WiFi 6 was not worth it for me ~ Journey of WiFi 6 AiMesh back to WiFi 5 UniFi ~

[Tech Focus]

Who is this for?

I've written this to share my personal experience to those who believe WiFi 6 is the best one can get right now. Yet, your experience is not quite as impressive as you expected. Perhaps, debating WiFi 5 vs. WiFi 6 because of different reviews. This is one technophilic person's real world experience who usually go off by specs with the latest technology. Here I specifically evaluated Asus AiMesh system for over a year.

Since there is variability and many environmental confounding factor, I would primary look relative number/trend over the absolute number if you are trying to apply to your home setting. I.e. A can be X times faster than B is more important data here than A is Y Mbps faster than B on these data.

Disclaimer: This is just my personal experience and may not be generalizable.

Introduction

I know many sites says WiFi 6 is still expensive, so value wise go for WiFi 5 Wave 2 product, but I want the best/fastest and feature-rich system even its expensive. We get what we pay for. This is my hobby so budget is not a primary concern. So screw all those reviews, I'd still get WiFi 6. Well, I was wrong.

This is by no means as controlled experiment as SmallNetBuilder's hot article. But my real world setting test agreed with SNB's result. This is a just little journey of how I came back from my over $1000 WiFi 6 setup using for 1 year back to WiFi 5 setup.

Mesh Network to Wired AiMesh

Past three years, I have been in search for home network system that can provide my home full, speedy, stable WiFi coverage. I've started with Orbi (AC3000 back then). Due to its instability, I moved onto Eero Pro x 3 setup. Then my obsession for speed (though mostly just looking at numbers rather than actual impact on day to day use), I had Cat 6 ethernet cable installed throughout my home. I then subscribed to 1 Gbps internet service, the first one in my area. To get the full benefit of 1 Gbps download speed, 10 Gbps capable NAS, I have yet went out search for new network system. I then found AiMesh. Brilliant concept and I thought I had discovered my dream system. I ended up spending over $1000 for AiMesh setup with Ax11000 as a router and a wired AiMesh node, 2 additional WiFi 6 capable Ax6100 routers as additional WIRED AiMesh nodes. Living with AiMesh system for over a year, I was decently happy mainly because of the great specs and I felt I'm totally up to date with new tech, WiFi 6. Oh and I didn't have my first WiFi 6 client until last November (iPhone 11 Pro) and it still remains only WiFi 6 client as of today amongst 50 WiFi connected home devices.

However, there was still one area in our house (despite 4 Access Points) that had poor WiFi signal where we recently started to use more WiFi and discovered an issue of intermittent internet disconnects/slow down. I've also had complaints from family members for occasional disconnect on their devices. At first, I was looking at ZenWiFi as one (technically 2 as they come in pair) more wired AiMesh node. However, knowing this won't really fix instability my family members were facing, I've started the forth round of the search for potentially different network system. As I do this, I saw various speed test results, mostly utilizing iPerf 3, across internet. The numbers were widely variable with interesting thing were some measures show their numbers on WiFi 5 system reached compatible range to that of WiFi 6.

Reality of WiFi 6

So I've decided to do my own internet speed test to see how much faster WiFi 6 really is. To be honest, up until this point, I really did not realize how WiFi speed really calculated. I was really going off by numbers. When I first saw WiFi 6 or giga-bit internet service, 4800 Mbps sounded like over 10x speed compared to my internet speed test result of 300 Mbps. I did not even know there was theoretical speed maximum in WiFi based on Wireless physical link.

The first part of test was using Asus Ax11000 as a router with different clients including WiFi 6 capable client i.e. variable here was client side.

I got about 30% speed gain, which is close enough to Intel's claim of 40% especially without using 160 MHz channel. But in reality with some practical range, the speed gain was 13%, which again is similar to what's been reported by other for 10% practical gain.

The speed gain is speed gain. I plan to buy more WiFi 6 capable products (though probably hold until 6E), one question came to my mind during this testing process. That is WiFi inefficiency ranges from 30-50%. Asus system looked only 50% efficient. So what if I had a highly efficient WiFi 5 system? It can potentially close the 10-30% WiFi 6 speed gain gap on all my WiFi 6 device (only 1 for now) and gives rest of my devices extra speed boost. Oh and yes, I still had main execute of looking for more stable system.

