is different so other variables come into play too. If you want to increase the range, different router may be one option but also other method like using an AP could be the answer. Before anything else I 'd try and find a best location for your router in your place. Luckily one router does more than good enough job for me and my family in my house. There is no router which does every thing super well. According to your need, if range is most important for you go for that one with best range, etc.

 

[djayjp]

May I ask what router do you have now? What were two routers you cancelled after ordering? Router ranker is
one factor for your decision making. Just a reference according to set test procedure. In real world every one's
environment is different so other variables come into play too. If you want to increase the range, different router
may be one option but also other method like using an AP could be the answer. Before anything else I 'd try and
find a best location for your router in your place. Luckily one router does more than good enough job for me and
my family in my house. There is no router which does every thing super well. According to your need, if range is
most important for you go for that one with best range, etc.

We just have the default ISP provided modem/router (which is stuck at 2.4Ghz). The interference in this apartment building is horrible (as confirmed via wifi analyzer) and there are regular connection drops to a variety of devices. I need a rock solid connection for VOIP and video streaming purposes. I cancelled the Archer C7 and the RT-N66U.

I just want the best router as determined by objective testing. After checking the best routers listed in the 3rd link, I'm still very confused. I checked the reviews for those top routers listed in that comparison, and the results in those individual reviews do not at all seem to support the results of the comparison (e.g., zero throughput at higher attenuation).

Edit: What I'm looking for is fairly straightforward, yes? So, what does SNB say is the best?

 

[djayjp]

I decided to go with the Archer C9. If that doesn't work, then I'll try the RT-AC68U. They both seem great in the range department (including with N clients).

I think I understand the various discrepancies. Between testing methods changing and a number of re-tests (involving vertical vs. horizontal orientation), I suppose it all makes sense in the end. Though the only thing I really still don't understand is how in the review of the RT-N66U it only gets about 2Mbps at location F, whereas in the charts it shows it getting 82.5Mbps....

 

[TonyH]

We just have the default ISP provided modem/router (which is stuck at 2.4Ghz). The interference in this apartment building is horrible (as confirmed via wifi analyzer) and there are regular connection drops to a variety of devices. I need a rock solid connection for VOIP and video streaming purposes. I cancelled the Archer C7 and the RT-N66U.

I just want the best router as determined by objective testing. After checking the best routers listed in the 3rd link, I'm still very confused. I checked the reviews for those top routers listed in that comparison, and the results in those individual reviews do not at all seem to support the results of the comparison (e.g., zero throughput at higher attenuation).

Edit: What I'm looking for is fairly straightforward, yes? So, what does SNB say is the best?

Another thing is your budget.... IMO. returning those two routers was alright. Else where in the forum you noticed newer -AC routers gives good result as a side enffect having newer more powerful hardware overall. If budget is
not a concern, buy newer ones(over buy for yourself). Do you want to put the modem router into bridge mode or use new router in AP mode with the modem router. Instead of VOIP, not interested in Skype? I chat with my grand son living in next province using skype(audio/video). Works 100% always Wednesday morning every week. Hope others will chime in for you.

 

[djayjp]

Another thing is your budget.... IMO. returning those two routers was alright. Else where in the forum you noticed newer -AC routers gives good result as a side enffect having newer more powerful hardware overall. If budget is
not a concern, buy newer ones(over buy for yourself). Do you want to put the modem router into bridge mode or use new router in AP mode with the modem router. Instead of VOIP, not interested in Skype? I chat with my grand son living in next province using skype(audio/video). Works 100% always Wednesday morning every week. Hope others will chime in for you.

I'm going to use it in bridge mode (if that is what I think it is-- not using the isp modem/router's wifi).

On that other topic lol, I'm using Fido's special data only rate plan (offer ended in March) to get 3GB for only $25/month on my phone, then using Freetone for free phone number calling (incoming and outgoing across US and Canada) and texting, as well as free voicemail and caller id (they also provide a new Canadian number for free). It's very reliable so far. As you know, calling uses very little data (about 0.5-1MB/min). Cheers from Vancouver

 

[TonyH]

I'm going to use it in bridge mode (if that is what I think it is-- not using the isp modem/router's wifi).

On that other topic lol, I'm using Fido's special data only rate plan (offer ended in March) to get 3GB for only $25/month on my phone, then using Freetone for free phone number calling (incoming and outgoing across US and Canada) and texting, as well as free voicemail and caller id (they also provide a new Canadian number for free). It's very reliable so far. As you know, calling uses very little data (about 0.5-1MB/min). Cheers from Vancouver

Hey, I am in Calgary. My daughter-in-law is tenured prof. at UVic. Used to teach at the Uof C. I spent about 20 some years field testing Multics mainframe new hardware at UofC CS dept. working with design engineers and guys at CISL , MIT. I quit USDOD military civilian job as C&I and M&O personnel on STRATCOM before then. Autodin/autovon was becoming Arpanet that time. Soon Internet was born and started growing like wild fire, LOL! Signals above that certain level it becomes stable and can sync good. Stronger signals also suffer less when interference occurs compared to weaker signal. Good luck with your decision.

