Can device with WiFi 5 (AC) benefit from WiFi 6 router (AX)?

[junzhi2002]

With the increasing usage of Zoom online class recently, my ISP included router just can't keep up with the demand of the bandwidth and keep crashing. Although I took off the shroud/cover of the router and add a fan blowing towards the router, the frequent crashing solved but occasion crash still exists. Yesterday I stumble on a cheap router and review says it's quite capable of handling heavy traffic, giving me low latency and high bandwidth Internet. Link below vvv

https://www.tp-link.com/us/home-networking/wifi-router/archer-ax10/

 

Almost all device on the network uses WiFi 5 which isn't same as the router.

Will it help to reduce the crashing or even solve the crashing and give me better internet experience? Or I should just get an even cheaper AC router to save more buck

 

IDK which information about the device I am using currently I should include. Please tell me which I should include.

[Electronics Wizardy]

Well the crash shouldn't be happening, even under heavy load, so im guessing a new usnit will fix this.

 

AX won't help with ac devices, it will run a mode that is compatable with ac devices.

 

 

 

[HenrySalayne]

15 minutes ago, Electronics Wizardy said:

AX won't help with ac devices, it will run a mode that is compatable with ac devices.

This is indeed true, but a more modern device might be a little bit better because of advances in technology. It could also be worse.

 

A real improvement can be made by increasing the number of parallel streams. Cheap APs will use only one or two while more expensive and better units might use 3. For maximum performance make sure your device's number is at least matched by your AP's number of streams, e. g. 2x2 or 3x3. Upgrading from 1 stream to 3 streams (if your device supports it) can almost triple the bandwidth.

[junzhi2002]

50 minutes ago, Electronics Wizardy said:

Well the crash shouldn't be happening, even under heavy load, so im guessing a new usnit will fix this.

 

AX won't help with ac devices, it will run a mode that is compatable with ac devices.

 

 

 

Wouldn't it improve anything? Like latency or something else? Or a decent AC router will perform same as the AX router?

On the technical specification, "budget-ish" AX WiFi router has more and faster cores then same price AC WiFi router, will that impact performance?

I have about 6~7 device connected almost 24/7 and max device connected on it will be about 10, in a 236,000 single storey house.

[Electronics Wizardy]

29 minutes ago, junzhi2002 said:

Wouldn't it improve anything? Like latency or something else? Or a decent AC router will perform same as the AX router?

On the technical specification, "budget-ish" AX WiFi router has more and faster cores then same price AC WiFi router, will that impact performance?

I have about 6~7 device connected almost 24/7 and max device connected on it will be about 10, in a 236,000 single storey house.

well all the new fancy stuff in ax won't be used, as both client and ap have to support it.

 

Other features in the router can be usedfull, but for most home uses, your liminted by the wifi, not the routing performance of the router. 

 

What bandwidth we thinking about?

[Alex Atkin UK]

2 hours ago, junzhi2002 said:

Wouldn't it improve anything? Like latency or something else? Or a decent AC router will perform same as the AX router?

On the technical specification, "budget-ish" AX WiFi router has more and faster cores then same price AC WiFi router, will that impact performance?

I have about 6~7 device connected almost 24/7 and max device connected on it will be about 10, in a 236,000 single storey house.

Latency, probably not.  Raw bandwidth, maybe, for the reason you described - the CPU is likely to be less stressed.  Some AC routers are so weak that pushing top-speed on WiFi can actually slow down the Internet as the CPU is maxed out.

There is also the fact a lot of AC routers only support 80Mhz channel width,. AX routers I believe usually support 160Mhz so you can push more bandwidth, although this can vary a lot as with such a large channel width crosstalk/interference can have a huge impact.  I get anything from 400Mbit to 700Mbit on a 160Mhz channel.

I'm not going to go into details of things like MU-MIMO as I'm not convinced it makes any difference whatsoever unless you have super high-end business grade kit that has REAL beamforming (antennas that can physically move the beam direction rather than tons of fixed antennas).

[RMerlin]

31 minutes ago, Alex Atkin UK said:

I'm not going to go into details of things like MU-MIMO as I'm not convinced it makes any difference whatsoever unless you have super high-end business grade kit that has REAL beamforming (antennas that can physically move the beam direction rather than tons of fixed antennas).

 

MU-MIMO requires a minimum of two clients that support MU-MIMO to bring any benefits - and these benefits remain fairly marginal overall.  A lot of MU-MIMO implementations are also broken (like the early ones on routers using the BCM4366 - the BCM4366E revision is required, with its beefier built-in CPU).

 

MU-MIMO clients still remain fairly uncommon as well.

 

35 minutes ago, Alex Atkin UK said:

There is also the fact a lot of AC routers only support 80Mhz channel width,. AX routers I believe usually support 160Mhz so you can push more bandwidth, although this can vary a lot as with such a large channel width crosstalk/interference can have a huge impact.  I get anything from 400Mbit to 700Mbit on a 160Mhz channel.

Note that to be able to use 160 MHz you need a client that supports it (very few do), and also to be in a region that allows the use of DFS channels (there isn't enough continuous spectrum to support 160 MHz if your country only allows non-DFS bands).

 

I've been able to hit 900 Mbps here with 160 MHz (slightly below 900 Mbps with my Huawei P30).  My laptop equipped with an Intel AX200 can saturate my NAS's 1 Gbps Ethernet link, however that's in part due to AX.

 

 

The main benefits for AC clients with an AX router is that quite often, the AX router will use newer, better components.  That can translate in slight range and throughput improvements (better amps and filters on the radio, better CPU both on the radio and on the router itself).  But don't expect anything earth shattering tho.