So to test this, I chose UniFi system. First I started with a single access point from UniFi. Very underwhelming spec device and honestly I had my mind set to use this UniFi access point to confirm/convince myself that I do have the best set up I can buy for now (AiMesh). The second test I conducted were comparing AiMesh node Ax11000 and Unifi AP (UAP-AC-HD) both connected to the Ax11000 router via Ethernet in the same room, same location.

My hypothesis of system efficiency can close the current gap of WiFi 6 speed gain was proven to be true, and actually the results were beyond my expectation. Because even on Wifi 6 connected device, UniFi WiFi 5 AP's speed outperformed Ax11000 AiMesh Node. The same night, I ordered more UniFi products and I haven't been happier.

Limitation

I can't say the same applies to all other WiFi 6 system. In fact, I can't even say stand alone router would underperform as much as I saw on my tests. Because I've never owned other WiFi 6 routers or Mesh network system. Also, I did not test Ax11000 as completely stand alone router i.e. did not disconnect AiMesh nodes; other than confirming I had 940 Mbps download at Ethernet port on the router. However, looking at SmallNetBuilder's hot article, I surmise other system may suffer similar issue. SNB's AiMesh review itself is not stellar so it is possible this test results only applies to Asus AiMesh system.

Also, Tri-band mesh system should really shine when wireless, dedicated backhaul are utilized in an environment where there is no ethernet backhaul access point option available so link between mesh nodes won't be the bottleneck. My tests are ethernet backhaul based so appropriate for current maximum WiFi throughput assessment but ignoring one of the main feature of mesh system.

Final Thought

So in conclusion, WiFi 6 isn't worth for me (for now). If there is anyone like I was, and blindly believing newer technology is better, WiFi 6 at current state can be inferior in performance to well built Wi-Fi 5 Wave 2 system. Have I given up on WiFi 6? I highly doubt it. After all, this is my hobby. When UniFI releases WiFi 6 access point, I am sure I will purchase it on the day 1. WiFi 6E? Bring it on.

Thank you for reading!

 

[Gar]

I haven't done the WiFi6 trip yet, but appreciate all you went thru and for writing it up, thanks.

 

[K-2SO]

Different class, purpose and price range equipment.

 

[Tech Focus]

Different class, purpose and price range equipment.

In my case, I went for high-end Asus models so it actually turned out UniFi could have been a cheaper choice for one to one match system.

Though with my obsession, I ended up spending more than AiMesh setup but I was able to sell all my AiMesh parts on eBay.

 

[K-2SO]

In my case, I went for high-end Asus models so it actually turned out UniFi could have been a cheaper choice for one to one match system.

UDM is a AIO home router.
Proper UniFi system will cost more. With 3-4 APs in right places will kill any home router. UniFi is not really high-end.

 

[Sharpbarb]

UDM is a AIO home router.
Proper UniFi system will cost more. With 3-4 APs in right places will kill any home router. UniFi is not really high-end.

What is an example of a proper UniFi system?

 

[L&LD]

I haven't compared Asus to UniFi, but you haven't really compared WiFi 5 to WiFi 6 either?

In an 3800 SqFt home, 4 AP's are definitely too much RF.

Testing WiFi 'right next to the router' is also a 'non-test'.

If the systems compared were not set up optimally independent of each other with regards to unique SSID's, location, channels used and the strength of the radios properly adjusted, this was not a real comparison of their inherent strengths.

When a single RT-AC3100 can cover a 5000 SqFt home on multiple levels even when not optimally placed, there seems to be something amiss here?

It may be the construction materials that a higher number of AP's are required, but it doesn't explain how WiFi 5 is better than WiFi 6 though.

That goes against all the tests I've done. And I guarantee that equipment will not be upgraded if the existing equipment can be made to work substantially better first.

Of course, I've also never used stock Asus firmware for longer than it takes to flash RMerlin firmware. Nor have I used any GT router either.

