Wi-Fi 6 Performance Roundup: Five Routers Tested

[thiggins]

We take a look at how five Wi-Fi 6 routers perform with a Wi-Fi 6 client.

Read on SmallNetBuilder.

 

[Trip]

Well done, Tim. Thanks for persisting and continuing to put out quality reviews, even if they are fewer over time.

 

[coxhaus]

Yes nice review. It looks like 2.4GHz is going to receive the strength it needs to play in the new high speed internet world.

 

[Hawk]

Well written article Tim, your review confirmed my suspicion of poor 5 ghz downlink performance, compared to uplink.

 

[thiggins]

Yes nice review. It looks like 2.4GHz is going to receive the strength it needs to play in the new high speed internet world.

Well, higher bandwidth and a little range boost help. But with 2.4 GHz channels as congested as they are, users may find it difficult to see a real benefit.

 

[Makaveli]

Great review.

I don't have any AX devices and when I do lets hope the firmware performance has improved on the AX 5Ghz side.

 

[sfalcon]

Glad to see your article again, very informative! good job!

 

[Val D.]

Wi-Fi 6 is the new MU-MIMO on a 2x2 router.

 

[keenanj]

Great article Tim it is interesting that the consumer grade routers are quite a bit slower than what i saw with the business class ews377ap and r730 https://www.keenansystems.com/wordpress/2019/10/03/exactly-how-fast-is-the-wi-fi-6-in-the-iphone-11/

 

[Razor512]

I wonder, have these ax routers done anything to improve the performance of purely ac clients?

 

[thiggins]

Great article Tim it is interesting that the consumer grade routers are quite a bit slower than what i saw with the business class ews377ap and r730 https://www.keenansystems.com/wordpress/2019/10/03/exactly-how-fast-is-the-wi-fi-6-in-the-iphone-11/

Could be the difference in clients.

 

[thiggins]

I wonder, have these ax routers done anything to improve the performance of purely ac clients?

That data is in the article.

 

[pete y testing]

Could be the difference in clients.

hi tim

could that not be the case with your tests and conclusions in only using the intel wifi chipset that it may be causing the variations seen by the difference in router chipsets performance as its noticable all the broadcom chipset routers performed worse

i know using the single client chipset gives you a fixed known platform to compare but could the intel its self have performance issue both in sync and interoperability

i am doing some testing on the asus rt-ax92u and its ax backbone is connecting at the full 2402Mbps sync rate at 80mhz am about to do some throughput testing , will be interesting how broadcon to broadcom perform compared to your results

 

[thiggins]

could that not be the case with your tests and conclusions in only using the intel wifi chipset that it may be causing the variations seen by the difference in router chipsets

Sure. There is always that chance. That's why I tested with AC STAs, too. Some of the odd behavior is common to using both the Intel AX STA and Qualcomm N/AC STA used in the octoScope Pals.

Remember too, that AP to AP chipset, which is the case with two routers forming a bridge, can produce different result from using a STA chipset (Samsung S10/ iPhone 11), even from the same vendor.

 

[sfx2000]

Interesting article...

I'm thinking folks should wait until next year on consumer gear - as both silicon and firmware are still under heavy development.

It took a couple of years actually for 802.11ac Wave 2 to get fully sorted across the different vendors, and even there, interoperability across different vendors is still a question

 

[pete y testing]

Remember too, that AP to AP chipset, which is the case with two routers forming a bridge, can produce different result from using a STA chipset (Samsung S10/ iPhone 11), even from the same vendor.

agreed tim

I'm thinking folks should wait until next year on consumer gear

also agree , at least till OFDMA , mu-mimo and other ax feature are fully enabled and working they way they should

 

[ng4ever]

Interesting article...

I'm thinking folks should wait until next year on consumer gear - as both silicon and firmware are still under heavy development.

It took a couple of years actually for 802.11ac Wave 2 to get fully sorted across the different vendors, and even there, interoperability across different vendors is still a question

Agreed.

 

[pete y testing]

hi guys

can someone confirm something for me

here in australia ive been told we only have spectrum to operate 160MHz which is between CH36-CH64.

so this would mean a tri band ax router would not be able to run 160Mhz if the 4 x 4 is on band 3 only

i see the ax1100 has 4 x 4 on band 2 and 3

is it capable of 160Mhz on band 2 and band 3

pete

 

[pege63]

The line where it says "there's no sense in making a large investment that may not pay off" says it all to me anyway!

 

[avtella]

hi guys

can someone confirm something for me

here in australia ive been told we only have spectrum to operate 160MHz which is between CH36-CH64.

so this would mean a tri band ax router would not be able to run 160Mhz if the 4 x 4 is on band 3 only

i see the ax1100 has 4 x 4 on band 2 and 3

is it capable of 160Mhz on band 2 and band 3

pete

In your region, on a triband or dual band router, doesn’t really matter which, only the lower channels can use HT160. The upper channel radio in your triband router/AP would be limited to HT80 which I assume you are referring to as “band 3”.

Here in the states however, we have a contiguous HT160 band in the upper 100-128 channels but that’s fully DFS and I believe with lower power limits than the 36-64 area and 149-165 area.

HT80+80 split bonding is a thing for non-contiguous 80 MHz widths but I don’t know of any clients that actually support it.