Question A few Questions about AC and AX Standards and Throughput

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
Hello,

Currently using an 8-year old router bridged with ISP's modem with 200/5 bandwidth.
My home network approximate topology:



Want to understand a few things about newest routers:

1 - Throughput

I saw 802.11ac wireless routers with:

2.5GHz frequency and bandwidth ranging from 300Mbps to 600Mbps
5GHz frequency and bandwidth ranging from 867Mbps to1733Mbps

Is difference noticeable between 2.5GHz 300Mbps and 2.5GHz 600Mbps that it would
be worth getting less expensive router (2.5GHz 600Mbps, 5GHz 1300Mbps) than
(2.5GHz 300Mbps, 5GHz 1733Mbps) one provided 2.5GHz works better at home?

I saw 802.11ax wireless routers with 5GHz with bandwidth of up to 3000Mbps which
people claim are faster than 802.11ac standard ones. But all I heard is that AX standard
is designed for more dense spaces like public Wi-Fi, hotels renting internet and airports
with many devices for simultaneous connections. AX is also more power efficient for IoT
devices as well.

Is it true that assuming my tablets and smartphones are AX compatible, I will get improved
throughput?

2 - Antennas

I saw AC an AX routers with 4 and even 6 antennas with 2 x 2.5GHz ones and 2 x 5GHz ones.
Also there was device with 2 x 2.5GHz, 2 x 5GHz and 2 x 2.5GHz + 5GHz combined. Is it the more
antennas, the better? Why would antenna have 2.5GHz + 5GHz as well?

People told me that for cheaper than a price of TP-Link Archer C80 AC router locally I could buy newest Huawei AX or Xiaomi Redmi AX router from China. Yet that routers from China don't have English web interface and warranty will be an issue too.
 
Last edited:
First and most important is you end device is half the connection. Buying a fancy router will not change the wifi support in the end device. The most common end devices,especially portable ones, only support a wifi encoding method that would be the same as a router that has a 1200 number on the box.

Most the stuff you see on routers is all marketing that makes people thing more is always better like when you can see more antenna.

The end device is also responsible for most problems with coverage. Unlike a router where most transmit at the full legal power most end devices transmit at lower power to save on battery or have a small size.

Wifi6 (ie 802.11ax) has been a huge disappointment. In real world test is does not perform much better than the older 802.11ac. The problem is the limited radio bandwidth on 5g. To get the highest speeds wifi6 need to run on a 160mhz radio band. The only way for them to do this is overlap radio bands that are used by things like weather radar. Because of the complexity of the rules with detecting weather radar a lot of device choose to only support 80mhz bands which is the same as 802.11ac. In addition using 160mhz makes the overlaping of neighbors wifi signals so you get even more interference.

The other huge issue with wifi6 is to get very high speeds it uses QAM1024. This very dense data encoding needs almost a perfect signal so it does not work well far from the router. Many people say it only works in the same room.

So most people that went out and bought wifi6 and replaced their end devices saw only small increases in the speed when you would have though the speed should almost double over 802.11ac.

The so called solution is wifi6e. This functions on the new 6ghz radio band. There is a lot more bandwidth there so it should be easy to get 160mhz blocks that do not interfere. wifi6e is still very new and expensive. Not a lot of end devices support it yet.


The archer c80 is good router and will be much faster than your current one......again if you also have very old end devices it will make no difference.

The archer c80 is using 3x3 mimo to get the higher speed. Your end device likely only have 2 antenna but if they have 3 they can use this extra feature. The way it implements the so called 600 speed on is not supported by a lot of end devices. It would be better called 450 and you only get that if the end device has 3 antenna you get only 300 if it has 2. And most important to remember is these represent some magic data encoding speed that you can never get, if you are very lucky you might get 1/4 to 1/2

Hard to say about devices from china in general. Many have very poor support. If you buy older technology ie 802.11ac rather than wifi6 then most the bugs have been worked out. If cost is a issue you might consider a router that only has a 1200 number on it. It all depends on if your end devices can support faster.
Something like a archer a6 is about 1/2 the price of archer c80 here. Key things to watch for is that it has gigabit lan and wan ports if you internet services is more than 100mbps.
 