Yes, I am a little skeptical here because my experience is very good with the two AX routers and the AiMesh AC routers I work with. Not that I am doubting your results in your home. Just that generalizing this for others may be short-changing them of the best possible WiFi in their homes.

 

[Tech Focus]

I haven't compared Asus to UniFi, but you haven't really compared WiFi 5 to WiFi 6 either?

In an 3800 SqFt home, 4 AP's are definitely too much RF.

Testing WiFi 'right next to the router' is also a 'non-test'.

If the systems compared were not set up optimally independent of each other with regards to unique SSID's, location, channels used and the strength of the radios properly adjusted, this was not a real comparison of their inherent strengths.

When a single RT-AC3100 can cover a 5000 SqFt home on multiple levels even when not optimally placed, there seems to be something amiss here?

It may be the construction materials that a higher number of AP's are required, but it doesn't explain how WiFi 5 is better than WiFi 6 though.

That goes against all the tests I've done. And I guarantee that equipment will not be upgraded if the existing equipment can be made to work substantially better first.

Of course, I've also never used stock Asus firmware for longer than it takes to flash RMerlin firmware. Nor have I used any GT router either.

Yes, I am a little skeptical here because my experience is very good with the two AX routers and the AiMesh AC routers I work with. Not that I am doubting your results in your home. Just that generalizing this for others may be short-changing them of the best possible WiFi in their homes.

Thank you for constructive feedback. I’d be truly curious to see your internet speed test result along with your AiMesh setup. I often came across on internet testing with 300 Mbps internet speed and say it’s perfect but that means internet speed is the bottle neck even on wifi 5. Others have a number that looks to go over 70% efficiency of wireless PHY. So I was truly confused. I kept taking the best number I see on internet and tried to buy new routers. So this time I finally conducted these test/experiment. Yes. it’s absolutely for my home environment setup. So it won’t generalize to everyone but hope it would be help or at least make aware such potential exist to some. As I was always doubting with over $1000 Wi-Fi 6 setup, there is no way any of Wi-Fi 5 can be better.

In order to validate my test without having expensive setup or test tools, I checked few expected things or common sense/facts and assessed if my test result have these trends. I didn’t put these detail on this forum. In summary, the results were within the expected range. So I concluded although it is far from perfect but a test with potential implication and may be some credibility. Also probably most can quickly do at their home if one so desire.

Custom firmware is certainly something I did not have. Last I checked I didn’t see ax11000 custom firmware.

As for 4 APs, I started one by one. First One ax11000 for couple weeks. Got second ax11000. I didn’t get ax92 pairs for at least several months after. Same thing happened for Orbi started two units and eventually got to 9000sqft worth devices. I think it’s my old home construction plus limited location of Ethernet jack.

I’m truly curious to see your numbers. The best number I got for my 2x2 MIMO WiFi 6 is little over 600 Mbps right next to router (5ft). I did some range test at 30ft, room next etc but these were secondary test.

I hope we are in agreement best home network system may not be as straight forward as buying the most expensive, latest technology or the highest spec systems.

P.S. I’ve added a disclaimer on opening post.

 

[Sachb]

It's strange, that for the IphoneX you're just able to achieve speeds of 450 -460 mbps. What channel are you using? Was your Aimesh configured properly?

I have an average setup which consists of Netgear R7800 ----> D-Link DAP 1860----> DIR-882, and achieve around 520 Mbps on my 2x2 devices such as LG G7+, Samsung S8+ etc. All of them achieve a link rate of 866 Mbps.

Have you tried tweaking the settings such as Wireless?

 

[Sachb]

When I connect to the Netgear R7800 X4s directly I achieve speeds up to 532 Mbps on all my 2x2 clients, which is basically maxing out my 500 Mbps connection.

 

[Tech Focus]

It's strange, that for the IphoneX you're just able to achieve speeds of 450 -460 mbps. What channel are you using? Was your Aimesh configured properly?

I have an average setup which consists of Netgear R7800 ----> D-Link DAP 1860----> DIR-882, and achieve around 520 Mbps on my 2x2 devices such as LG G7+, Samsung S8+ etc. All of them achieve a link rate of 866 Mbps.

Have you tried tweaking the settings such as Wireless?

