MY experiences with over 2 weeks on Asus XT12 -my thoughts.

[mstar]

Hello all

Background

1. Only 1 XT12 no node. I am using it as a standalone router only.
2. Plugged into ISP (virginmedia) router with 1gig connection
3. Default firmware used for testing.
4. Default settings used except following: 1) split bands now i have 2.4, 5-1ghz, 5-2ghz (i don;t have any node so this is free for clients to connect)
5. 2.4ghz set to 20hz and auto channel selection everything else left as default. not including 12-13 channels
6. 5gz -1 and 5ghz 2 all defaults with auto channel selection Auto select channel including DFS channels.
7. 2.4ghz network is used mostly for IoT devices, doorbell, echo show. However when i go in the garden the range of 5ghz drops and i have to connect to 2.4ghz

1. 2.4ghz Wifi channel -auto or manual selection results.

For 1 week as L&LD mentioned in his threads ages ago. I put my Asus XT12 on auto channel selection. I noticed most traffic (neighbour on auto it seems), seems to be on 1,6,11. I never seen any network on any other channel.
My XT12 picks a variety of channels e.g 1,6,7,11,9. It seems my router likes channel 7 most days which is surprising as the strongest channel next to me is on channel 11. Most on 1,6 channels. only 1 (strongest) network is on 11.
I noticed no degrade of connection the download and latency seem VERY similar not sure as most of these networks are distance away it doesn't matter.

Question i had is i like to be on channel 11, one week i noticed HARDLY no network goes on this channel -except 1 network who never moves from this channel. I did some investigations its my next door neighbour!!! Ideally i like to shift this network so i get this 11 channel myself as it seems most are on 1,6 channels. If i manually put my network on 11 and just camp on that channel for days. Will neighbours network move? as we both strong networks by dbm. I am not even sure they are on auto channel selection as it seems they camp on channel 11 for over 1 week. For now i will manually set to 1 as it has the lowest signal strength networks (4 of them). I could put on 13 but not sure it interfere with channel 11 network who is m y neighbour.


2. 5gz is solid. nobody in 1 week is on the auto selection channels i am on 5ghz just camping on channel 52, 5ghz-2 on channel 100. All other 5ghz networks are on 36 and 48.
It seems auto working good on here and i am getting better speed on 5gz-2 for some reason not sure why. I suspect as this is the dedicated backhaul connection (if i had a mesh node) so its faster?

3. I am surprised the range is VERY similar to my previous router (Sagemcom F@st 3896 virginmedia hub 5), maybe i was expecting a large jump in range but using signal meter i am only getting 4-10dBM more than my old router.
Hard to comment on speeds as wifi has so many factors. As L&LD said those websites for speedtests are not the honest true picture.

4. Placement of the router i was thinking the antennas go all directions. unfortunately as my ISP connect comes from side of the house under the stairwell most signal is wasted going outside (not to brickwall to neighbours house but to side driveway). and also under the 7th step of the stairway. I read from L&LD that as i have like 4.5 cm from the 7th step on the stairwell should i place it to get more clearance from the top of the router so it enables the wifi waves to form more to enable penetration? or its fine as it is? I cannot move this without the ISP moving the connection coming to my house :-(

5. Settings i used was from here but the settings based on L&LD and others i will change following but not sure what will happen:

• In the Wireless, Professional tab 2.4ghz,
  • Choose the 2.4GHz band and change the Preamble Type to Short. [i read this maybe make 2.4ghz faster but may cause device compatibility? most of my 2.4ghz devices are also 5ghz but i keep them on 2.4ghz)
  • Disable Universal Beamforming. (this was on in all routers i had not sure what happens this being off will it improve speed/reliability?? i assume keep explicit beamforming as enabled??)
• In the Wireless, Professional tab 5ghz,
  • Choose the 5GHz band and change Universal Beamforming to Disable.(this was on in all routers i had not sure what happens this being off will it improve speed/reliability??, i assume keep explicit beamforming as enabled??)
  • Make sure you hit Apply before moving away from this page.

6. I do not know to enable/disable dfs channels or enable 160 MHz, as i have no way to test this work, will devices struggle to connect if i enable this? as far as i can see as the crow flies there is a military airspace is 7.9miles away. Which isn;t used much but still active.

---

Overall maybe its not set up and i need to upgrade to latest firmware but not noticed a DRASTIC change in wifi performance but my next task is to get a node to create a mesh for my new upper floor where signal quality on 5ghz is around 43-52dBm

 

[Bhoot®]

Got AXE16000 main router + 3x XT12s in wireless aimesh. I also thought about pro settings but just left them mostly to default.
For closest node, average 920+Mbps up/down - very happy (Truenas Scale VM + Win10 Plex VM). Mid-range node is 400Mbps (family iPads). Last node, need to setup again as I was testing out stuff.