Reactions: Boris_yo

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
The archer c80 is using 3x3 mimo to get the higher speed. Your end device likely only have 2 antenna but if they have 3 they can use this extra feature. The way it implements the so called 600 speed on is not supported by a lot of end devices. It would be better called 450 and you only get that if the end device has 3 antenna you get only 300 if it has 2. And most important to remember is these represent some magic data encoding speed that you can never get, if you are very lucky you might get 1/4 to 1/2
How to check for certain how many antennas my smartphone and tablet have? Through technical specifications sheet?

What if I connect through 5Ghz though? If one antenna on 2.5Ghz channel bandwidth is capable of 150Mbps throughput and 2 antennas 300Mbps throughput, how much would that be on 5GHz channel connection?

Hard to say about devices from china in general. Many have very poor support. If you buy older technology ie 802.11ac rather than wifi6 then most the bugs have been worked out. If cost is a issue you might consider a router that only has a 1200 number on it. It all depends on if your end devices can support faster.
Something like a archer a6 is about 1/2 the price of archer c80 here. Key things to watch for is that it has gigabit lan and wan ports if you internet services is more than 100mbps.
I get 50Mbps (1/6) through my Edimax 300Mbps 2.5GHz router on smartphone and tablet (both support 802.11ac). By the way, I get same speed through ISP's 2.5GHz 300Mbps modem. If I buy 1200Mbps router (867 Mbps through 5GHz channel + 300 Mbps through 2.5GHz channel) will I be getting same throughput ratio assuming the channel used is 5GHz? Which comes to 145 Mbps more or less with 867 Mbps 5GHz? If that will be the case, should I go with higher 5GHz speeds like Archer C80 1900 Mbps? (1300 Mbps through 5Ghz and 600 Mbps through 2.5GHz) provided tablet and smartphone can go above 145 Mbps?
 
The devices might tell you how many antenna. You can to a point figure this out by looking at the speed numbers. The so called 867 is not really a speed it is the encoding rate and with that number you can tell how many antenna/mimo channels it uses as well as the channel width etc.
It tends to be fairly messy but the numbers can tell you a lot when you learn to read the tables.

https://docs.google.com/spreadsheets/d/e/2PACX-1vQXoEYLGWrR1aGyGaTXOOaDQSPLfeC4rv70KRFuRP6eZ5fL-Ku_YI6DgS6zZMNyIhQpQmnKQ1O7abij/pubhtml?gid=1367372895&single=true

It might be easier to see if the specs on the device tell you.

In general you get a higher percentage on simpler encodings. The more complex the data encoding method is the less percentage you get. I would suspect you will not get much over 150mbps even if you use the 1300 data encoding on the 5g radio band.

What is important to remember is for portable devices you really don't need huge amounts of transfer speed. It is not like you can download a 60gbyte game into your phone. When you look at the more common things like say streaming even 4k netflix it uses only about 25mbps.
Too many people get hung up on speedtest numbers when it really doesn't matter once you get above a certain point. Kinda why buying a internet plan that has a bigger rate is just a waste of money for some people. The only things that a larger rate really helps is large file downloads.
All depends how much downloading you do in a month and how much total time you save. I mean it is really cool to watch a game download in 2 minutes rather than 5 but does it really matter in the big picture.
 
Reactions: Boris_yo

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
In general you get a higher percentage on simpler encodings. The more complex the data encoding method is the less percentage you get. I would suspect you will not get much over 150mbps even if you use the 1300 data encoding on the 5g radio band.
Hello. I read that Xiaomi 4A Gigabit router although has 4 antennas and 2 streams is weaker than TP-Link Archer C80 with 4 antennas and 3 streams when it comes to distance on 5GHz channel. What makes former slower than the latter? MIMO is those streams by the way or something different?

I checked portable devices used at home but could not find detailed wireless component's information. I don't believe device manufacturers allow access to detailed technical specifications sheet. Nowhere is encoding rate mentioned so I don't have necessary information to track it back to the number of antennae used in them. I was just curious for a challenge to see it myself using the table you provided.