That's a good number to know. Since your setup is also WiFi 5 based and getting up to what' expected range, I interpret that as yet another example of efficient (perhaps simply more mature) WiFi 5 technology can compensate/outperform still inefficient/immature WiFi 6 system. As far as AiMesh set up I had, I know one time I had MU-MIMO, explicit beam forming and airtime fairness turned off for stability issue based on comments from this forum. But I believe I eventually turned back on MU-MIMO and beam forming (though can't remember if turned back off again). But my first test was really meant for client comparison.

Other than above, setting change/no change, I didn't do any special tweak i.e. no explicit channel selecting etc. That's same for Unifi AP. Having said this, there is variability on absolute number for these test even I tried to ensure no streaming or gaming taking place while I was performing this test. The biggest swing I saw was iPad connected to Ax11000 AiMesh Node, one time I got over 500 Mbps average. But all the other occasions, I get 400 Mbps and sometimes even upper 300 Mbps. So I really tried to look relative numbers more than absolute number for these test i.e. with same setting/environment how one variable affect another.

 

[Sachb]

That's a good number to know. Since your setup is also WiFi 5 based and getting up to what' expected range, I interpret that as yet another example of efficient (perhaps simply more mature) WiFi 5 technology can compensate/outperform still inefficient/immature WiFi 6 system. As far as AiMesh set up I had, I know one time I had MU-MIMO, explicit beam forming and airtime fairness turned off for stability issue based on comments from this forum. But I believe I eventually turned back on MU-MIMO and beam forming (though can't remember if turned back off again). But my first test was really meant for client comparison.

Other than above, setting change/no change, I didn't do any special tweak i.e. no explicit channel selecting etc. That's same for Unifi AP. Having said this, there is variability on absolute number for these test even I tried to ensure no streaming or gaming taking place while I was performing this test. The biggest swing I saw was iPad connected to Ax11000 AiMesh Node, one time I got over 500 Mbps average. But all the other occasions, I get 400 Mbps and sometimes even upper 300 Mbps. So I really tried to look relative numbers more than absolute number for these test i.e. with same setting/environment how one variable affect another.

I think you didn't mention few thing such as security type and the wifi channel you're running these tests on.

Yes, Wifi 6 needs to mature.

Sent from my LM-G710 using Tapatalk

 

[Tech Focus]

I think you didn't mention few thing such as security type and the wifi channel you're running these tests on.

Yes, Wifi 6 needs to mature.

Sent from my LM-G710 using Tapatalk

I have already sold my AiMesh setup so I don't have those info anymore. But I had never explicitly set them so it was always auto setting. I know this is considered a bad habit from network enthusiasts; however, this is what a lazy network hobbyist does.

I do not like to dig into setting for variable parameter changes that may require me to configure every time the system reboots. Honestly, I had tried to do some tweaks before when I got Asus Ax11000 as I was excited. For example, at one point I tried changing RSSI to see if I can make roaming better on AiMesh system, but I was NOT good at it. It really didn't help. So I've decided to trust and just go with default manufacture setting understanding it is suboptimal.

As far as security goes, I did have Ai Protection turned on but I didn't do any extra tweaking or setting change. I do not recall if Asus had explicit way of changing IDS or IPS feature, Deep Packet scanning. At least I confirmed at the router, I had 940 Gbps download internet speed via ethernet so I assumed whatever the process taking process on the back, it did not bottleneck my internet throughput.

UPDATE:

I've decided to give my new tool a try (UniFi Controller) and scanned channel. I had a suspicion but using Netspot confirms that I think previous owner left old n/g Netgear Wifi router/access point in the middle of house, which I can NOT find. It uses channel 6 and currently UniFi 2.4 GHz is set to channel 6. I assumed AUTO setting will pick the best channel unless dynamically surrounding are changing, but I can't imagine this hidden router changing its channel, so UniFi must have picked channel 6 without scanning. I didn't try WiFi AI (UniFi way of auto scan at certain time and switch channel). One of AP is seeing the interference. So I changed it to channel 1 as that looks wide open.

Is it common practice or preferable to use separate channel for each AP on 2.4 GHz band single SSID network? It looks like 5GHz chose separate channels for each AP.