 

[stevech]

I've literally spent *hours* carefully combing through the data on SNB. I have now cancelled two separate router orders (which is quite unlike me) upon coming across new information. The more data I came across today, the more confused I became, as many articles (and recommendations by its author) seem to contradict each other, to the point that it has me questioning the legitimacy of such. All I really want is a long range 5Ghz downlink oriented router (with at least 20Mbps at, preferably, location F)-- any other considerations (such as "class") are purely secondary. I am not an expert.

At first, I thought the Archer C7 (AC1750) would be the best as the router charts show it providing the highest throughput at location F of any router in the charts at ~115Mbps (using the method just prior to the current test method). It supposedly roughly equals the RT-AC66U in this respect:

http://www.smallnetbuilder.com/wire...ess-dual-band-gigabit-router-reviewed?start=3


Then, however, I came across a post by the site's author, who linked to a particular comparison chart (such comparisons are seemingly unavailable to users). It recommends going with the RT-N66U, as it bests its aforementioned AC cousin with 85Mbps at 45dB of attenuation vs. 0Mbps at 39dB (however, in the RT-N66U review, it is explicitly stated that the router only operates at less than 2Mbps, yes 2Mbps, at location F):

http://www.smallnetbuilder.com/tool...rk-knight/1140-asus-rtac66u/1226-asus-rtac56u


So, I figured the RT-N66U is the best of all the routers mentioned for 5Ghz downlink range. Then, I came across this post which, again, seems to contradict all the above data, showing up to a ~400% improvement by a number of AC1900 class routers compared to the RT-N66U at location D:

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance


So, which is correct, the router charts, the forum post by thiggins (the second link), or the results of the last test which pitches the, presumably, best N class router against newer AC ones...?

What is the best router for this purpose?


*Edit: Then there's this recent comparison:

http://www.smallnetbuilder.com/wire...32896-will-a-new-ac-router-bring-you-wifi-joy

Soooo confused....

*Edit2: Then I also came across this, which states that the RT-AC66U produces 52Mbps at location F!!!:

http://www.smallnetbuilder.com/wire...1750-gigabit-router-reviewed?showall=&start=3

I use/recommend the modest cost dual band non-AC WiFi routers... because things that use WiFi are mostly handhelds not high-demand desktops. Example: I've been using an ASUS RT-N56U. A year younger dual band is what I'd buy now. The ASUS just runs. Never needs a reboot. It's 1+ years old, dual band. My wife's iPad in 2.4GHz hammers it hours every day, streaming internet cooking videos, etc.

From another room. My Samsung Galaxy S6 phone is another light user - but I ran speedtest.net on it just to see.. and it reports, consistently, 100+ Mbps down and 10+ up. From a room away. On 5GHz. The phone like most handhelds, switches bands as the urge hits it, and rarely switches back to 5GHz. This is unavoidable with most consumer WiFi due to the IEEE standards.
A desktop PC or immobile laptop on a desk, should connect to the router by cat5 cable, IP over power wiring, or IP over existing TV coax cable (MoCA), or use WiFi if it has to be, and reduce expectations with an immobile PC.

Also... the indicated WiFi speed is the raw burst rate on WiFi. The net speed after the overhead of WiFi and IP is about 60% of the indicated WiFi speed, assuming you don't lose speed due to a neighbor that is a bandwidth hog and is near your chosen channel.

So I urge people to not over-spend.

 

[thiggins]

First, if you have a question specifically for me, it's best to PM me, i.e. start a "Conversation". I don't have time to read every post.

You are trying to compare results from many different test methods, which will indeed lead to madness, or at least a bad headache.

The easiest way to do the comparison you want is to go to the Router Ranker and select 5 GHz Range instead of the default Total router Ranking. Here's the rank for AC1900 class routers.
http://www.smallnetbuilder.com/tools/rankers/router/ranking/AC1900/rev8/133

Repeat for other class routers.

You can't accurately compare between classes below AC1750, even when using the same test method because we test with a 3x3 client. So classes below AC1750 are going to show lower throughput because they can't reach the maximum 3x3 link rates.

You can compare 5 GHz results for AC1750, AC1900 and higher class products. Test results on products with 4x4 5 GHz radios are limited by the 3x3 test client.