  • Samsung Galaxy A71 Wi-Fi specs and I only found this: 802.11 a/b/g/n/ac 2.4G+5GHz, VHT80

  • I checked also Samsung Galaxy S6 Lite Wi-Fi and only found this: 802.11 a/b/g/n/ac 2.4G+5GHz, VHT80 MIMO

  • Here is Wi-Fi specs for Samsung Galaxy A5 2017: 802.11 a/b/g/n/ac 2.4+5GHz

  • Here is Wi-Fi specs for iPad 9th Gen 2021: Wi-Fi (802.11a/b/g/n/ac); dual band (2.4GHz and 5GHz); HT80 with MIMO

  • He is Wi-Fi specs for my old iPad Mini 2: Wi-Fi (802.11a/b/g/n); dual channel (2.4GHz and 5GHz) and MIMO
 
Wifi specs on mobile devices tend to be a challenge sometime. I don't know why they don't more clearly state the maximum speed like most pc or routers devices do. Some vendors do and other do not. I suspect it is because the end consumer of these devices does not have this as a priority.

So to some extent those do tell you. The stuff with HT80 means it can support up to 80mhz channels. This is pretty standard for 802.11ac, that is kinda what was the major difference but very early generation stuff could only do 40mhz channels like 802.11n
The key question would be what does "mimo" really mean. It should mean it has at least 2 antenna but it might have more. I suspect they would say something if it does. Most end device only have 2 at most and it is not uncommon for it to have only 1.

High wifi transfer speed does not seem to be a key concern on portable devices. Kinda makes sense since you really only need very high speed to download huge files and there is very little storage in a portable device. Mostly portable devices are used for consuming online content that is downloaded much more slowly like say netflix or web surfing.

You have to be very very careful about combining the concept of distance and speed. This is what a lot of the router manufactures and some of the so called professional "reviewers" do. There is no consistent way to qualify the tests so everyone does it different, using the method that makes their product look the best.
So one person may test saying anything less that 500mbps is unusable and another saying anything less than 10mbps is unusable. The only actual test is to measure the energy the radio puts out in DB. This is what the FCC tests and unsurprisingly every router is very near the legal maximum.

A example would in the old days if I transmit voice over radio it would go much less distance than if I used morse code. So which if those has less distance....kinda depends on how you ask the question. I mean a message that is delivered via morse code is much much fast than a voice one where you can not detect the voice at all.

Number of physical antenna on devices is mostly for show. Some routers like the xiaomi router use seperate antenna for 2.4 and 5. This tends to be unnecessary since 5ghz is almost a direct multiple of 2 times the 2.4ghz. So it is easy to design a antenna that optimum for both frequencies.
Other devices do silly stuff like use different antenna for transmit and receive. Mostly this is to just to play the marketing game "more is better" to the end consumer. When engineered correctly you only need 1 antenna per mimo stream no matter how many radio chips you hook to it. I don't know why the archer has 4 when it uses
3x3 mimo. It likely is something to with the MU-mimo stuff. Mu-mimo is another messy topic that tends to be mostly marketing hype and mostly is something they test in labs rather than provides a significant benefit to end users. Again most end devices do not support mu-mimo.
 

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
Number of physical antenna on devices is mostly for show. Some routers like the xiaomi router use seperate antenna for 2.4 and 5. This tends to be unnecessary since 5ghz is almost a direct multiple of 2 times the 2.4ghz. So it is easy to design a antenna that optimum for both frequencies.
Does that mean that Archer C80 using all 4 antennas in 5GHz band will be more effective at higher range than Xiaomi using only 2 antennas on 5GHz band (while other 2 antennas on 2.5GHz band being idle)? Also Archer C80 has beamforming technology that Xiaomi doesn't.
 
Again the number of antenna have nothing to do with how far the signal goes. Mimo means they are transmitting 2 different signals one on each antenna. They hope they can combine the 2 signals on the far end to get more bandwidth. It is just 2 different radio transmissions it does not affect the distance a each signal goes.

The concept of beam forming is fine when you can engineer it for a specific problem. I mean they have done this for years with tv antenna towers to feed the signals down say a valley.