 

[Sachb]

I have already sold my AiMesh setup so I don't have those info anymore. But I had never explicitly set them so it was always auto setting. I know this is considered a bad habit from network enthusiasts; however, this is what a lazy network hobbyist does.

I do not like to dig into setting for variable parameter changes that may require me to configure every time the system reboots. Honestly, I had tried to do some tweaks before when I got Asus Ax11000 as I was excited. For example, at one point I tried changing RSSI to see if I can make roaming better on AiMesh system, but I was NOT good at it. It really didn't help. So I've decided to trust and just go with default manufacture setting understanding it is suboptimal.

As far as security goes, I did have Ai Protection turned on but I didn't do any extra tweaking or setting change. I do not recall if Asus had explicit way of changing IDS or IPS feature, Deep Packet scanning. At least I confirmed at the router, I had 940 Gbps download internet speed via ethernet so I assumed whatever the process taking process on the back, it did not bottleneck my internet throughput.

UPDATE:

I've decided to give my new tool a try (UniFi Controller) and scanned channel. I had a suspicion but using Netspot confirms that I think previous owner left old n/g Netgear Wifi router/access point in the middle of house, which I can NOT find. It uses channel 6 and currently UniFi 2.4 GHz is set to channel 6. I assumed AUTO setting will pick the best channel unless dynamically surrounding are changing, but I can't imagine this hidden router changing its channel, so UniFi must have picked channel 6 without scanning. I didn't try WiFi AI (UniFi way of auto scan at certain time and switch channel). One of AP is seeing the interference. So I changed it to channel 1 as that looks wide open.

Is it common practice or preferable to use separate channel for each AP on 2.4 GHz band single SSID network? It looks like 5GHz chose separate channels for each AP.

I use seperate SSID for both the 2.4 ghz and 5 ghz. This helps you know which band you're connected to, eventhough I have smart connect option, I chose to create seperates for it.



Sent from my LM-G710 using Tapatalk

 

[mjc775]

Interesting research- thanks for taking the time to post it.

I recently upgraded from a Netgear R7000 to RAX80. In our 2400 sq ft 1990s era wood construction home, the AP is upstairs on one end of the house while the family room, kitchen and small backyard is on the lower level on the opposite end.

In summary, my experiments have shown that a clear WiFi channel made lots of difference in speed and reliability. With the help of WiFi Analyzer on my Android phone, I was able to find one - on 5 GHz DFS Channel 104. On my 400/20 Mbps Internet connection, I’m able to achieve 250-400 all over my property. I purposely am only using an 80 MHz channel, since my limited internet speed wouldn’t benefit from a 160 MHz channel. While experimenting I was kicked off the lower DFS channels in the 50-60 range. Unfortunately my Dish Hopper 3 doesn’t recognize DFS channels, but my MBP and iPhones do.

Since there is overlap amongst WiFi channels, when experimenting and selecting a “clear” non-DFS channel with a neighboring network on a different yet nearby channel, the speed would degrade most noticeably the further from the AP I was.

Right now I’m happy with my WiFi and will be holding off on purchasing new equipment until WiFi 6E’s 6 GHz channels are available on an AP and iPhone.


Sent from my iPhone using Tapatalk

 

[Tech Focus]

Right now I’m happy with my WiFi and will be holding off on purchasing new equipment until WiFi 6E’s 6 GHz channels are available on an AP and iPhone.


Sent from my iPhone using Tapatalk

Thank you for sharing your experience as well. At this point, I agree waiting for 6E seems a way to go.

 

[L&LD]

@Tech Focus (replying to your post: 8).

% efficiency isn't what makes a network functional (or 'worth it') for myself or my customers. Lowest latency is what matters most, for most. When a given WiFi environment is optimized fully with the budget and hardware available, then the % efficiency is a side story as it is a variable with too many dependencies to worry about.

We fully agree that home WiFi isn't straightforward. And apologies for taking so long to respond back, just very busy the last few days and I wanted to answer this as fully as possible. I use an RT-AX88U with a 1Gbps symmetrical up/down ISP connection.