 

[L&LD]

@mstar,

Testing anything but the current firmware (and specifically, the fork of that firmware, you want to use) is just wasted effort for little (at best) gain. This exercise is best reserved if you're testing the hardware for failure, not if you're expecting to set up your network optimally (one and done).

1) I don't recommend Auto Channels (ever). I also don't recommend setting a manual channel other than 1, 6, or 11 by 'logic' either.

Only testing in your actual environment will determine the best, overall, Control Channel to use. Your testing should be conducted from multiple locations around the areas you want coverage, with a plugged-into AC power laptop and other client devices you may normally use at those locations. The fastest download speeds should not be the sole determination of this 'best channel', the lowest latency observed (over the entire area, and different client devices) should be. That is what gives you a fast, and optimized Control Channel and correspondingly great network experience.

Don't worry about how many APs are using the channel you're considering/testing (you test them all). Your router (and everyone else's) needs to sync with the rest of the routers they 'see', and then proceed and share those time slices with each other. When you're choosing between control channels like 2, 3, 4, 5, and, 7, 8, 9, 10, your router has potentially double the routers to sync with (because it will 'see' routers on CC 1 and CC 6, or CC 6 and CC 11). So even though the paper specs may indicate to you the best channel it is usually suboptimal (if not outright the worst one).

When you simply test (and keep good notes), with a variety of your client devices (preferably not just all handheld ones), in a variety of areas, and only with the standard Control Channels, the notes you kept will indicate which is the best channel overall (note: there will be no 'perfect' channel). This is because this method takes not just your router (and its tuning bias to certain channels), just your client devices (and their everchanging drivers), nor just your WiFi environment (which may be changing by the minute, depending on your location), but also takes into account non-WiFi interference too, which is explicitly 'ignored' by all the indicators you've relied on so far (but also/particularly WiFi 'apps', that not only give inaccurate suggestions (because of all the above, and also because they're not taking into account Channel Utilization percentage into their recommendations), but also because just using most/all of them on your network will skew (negatively) the results you will see otherwise).

The bottom line, WiFi isn't a science, so forget the theory when you want to apply it to your network. You need to roll up your sleeves, do some leg work, and get the settings (Control Channels, in this case) with the least compromises, overall, for the specific router, clients, and environment, for you.

2) The points I made above are all valid (except for the specific Control Channel numbers, of course) for all bands, including 5GHz bands and soon, the 6GHz bands too. Testing, taking notes of the settings with the least latency and the greatest speeds/coverage (in that order) will get you the best setup possible in your environment.

3) Range depends on too many factors (all of the above, including home construction materials, etc.), but what newer routers offer is much greater throughput than previous models to the same client device, at the same distance (to me, that indicates a greater range). At least, properly set up ones. Again, I wouldn't worry about the dBm numbers on their own (unless you have highly calibrated and expensive equipment to measure this with). The only thing that matters is real world results. And for range, this has already been achieved (for me, with the RT-AC3100, many, many years ago now).

4) I don't understand this point? You can't get a longer Ethernet cable to experiment with the router placement? The ISP equipment doesn't have to move. Just the router you're trying to optimize for the space it will be used in. Don't concern yourself too much with WiFi going 'outside' (it will anyway). Concern yourself with best coverage within your main areas instead. (Don't forget that WiFi signals will bounce from exterior walls and may enhance (or be detrimental) to range, speeds, and latency too. There is no 'science' here to help. Simply try different router locations, orientations, and a few antennae positions to dial in what works for you, now.

6) With a military air-base so close, do not enable DFS or 160MHz width on the 5GHz band. If you do, your network experience will only suffer.


What router did you upgrade from? What node are you considering for the upper floor? On its own, -43dBm to -52dBm signal strength is excellent (for one floor away).

If you're expecting to put one router above an other (on different floors), do not expect good results.

For best (wired) AiMesh placement, put the routers at diagonally opposite corners (further distance possible apart), and ideally, a floor (or two) away from each other too.

Hope the above is helpful to you.

 

[mstar]

Thanks L&LD, as always a great insight in this murky world of networking for me!

You are right on the points 1) I am going to camp on channel 1 for now. I am seeing good results so for now i will continue monitoring on this channel. Hopefully its reliabile and i wont get to much issues as there is a LOT of networks in my area.

2). Yes i am happy with my 5ghz results so for not i will leave this as is. Its tempting to tinker around but i am going to leave it as is.