It is mostly marketing hype when you consider a home wifi environment. I mean the most direct path might be directly through a wall that absorbs a lot of the signal. Most wifi in a home environment is bouncing around off walls etc. You many times will see huge difference in the signal level if you open and close doors since the signal bounces around. This random bouncing of the signals is kinda what makes mimo work. The 2 or more different radio transmissions taking slightly different paths and arriving at slightly different times is how the devices can separate the signals.

When you use beam forming and mimo you are forcing the signals to follow the same path which reduce the ability to separate the overlapping signals. If you look at outdoor wireless bridge equipment which uses directional antenna instead of beam forming to send the signal long distance they solve the mimo issue by mounting
the 2 directional antenna so the signal is 90 degrees from each other. So they send 1 signals with horizontal and the other vertical. This allows the signals to be more easily separated on the far end.

But again beam forming does not matter if the end device does not also support it. The beam forming may make it so the end device can hear the router better but it still must transmit its signal back and if that signal will likely not follow the same path as the one from the router.
 
Reactions: Boris_yo

gggplaya

Distinguished
But again beam forming does not matter if the end device does not also support it. The beam forming may make it so the end device can hear the router better but it still must transmit its signal back and if that signal will likely not follow the same path as the one from the router.
This is exactly why I think beamforming causes more problems than it's worth. Client devices believe the signal strength is strong than it actually is, and turn down the gain on their wifi transceivers. They then tend to hang while waiting for a response, but in reality the router never received the signal in the first place, so the client device tries a few more times and finally the router receives the signal and responds back. I've seen this happen numerous times and solved it by turning off beamforming.

Beamforming will cause the client device to show a much higher signal strength. So that may be why people think it's better using their wifi analyzer apps. But as I said, it can cause more problems than it's worth. Which is why I turn it off on all my devices and friend's devices.
 
Reactions: Boris_yo

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
Beamforming will cause the client device to show a much higher signal strength. So that may be why people think it's better using their wifi analyzer apps. But as I said, it can cause more problems than it's worth. Which is why I turn it off on all my devices and friend's devices.
How to check if my Android and iOS phone and tablet have beamforming?
 
Good luck with that. Apple tends to not implement technology quickly and even when it does it is hard to find very detailed information about which exact chips they are using in particular devices. Android is too generic a term some vendors you can find a lot of information about the hardware used and other tell you pretty much nothing.

In the end I strongly suspect if you do manage to get end device that support beam forming and you test with it turned on and turned off you will not find a significant difference. The problem is there is no good way to test this. The so called laboratory control test do not mean much because they don't represent a real life install. You also have the reverse with all the idiots on youtube testing it in their house and trying to say that somehow allows them to rank one router over another.
 

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
The end device is also responsible for most problems with coverage. Unlike a router where most transmit at the full legal power most end devices transmit at lower power to save on battery or have a small size.
Hello. I wanted to ask you about modems/routers that ISPs lease to customers on a monthly basis (or sell) versus routers from known brands bought by customers.

It's been several times where I have been told to set ISP's modem/router in a bridge mode and buy my own router because the transfer rate difference is significant and signal strength is better.

But you mentioned that all routers transmit at full legal power. Why is there a noticeable difference in transfer rate and signal strength between ISP's modem/router and some branded router on the market?

Thanks
 

kanewolf

Titan
Moderator
Hello. I wanted to ask you about modems/routers that ISPs lease to customers on a monthly basis (or sell) versus routers from known brands bought by customers.

It's been several times where I have been told to set ISP's modem/router in a bridge mode and buy my own router because the transfer rate difference is significant and signal strength is better.

But you mentioned that all routers transmit at full legal power. Why is there a noticeable difference in transfer rate and signal strength between ISP's modem/router and some branded router on the market?

Thanks
The biggest difference between an ISP router and a personally owned router, IMO is the ability to run second source firmware. You just can't do that on an ISP device. So if there is a feature you want or a security patch you MUST have, then you may not get it with an ISP device.
Performance is usually a function of cost. If you are willing to spend $300 on a router, then you will probably get better performance than you will with an ISP device. If you spend <$100 your performance will probably be similar. WIFI performance is not something that is easily predicted. Your house, your neighbors, your client devices, even your country can impact WIFI performance. We can't easily predict how WIFI will work for you.
We can provide general guidelines, like get as close to the WIFI source as possible and move all stationary devices to a wired connection. That is how you improve your WIFI. It isn't getting a "magic" router.
For you original title question -- AC and AX performance will be the same for most people. Most client devices that people have are not AX. So the performance is equivalent to AC.
Wait for WIFI 6E (6Ghz) to become more available and get that as your router. That router with a 6E client WILL be a big improvement.
 