I can experience higher speeds in my home with a single, well placed, RT-AX88U than with an RT-AX58U in wireless AiMesh mode. Yes, the second wireless AP makes things worse for AC and AX clients.

The best use of the second router in my home was in AP mode (wired) which not only saturated the wireless signal nicely but also gave me a 4 Port switch at the other end of the house too and also allowed me the benefit of choosing which to connect to (with unique SSID's, control channels and channel widths).

For this particular customer with a very large family, the story may sound similar to many. They originally had an 80/15 Mbps d/u ISP connection when I first met them and because they expected to have 1Gbps service sooner than later, they opted for an RT-AC86U over an RT-AC3100 per my recommendations. Things were greatly improved over the ISP supplied router and wireless 'extenders' they had littered all over the home (3 of them). The home had Ethernet runs at good/strategic locations, but an additional router/AP at that time wasn't deemed cost-effective for them (and the performance the single 'AC86U provided backed this up when compared to their ISP service plan limits). This network ran with AiProtection, QoS, and a few choice scripts (FreshJR QOS, Skynet, Diversion) with minimal maintenance for almost a year.

Then things changed. Their two eldest were to move out and, the home could get 1Gbps service finally. After the new ISP service was installed, the RT-AC86U seemed to be overwhelmed, but in reality, what they were seeing was just the difference from a 'great' connection vs. just an 'okay' one (depending on where they were in the home and what mobile, handheld client device was being used then). And that made it look like the 'AC86U wasn't performing as it should with their AC clients. They asked what could make it better and I mentioned the 2x 'AC86U AiMesh network that worked very well for another customer. Then they asked what I had.

The RT-AX88U was tested in their home and (of course) was found to be superior to the 'AC86U. They kept it.

Then things changed again a few months later. The two eldest moved back in and brought with them a whole lot more devices than they had when they moved out.

Using a device from the bedrooms or the garden (furthest/most obstacles from the main router when it was at its original location) caused the 1Gbps service effectively to be worse than what they had previously when everyone was at home (and now, everyone was). Some of this was caused because QoS was not used now. Some issues were because the usage of the network had changed too. For example, the 'main areas' had morphed from the living room/great room to the bedrooms/garden being just as important, and even more so, when the number of devices was considered and their load on the network. Particularly with a newly built greenhouse that 'needed' WiFi internet to make it greener, the rest of the devices were fighting tooth and nail for their time-slice of the WiFi pie they demanded too.

For this customer, the added AP in wired AiMesh node made a world of difference (and they still have no AX devices themselves today). The RT-AX58U was also cheaper than an RT-AC86U for them and I believe it provides a better match to their AX main router too. If this 'science experiment' didn't work (I had it tested in my home to make sure it would though), the RT-AC86U would have been the next 'top' choice for an AiMesh node for them.

With a three-level home at just under 4100 SqFt, the RT-AX58U was what pushed the 1Gbps service they had to acceptable levels for all areas with the RT-AX88U as the main router too. With their AC class client devices, sustained 500Mbps throughput, and better is achieved everywhere they need it most. Including the back garden. For up to 1Gbps total and two clients simultaneously.

In reality, no two clients need over 500Mbps sustained throughput indefinitely, of course. The 1Gbps total throughput could now be shared with a dozen or more clients simultaneously throughout the home and each could still have more effective speeds than with their previous ISP connection on just a single client and the previous RT-AC86U from any normal distance. This improvement was more than they were expecting.

With my AX equipped laptop, the performance average in their home was close to 750Mbps when the routers were set to 160MHz channel width (about two dozen measurements in 6 regular 'living areas'. The kids noticed that their shows (two or three streams) kept playing while I was doing 'my' tests too.

The routers were placed with the main one on the first floor (walkout basement with living room above it) and the second one on the second floor and about 80' away from the main router (above the second router is the bedrooms) and also provided the back yard coverage too. This placement was different than when the single router was used (I moved it away from the other router) and the third floor (also wired) was tested but provided worse results overall. The final settings were 80MHz bandwidth for 5GHz in AX mode (no issues noted with their client devices and they were then ready for new devices at any time), separate (and new) SSID's, and a slight reduction of the power output from the main router to 'good', in addition to the M&M Config suggestions too, of course.