 

[Alex Atkin UK]

7 minutes ago, RMerlin said:

 

MU-MIMO requires a minimum of two clients that support MU-MIMO to bring any benefits - and these benefits remain fairly marginal overall.  A lot of MU-MIMO implementations are also broken (like the early ones on routers using the BCM4366 - the BCM4366E revision is required, with its beefier built-in CPU).

 

MU-MIMO clients still remain fairly uncommon as well.

 

Note that to be able to use 160 MHz you need a client that supports it (very few do), and also to be in a region that allows the use of DFS channels (there isn't enough continuous spectrum to support 160 MHz if your country only allows non-DFS bands).

 

I've been able to hit 900 Mbps here with 160 MHz (slightly below 900 Mbps with my Huawei P30).  My laptop equipped with an Intel AX200 can saturate my NAS's 1 Gbps Ethernet link, however that's in part due to AX.

 

 

The main benefits for AC clients with an AX router is that quite often, the AX router will use newer, better components.  That can translate in slight range and throughput improvements (better amps and filters on the radio, better CPU both on the radio and on the router itself).  But don't expect anything earth shattering tho.

 

I know, I used two laptops with AX200 clients, even moved them to opposite sides of the nanoHD, but the speed was an even-split between them of half of the speed a single client can get.

As its a 4x4 MU-MIMO device that can split into dual-2x2, according to the bold claims it should have been able to operate both at full-speed (well it only has a Gigabit LAN port, so that would be a limit, but it should have combined over 700Mbit).  Maybe its another broken implementation, but that's kinda the point isn't it, its not worth paying more for a feature that probably doesn't actually work.

[RMerlin]

1 minute ago, Alex Atkin UK said:

but that's kinda the point isn't it, its not worth paying more for a feature that probably doesn't actually work.

 

Exactly.  MU-MIMO is not something you should be worrying about when shopping for either an AP or a client.

 

So far, initial tests about OFDMA seem to indicate that this will be a similar case: nice technology on paper, next to no measurable improvement in real-life scenarios.  Tim Higgins at SmallNetBuilder has done some fairly extensive testing of OFDMA recently, and the results were disappointing.

[Alex Atkin UK]

23 minutes ago, RMerlin said:

 

Exactly.  MU-MIMO is not something you should be worrying about when shopping for either an AP or a client.

 

So far, initial tests about OFDMA seem to indicate that this will be a similar case: nice technology on paper, next to no measurable improvement in real-life scenarios.  Tim Higgins at SmallNetBuilder has done some fairly extensive testing of OFDMA recently, and the results were disappointing.

I do tend to keep an eye on SmallNetBuilder, but I have also seen his testing show less performance than I was getting on the same hardware.  So its pretty hard as firmware is an ever moving target.

I'm really hoping he is wrong with OFDMA, but considering its a technology that is also used on LTE and that has absolutely atrocious performance when the tower is loaded, it doesn't look promising.

[junzhi2002]

16 hours ago, Alex Atkin UK said:

I do tend to keep an eye on SmallNetBuilder, but I have also seen his testing show less performance than I was getting on the same hardware.  So its pretty hard as firmware is an ever moving target.

I'm really hoping he is wrong with OFDMA, but considering its a technology that is also used on LTE and that has absolutely atrocious performance when the tower is loaded, it doesn't look promising.

 

17 hours ago, RMerlin said:

 

Exactly.  MU-MIMO is not something you should be worrying about when shopping for either an AP or a client.

 

So far, initial tests about OFDMA seem to indicate that this will be a similar case: nice technology on paper, next to no measurable improvement in real-life scenarios.  Tim Higgins at SmallNetBuilder has done some fairly extensive testing of OFDMA recently, and the results were disappointing.

So the solution is to just change it to a decent AC router?

19 hours ago, Electronics Wizardy said:

What bandwidth we thinking about?

300Mbps download and 100Mbps upload capped.

[RMerlin]

9 hours ago, junzhi2002 said:

So the solution is to just change it to a decent AC router?

Right now it's probably a smarter purchase, yes, until lower-priced AX routers start to come to market.  They are just starting to appear, with products like the Asus RT-AX56U.  There is little point in overpaying to get AX right now.

 

[junzhi2002]

8 hours ago, RMerlin said:

Right now it's probably a smarter purchase, yes, until lower-priced AX routers start to come to market.  They are just starting to appear, with products like the Asus RT-AX56U.  There is little point in overpaying to get AX right now.

 

I found one which is TP-Link Archer AX10, don't know it's good or not but I cost about $80. Review says that it deliver good speed with AX and I think it will future proof a little when AX devices start to become common. But I think I will mark this question as solved after this  

[Falcon1986]

9 hours ago, junzhi2002 said:

I found one which is TP-Link Archer AX10, don't know it's good or not but I cost about $80. Review says that it deliver good speed with AX and I think it will future proof a little when AX devices start to become common. But I think I will mark this question as solved after this  

This article from smallnetbuilder.com came out last October, but is still relevant today. There doesn't seem to be that much difference between AX and AC at this time. That's not to say that AX isn't an improvement; but very few client devices can take full advantage of AX and not all of WiFi 6's features are activated on all WiFi 6 wireless routers at this time. That's because the standard is still being worked out.

 

I wouldn't buy a WiFi 6 wireless router now with hopes of "future-proofing" anything. You can't future-proof with something that's still in the making, which might change in a few months. If you just want a better-performing wireless router and the cost is within your budget... and it just happens to support AX, then that would be a better reason.

[Alex Atkin UK]

The biggest improvement will come from WiFi 6e IMO, where you get a ton of new channels in the 6Ghz range that being even shorter range than 5Ghz should be easier to avoid any crosstalk.

Lots of people wont be happy with that of course, needing more Access Points in the house, but its literally the ONLY solution to improving WiFi reception in built-up areas.