Though the only thing I really still don't understand is how in the review of the RT-N66U it only gets about 2Mbps at location F, whereas in the charts it shows it getting 82.5Mbps....

That's because the Charts have the latest data, which is from the retest

 

[djayjp]

Hey, I am in Calgary. My daughter-in-law is tenured prof. at UVic. Used to teach at the Uof C. I spent about 20 some years field testing Multics mainframe new hardware at UofC CS dept. working with design engineers and guys at CISL , MIT. I quit USDOD military civilian job as C&I and M&O personnel on STRATCOM before then. Autodin/autovon was becoming Arpanet that time. Soon Internet was born and started growing like wild fire, LOL! Signals above that certain level it becomes stable and can sync good. Stronger signals also suffer less when interference occurs compared to weaker signal. Good luck with your decision.

Thanks Tony! I just saw these responses now. I have no idea how to respond to 'conversations'. But I checked out the r7800 that you mentioned-- it only gets 4.5mbps 5ghz downlink throughput "at location" (pretend location) F.... ;P Anyway, I've got the Archer C9 now and it seems to be working well with my Galaxy s6, but I'm having major problems with my Archer T2U cheapo ac600 usb adapter that I also just got... ugh. I went from streaming 1080p/60 youtube to 144p.... Yes. I'm thinking the windows 8.1 driver might have something to do with it (I have win10). I just ordered the Netgear 6210. We'll see how that goes.

 

[djayjp]

I use/recommend the modest cost dual band non-AC WiFi routers... because things that use WiFi are mostly handhelds not high-demand desktops. Example: I've been using an ASUS RT-N56U. A year younger dual band is what I'd buy now. The ASUS just runs. Never needs a reboot. It's 1+ years old, dual band. My wife's iPad in 2.4GHz hammers it hours every day, streaming internet cooking videos, etc.

From another room. My Samsung Galaxy S6 phone is another light user - but I ran speedtest.net on it just to see.. and it reports, consistently, 100+ Mbps down and 10+ up. From a room away. On 5GHz. The phone like most handhelds, switches bands as the urge hits it, and rarely switches back to 5GHz. This is unavoidable with most consumer WiFi due to the IEEE standards.
A desktop PC or immobile laptop on a desk, should connect to the router by cat5 cable, IP over power wiring, or IP over existing TV coax cable (MoCA), or use WiFi if it has to be, and reduce expectations with an immobile PC.

Also... the indicated WiFi speed is the raw burst rate on WiFi. The net speed after the overhead of WiFi and IP is about 60% of the indicated WiFi speed, assuming you don't lose speed due to a neighbor that is a bandwidth hog and is near your chosen channel.

So I urge people to not over-spend.

Thanks for the ideas, Steve, though I ended up going AC instead of N because of this article:

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance

 

[djayjp]

First, if you have a question specifically for me, it's best to PM me, i.e. start a "Conversation". I don't have time to read every post.

You are trying to compare results from many different test methods, which will indeed lead to madness, or at least a bad headache.

The easiest way to do the comparison you want is to go to the Router Ranker and select 5 GHz Range instead of the default Total router Ranking. Here's the rank for AC1900 class routers.
http://www.smallnetbuilder.com/tools/rankers/router/ranking/AC1900/rev8/133

Repeat for other class routers.

You can't accurately compare between classes below AC1750, even when using the same test method because we test with a 3x3 client. So classes below AC1750 are going to show lower throughput because they can't reach the maximum 3x3 link rates.

You can compare 5 GHz results for AC1750, AC1900 and higher class products. Test results on products with 4x4 5 GHz radios are limited by the 3x3 test client.

That's because the Charts have the latest data, which is from the retest

Thanks for your reply, Tim! It's a great site! I really think you should do both an enclosed, controlled box test and an open air, real-world tests for all reviews, as well as including an AC phone client or laptop. As you yourself have admitted, the results can be tremendously different (better and worse)!

Oh, and if you could do another roundup of usb adapters that would be amazing

 

[TonyH]

Thanks Tony! I just saw these responses now. I have no idea how to respond to 'conversations'. But I checked out the r7800 that you mentioned-- it only gets 4.5mbps 5ghz downlink throughput "at location" (pretend location) F.... ;P Anyway, I've got the Archer C9 now and it seems to be working well with my Galaxy s6, but I'm having major problems with my Archer T2U cheapo ac600 usb adapter that I also just got... ugh. I went from streaming 1080p/60 youtube to 144p.... Yes. I'm thinking the windows 8.1 driver might have something to do with it (I have win10). I just ordered the Netgear 6210. We'll see how that goes.