AP (separate SSID's) mode was also tested at their location, but they preferred the 'ease' of AiMesh instead, even though AP mode was noticeably better to me in-between the two routers, as I could choose which to connect to with my AX capable laptop.

I almost never test at 'next to router' distances as most installations will never be used like that anyway, not to mention that the results are not indicative of what the router and/or client device are really capable of, real-world. This test is usually deceiving as most older routers show much better results 'up close' but usually have the worse performance at normal/greater ranges than newer routers do. I believe (and I agree) that newer routers are built and designed to give the best-balanced performance, and not just 'wow' numbers at unrealistic distances to the detriment of the rest. At least the RMerlin powered Asus routers I work with.

The 5 main areas (living room, great room, kitchen/sitting area, bedrooms, and garden) were mostly/directly served by the two routers and gave more than the performance increase expected. The ancillary areas were better served overall, but now, the difference from best to good enough was even more dramatic, if even more usable than before. (10x on AC clients and closer to 40x on my AX laptop).

I am not sold on today's so-called 'TriBand' routers for use as wireless AP's/AiMesh nodes for the results stated above. Not only do they pollute the RF bands needlessly, but they also do so without giving the results they should either (in most cases).

With fixed control channels, bandwidth, and separate SSID's for each RF band in a wired AiMesh configuration, I left them with the idea that the network may become even more fluid in time. This may happen if the surrounding AP's are using Auto channels/bandwidth and they eventually decide to stay away from the channels they now use.

I also want to apologize for omitting the links I meant to include in my original post above. Oops!

(This is what is available in the link in my signature, browse at your pleasure, but the RT-AX88U upgrade and the RT-AC3100 Report may have more info you're asking for).

M&M Config https://www.snbforums.com/threads/n...l-and-manual-configuration.27115/#post-205573

Sanitize Network https://www.snbforums.com/threads/rt-ac66u-slow-wan-to-lan.12973/page-3#post-269410

Control Channel Set up https://www.snbforums.com/threads/a...details-in-the-description.55582/#post-472051

RT-AC3100 Report https://www.snbforums.com/threads/s...-go-with-the-rt-ac1900p-v3.34748/#post-281391

Nuclear Reset https://www.snbforums.com/threads/major-issues-w-rt-ac86u.56342/page-4#post-495710

RT-AX88U Upgrade https://www.snbforums.com/threads/b...ta-is-now-available.60037/page-31#post-531024

Why a new SSID? https://www.snbforums.com/threads/i...-ax88u-rt-ac86u-node.60551/page-2#post-532915

 

[Sachb]

@Tech Focus (replying to your post: 8).

I can experience higher speeds in my home with a single, well placed, RT-AX88U than with an RT-AX58U in wireless AiMesh mode. Yes, the second wireless AP makes things worse for AC and AX clients.

The best use of the second router in my home was in AP mode (wired) which not only saturated the wireless signal nicely but also gave me a 4 Port switch at the other end of the house too and also allowed me the benefit of choosing which to connect to (with unique SSID's, control channels and channel widths).

This also applies to Wifi Extenders, people always complain of extenders being slow etc. The moment you turn your Wifi Extender to an AP, the real benefit comes into place.

Although I don't have a huge apartment, just a normal 1700 sq ft 2 bhk Home with Thick concrete walls, I do get speeds around 520 mbps throughout the house, but not always.

Then the added signal indicator on the D-link DAP 1860 helps me get close to what my Main Router is providing. So speed loss is minimum across the house. 3 green lights and I'm good to go.

 

[psychopomp1]

OP, I suggest you try testing any client with an Intel AX200 chipset (wifi 6 2x2). I have such a PCI card fitted in my desktop PC (TP Link TX3000E) - PC is upstairs and router (Netgear RAX200) is downstairs. The TP Link wifi 6 card connects at the full 2.4 Gbps to the router with real world throughput of around 1 Gbps. No way on earth would i be getting that over wifi 5.

 

[K-2SO]

No way on earth would i be getting that over wifi 5.

I can do Gigabit with my 2 RT-AC86U. Router and Bridge. And ~50% cheaper than your RAX200 alone.