"It's been several times where I have been told to set ISP's modem/router in a bridge mode and buy my own router because the transfer rate difference is significant and signal strength is better. "

But WHO told you. Most times the people saying stuff are fools. They just google search find some post and repeat without any understanding. Then you get more idiots who find that post and repeat it again and again. You get massive amounts of wrong information being spread. It seems to mostly be gamers who seem to believe anything and they think they are experts on every subject because the can use google search.

There is very little scientific testing that you find on wifi stuff....well at least in so called reviews. You have reviewers who have degrees in journalism testing stuff. These type of people lack the training to setup proper testing. I mean a lot of them test it in their house and then think that results in their house can be translated to someone else's. This is the key problem with wifi. The interaction of the radio waves and the house is the key thing that causes difference but every house is different so there is no way to do any form of scientific comparison.

The only results you can trust you find in the FCC sites. The router manufacture is required to test in a certain way and submit the reports that are then made public. Now these reports are so detailed they can be very confusing to get useful data out of the massive documents.
I have read lots of these and what I find is the key numbers that show output power are all very similar. But the only way you really know for sure is to check the routers you care about. I am sure there are some that put out less power. I know the portable ones put out much less in many cases.

But even with the FCC reports it means very little. All you have is a set of data testing in a environment that is nothing like a house. It is pretty much a big open room where the wall absorb all the signals and nothing gets in or out.
It can be as minor as the spacing between the antenna between different routers is slightly different and the interaction between the antennas and the wall near the router is very different.

In the end there is no way to determine how well something will work in a particular house without actually trying it. Years ago when wifi chips and amplifiers were much more expensive you did find differences between very cheap ISP routers and the higher end consumer routers. Now days they have figure all the details out and there is no real cost savings to using say a lower power microwave amplifier chip.
 

Boris_yo

Distinguished
Feb 14, 2010
135
3
18,585
0
For you original title question -- AC and AX performance will be the same for most people. Most client devices that people have are not AX. So the performance is equivalent to AC.
Wait for WIFI 6E (6Ghz) to become more available and get that as your router. That router with a 6E client WILL be a big improvement.
bill001g also told me this before. I thought about ordering TP-Link Archer C80 but decided to order TP-Link Archer C6U AC1200. It has USB port for media, 4G modem and 4 times the ROM memory compared to the former. It's almost $20 cheaper than C80 though despite having more functionality.

But WHO told you. Most times the people saying stuff are fools. They just google search find some post and repeat without any understanding. Then you get more idiots who find that post and repeat it again and again. You get massive amounts of wrong information being spread.
Customers of other ISPs who noticed the difference when they switched modem to bridge mode and started using their private router. They hate that ISPs charge rental fee for their crappy modems when customer can save long-term buying their own private router. My ISP charged rental fee for modem and another rental fee for wireless home network. Yes, they charged me additional fee for basic wireless functionality.

I also heard that ISPs limit their wireless functionality of their modems depending on country and region. How true is that and what do they limit there I don't know.
 
The rules for output power and channels allowed are actually different in different countries. For example channel 13 is allowed in the UK but not in the USA. Some of the channels 5g channels can only be used at 200mw or even 20mw in some countries but can be used at 1000mw in the USA.

The people not liking to pay the extra fee I can see being the largest factor. What you will also find though is in some of the cases where you have no option but to keep the ISP box like fiber installations the boxes are better than most. ATT latest fiber modem/router that supports wifi6 uses the exact same wifi parts as the top of the line asus gt-ax-11000. It even has 5gbit ethernet. In fact almost every router that is wifi6 uses one of the 2 parts either from broadcom or qualcomm. There currently are no other chips that have fcc approval.

You can believe some random facebook user or you can do the research yourself. The FCC data base is public and you can get all kinds of good information including direct links to the fcc documents from sites like deviwiki
 
Reactions: Boris_yo

ASK THE COMMUNITY