AC6210 is not working well either at the moment. AC6200 I have went thru the same pain when first came out. It works just fine now. I hope AC6210 will work well soon when all the bugs are taken care of. Over all in my house R7800 is best as far as signal strength goes compared to Linksys EA8500, Netgear R7500, R7000, etc.. We never have problem live streaming to home theater using 5GHz band. All PC, desktop in the house are on W10 with auto update disabled. My laptop is still on W7 Ultimate and Ubuntu dual boot. Have fun with new router. Often I find from corporate surplus pile, UTM class routers. Good toy to play with
learning something new always.

 

[djayjp]

AC6210 is not working well either at the moment. AC6200 I have went thru the same pain when first came out. It works just fine now. I hope AC6210 will work well soon when all the bugs are taken care of. Over all in my house R7800 is best as far as signal strength goes compared to Linksys EA8500, Netgear R7500, R7000, etc.. We never have problem live streaming to home theater using 5GHz band. All PC, desktop in the house are on W10 with auto update disabled. My laptop is still on W7 Ultimate and Ubuntu dual boot. Have fun with new router. Often I find from corporate surplus pile, UTM class routers. Good toy to play with
learning something new always.

Uh oh... that doesn't sound good at all. Did you update the firmware? There was an update for it this April I think. I've also heard good things about the a6200. That's interesting about the r7800. It could be explained by the discrepancies Higgins has noticed between box testing and open air-- that the results can be completely different...! It hasn't been tested open air afaik.

 

[TonyH]

Uh oh... that doesn't sound good at all. Did you update the firmware? There was an update for it this April I think. I've also heard good things about the a6200. That's interesting about the r7800. It could be explained by the discrepancies Higgins has noticed between box testing and open air-- that the results can be completely different...! It hasn't been tested open air afaik.

Between the router and family room where our home theater is located, first time 5GHz came up below 60db. 2.4GH is 6-7db better than 5GHz. A6200 initially was unusable. Just loading the driver only got little better, then eventually good driver came out and it is rock solid performer now.

 

[stevech]

6dB advantage of 2.4GHz over 5GHz is just where laws of physics says it will be, if the transmitted power is the same in both bands.

 

[djayjp]

Between the router and family room where our home theater is located, first time 5GHz came up below 60db. 2.4GH is 6-7db better than 5GHz. A6200 initially was unusable. Just loading the driver only got little better, then eventually good driver came out and it is rock solid performer now.

So far my a6210 is working alright, certainly MUCH better than the TPLINK T2U-- with that I was getting 10% packet loss and completely regular 3500ms ping times to modem following the lost packet...! With the a6210 I'm still getting the same 10% loss, unlike my android phone at the exact same spot (which is 0%), but the ping times are all 1-2ms as expected (one odd one was 500ms though, but that was out of 500 pings). I'll call netgear tomorrow.... Have you tried the "ping 192.168.x.1 -t" command (ctrl-c to stop)?

*Edit.... I keep seeing it drop down to 173 or even 59Mbps link speed every few seconds.... ughhhhh

*Edit #2: So it seems that link speed issue is a power saving thing. When I'm actually using the wlan it never seems to drop below ~700mbps.

*Edit #3: after further testing, now instead of dropping packets, it appears the regular ~3300ms latent ping to modem has returned.... It happens almost exactly every 5th-10th ping. Any ideas?

*Edit #4: did testing from router to modem and all is well there.... So two totally different (assuming chipset isn't the same ha) usb wifi adapters, same pc, and both exhibit almost exactly the same issue....

 

[djayjp]

Not

Between the router and family room where our home theater is located, first time 5GHz came up below 60db. 2.4GH is 6-7db better than 5GHz. A6200 initially was unusable. Just loading the driver only got little better, then eventually good driver came out and it is rock solid performer now.

Not sure how to check dB signal strength in windows, but on my phone it reports -56dB which is pretty good I think.

 

[pete y testing]

but on my phone it reports -56dB which is pretty good I think.

so how it goes is 20 to 30db is excellent , 30 to 40db very good , 40 to 50db good , 50 to 60db good /average , 60 to 70db is very average and anything above -70db poor

but this also depends on the client adapter and its antennas etc as mobile phones have very poor antennas when compared to most other things

 

[TonyH]

Not

Not sure how to check dB signal strength in windows, but on my phone it reports -56dB which is pretty good I think.

Are you using inSSIDer or Acrylic? Sounds like you may have
poor s/n problem or interference on your channel just guessing. I am not using A6200 daily, it is just back up piece
for my audio station in the basement. For home theater HTPC I use Bigfoot Killer wifi Card which is only -N mode but very reliable and works just fine for our needs. Other laptops have Intel 7260. In the house speed test shows always 50/5 or little more which is what my ISP provides.