Wifi 6 router is it worth it with Fibremax ?

[Cedrik]

Greetings,

Am looking at the possibility of getting a Wifi 6 router as I now have Fibremax 300 installed. Apart from future proofing, are there any compelling reasons to get one / not to get one.

I must say I'm impressed with the Wifi capabilities of the new Smart Router, but am looking more to ther long term.

[psychopomp2]

Yep, the EE supplied router should give you wired speeds of 900 Mbps on their FibreMax 900 plan. But I'm almost 100% certain it won't give you 900 Mbps over wifi, for that you will need a wifi 6 router and of course wifi 6 clients as well.

[soupladel]

@psychopomp2 I have an Asus Ax11000 on Fibremax 900 and i only wish i could get wifi speeds that high, mine tops out about 600mbps, but i get 900Mbps consistently when using wired.

Have spent near £400 on the Ax11000 i am not inclined to go buy another router just to determine if the AX11000 is faulty, so for me the upgrade to wifi 6 hasn't so far been worth it

[mikeliuk]

Hi @soupladel ,

Is the device topping out at 600 Mbps near the router and making a WiFi 6/802.11ax connection?

Do two such devices running a bandwidth test each obtain around 450 Mbps at the same time?

[soupladel]

@mikeliuk Yup, I have confirmed that any devices tested are connect to the 5GHz_1 band at AX speeds with 160Mhz.  I have tried a variety of distances and lines of sight to rule out it being an environmental problem and none of it has helped.

The one thing i had neglected to do was as you suggested and run two devices at the same time  at the same location and it turns out that the combined speed of both amounts to roughly 600Mbps.

So it seems on that basis that the problem is almost entirely down to the router not being able to supply full speed by wifi, so the question is how do i convince a retailer of that now in my attempts to return it?

Or is there something i am missing that might be limiting the wifi throughput of the router? i dont have QoS enabled and i am not using anything like smart connect either.

[XRaySpeX]

@soupladel : Only the statutory 14-day cooling-off period if still applicable if you bought online or by phone.

[soupladel]

@XRaySpeX the router at this point is approx 5 months old, so well outside any initial cooling off period, but on the basis it is covered by at least a 12 month guarantee, at the very least a repair should be offered at this point, the problem is getting them to believe it is in fact faulty.

[mikeliuk]

Hi @soupladel ,

On the single-device number, you may be hitting a client-side limitation such as found at https://www.technologyfocus.net/network-how-much-faster-is-wifi-6-than-wifi-5/ with 2x2 MIMO capping at around 600 Mbps and 3x3 MIMO hitting 700 Mbps.

Two devices of the same capability hitting an aggregate cap of 600 Mbps is unexpected and the best play is probably ensure the two 5 GHz radios have their own SSID (if possible) and to try a simultaneous benchmark that way.

5G-1Hz AX: 4x4 (Tx/Rx) 1024 QAM 20/40/80/160MHz, up to 4804Mbps
5G-2Hz AX: 4x4 (Tx/Rx) 1024 QAM 20/40/80/160MHz, up to 4804Mbps

https://rog.asus.com/uk/networking/rog-rapture-gt-ax11000-model/spec

The AX11000 has been found to perform below its headline rate and below competing products but on balance I don't think it is sufficiently poor to justify a return in all circumstances. It's possible that a later firmware version could perform somewhat better. https://www.tomsguide.com/us/asus-rog-rapture-gt-ax11000,review-6373.html

I think you would need to show transmission errors to argue there is some fault in the silicon somewhere.

[soupladel]

@mikeliuk i think you have hit the nail on the head because when i try the same two devices on separate SSIDs, i get a result closer to full speed. therefore i am just going to have to accept i have a router that only really operates at roughly 75% of its advertised speeds in a single device scenario over wifi

[mikeliuk]

It's good to hear you've now got new methodologies to benchmark and characterize the performance of your hardware which, being cutting-edge, is typically quite expensive with a premium over older technology.

My feeling is that 75% of some cap/bottleneck for a single device is very good and could well be explained if it is a 3x3 MIMO device or below. Possibly a 4x4 MIMO client, if one exists, could achieve better so it may not all be down to any bottleneck with the router.

The high bandwidth capability of the router is actually bottlenecked by the internet WAN so a fairer test of the WiFi capability would actually be to transfer data within the LAN, for example using multiple phones or laptops to access files which are on a server directly wired to the router (e.g. purchased 4K UHD movies or lossless music). For Linux users, synthetic iperf tests could be used to see how capable the router is.

Making a note that using the third radio would additionally help to show the switching capacity of the router.

2.4GHz AX: 4x4 (Tx/Rx) 1024 QAM 20/40MHz, up to 1148Mbps