3). Thanks i guess i was getting to fasinated with my signal meter comparing to my last router and not seeing a drastic jump in dBm. Your right in what your saying

4) Sorry to clarify my ISP modem is cabled to the ISP wall point and then the ISP modem is connected by ethernet to my XT12. I would get a longer cable yes, problem is wife hates my XT12 and doesnt want it on show as she "wants to look at facebook and shopping so doesn't care she only gets a fraction of 1gig connection" lol My next DIY task is to buy a bigger cable and move this away from under the stair well and compare results.

6) Asus had DFS channels set as default i probably search the disadvantage of disabling these but will i notice a jump in performance if i disable this? Also i was about to test the 160Mhz but not sure if this will cause issues with network or not. I guess i will try and see. But happy to understand if people noticed any changes in performance, reliability etc.


right next task is to upgrade to next stable firmware.

 

[L&LD]

4) Your wife's needs need to be balanced with your own needs. You get an equal vote too. If you're paying for a 1Gbps connection, at least one of you should enjoy it.

6) You shouldn't notice a jump in performance if DFS was disabled. What may be noticeable is that you'll see fewer WiFi interruptions/disconnects. 160MHz bandwidth may need DFS Channels, depending on what Control Channel was used (so, same possible issues as above). If you can set 160HMz bandwidth, and the environment and your client devices were both friendly to that change, you will see greatly increased throughput. Worth trying, IMO.

When (if) you upgrade to RMerlin firmware, be sure you start fresh once more to test it properly. Full reset, minimal and manual configuration. Don't assume what worked now will work then (test, verify).

 

[mstar]

4) Your wife's needs need to be balanced with your own needs. You get an equal vote too. If you're paying for a 1Gbps connection, at least one of you should enjoy it.

6) You shouldn't notice a jump in performance if DFS was disabled. What may be noticeable is that you'll see fewer WiFi interruptions/disconnects. 160MHz bandwidth may need DFS Channels, depending on what Control Channel was used (so, same possible issues as above). If you can set 160HMz bandwidth, and the environment and your client devices were both friendly to that change, you will see greatly increased throughput. Worth trying, IMO.

When (if) you upgrade to RMerlin firmware, be sure you start fresh once more to test it properly. Full reset, minimal and manual configuration. Don't assume what worked now will work then (test, verify).

Thanks L&LD, i should apologise i am in home network, for years i been using ISP equipment until now
I have actually switched on 160hz with DFS so i will monitor see if i get any dropouts/interupts. BTW in the asus app is there not a easier way to see if i get disconnects other than look in the general/wireless log and search for something like "disconnect"? i guess i am not sure i can see if they happen as i am not always at home and i am quite sure there isn't a alert in the asus app/web UI that tells you at x time, wifi disconnected etc ?

 

[L&LD]

You're welcome mstar. And no need to apologize! I was where you are a decade ago.

I'm not one to look for issues (in logs, or elsewhere). I just flip a setting and continue keeping my notes and comparing the network experience as a whole, to what I was experiencing before the change. The details don't really matter, the forest is what is important to see.

With a military airport so close by, I don't see how you won't get issues. But testing is fun too.

Enjoy it!

 

[mstar]

Yes i know i will monitor now. I have just set my channel bandwidth to 80mhz (160 enabled). So much info on here i spend most my time searching lol

Still unsure what manually to set the control channel so i set it to 52 (which is the auto setting channel), the neighbours are all on channel 48.
Could i benefit moving more away from this 52 channel to something like 60 or 64??

my 5ghz-2 is camping on for nearly 2 weeks on channel 100 (80mhz 160 enabled), i cannot see any network this far down the spectrum and this channel is the fastest.

 

[mstar]

Hello all some reason the router (auto) setting went to channel 36 (40mhz) where there is like 3 other networks on and 1 network on channel 48 (my neighbour), and my speed dropped from 800mbps to 350-410, so i put it on channel 52 (80 mhz) again, and speed went up again -i think i read why but didnt think it have that much effect on speed.

Also shall i move it MORE further away from channel 52 i know i am getting seriously out of my depth here and really do not know the impact

 

[L&LD]

Don't worry about other APs on your channel. Simply test each one, as comprehensively as you can, keep good notes, then chose the Control Channel that works the best for your client devices, router, and WiFi environment too.

Control Channel Setup 2021

Reset Mini Guide + Control Channel Setup Details

Control Channel Setup (more)

You do not need to use an 'app' or other program to see what APs are around you (and note that when you do, it will be detrimental for your network (i.e. the act of scanning for networks degrades your WiFi experience).

Just check each channel, use a standard 'test' for your environment, test not only for maximum download speeds, but better yet, lowest latency on the entire network too. Use the notes you do on the above to determine what the 'best' CC is for you. Note there is no 'perfect' CC, only one with the least compromises.