What's new

Starting to despair

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

It made sense 10-20 years ago.
As @drinkingbird wrote - set 20 MHz and auto, see what the router will choose most often and check how it works.
Yes, I've done that and I'll keep a check on it over the next few days. I'll soon know if there is a problem as my Canon network printer will either run slow when printing or the scanner will return an error message. I only have a few devices that use the 2.4 GHz band now, fortunately (printer/sat nav/ Amazon Kindle/ Honeywell evohome controller and one of my security cameras), but I'm glad that this hasn't turned out to be a hardware issue, even though I thought it was initially, hence the reason I sent my other router back for a refund.
 
the abysmal data transfer speeds on channel 11
Why do you need this channel? You checked, you know it doesn't work well. Use another (or Auto channel). Be happy, it works. That's all.
 
This is how I currently have the settings in the router:
View attachment 51731

And this is the result of the speed test I have just carried out. This is using my laptop in the room below where the router is mounted:


For comparison, this is what I am currently getting on the 5 GHz band (same location as the chart above):

Sorry for some reason I thought you had an AC88 (which runs 2.4Ghz at N speed). Disregard that part, what you're getting is very good for 2.4 still.
 
This is what I have been using to check local Wi-Fi channels. My 2.4 GHz channel is the one at the top of the list.

All that tells you is neighboring networks. Not channel utilization or interference. Your real world testing is telling you 11 is no good
 
When set to auto, the router only rescans when the wifi is bounced hence why I say you can have it reboot every night at 3AM or whenever so it rescans.
This is not quite accurate. When set to Auto acsd is constantly collecting information (noise, glitches, interference, etc.) for all channels to calculate a "channel switch candidate score" (except for DFS channels that include weather radar). Then approximately every 30 minutes (ACS_CSTIMER) it will decide whether an alternate channel's score has exceeded a certain threshold to be worth switching channels. This can be seen in the log extract below (I stole this from a different thread as it was a good example :)).

Code:
Dec 19 09:29:45 acsd: acs_set_chspec: 0x100b (11) for reason ACS_CSTIMER
Dec 19 09:29:45 acsd: eth6: selected_chspec is 100b (11)
Dec 19 09:29:45 acsd: eth6: Adjusted channel spec: 0x100b (11)
Dec 19 09:29:45 acsd: eth6: selected channel spec: 0x100b (11)
Dec 19 09:29:45 acsd: eth6: txop channel select: Performing CSA on chspec 0x100b
Dec 19 10:03:22 acsd: acs_set_chspec: 0x1008 (8) for reason ACS_CSTIMER
Dec 19 10:03:22 acsd: eth6: selected_chspec is 1008 (8)
Dec 19 10:03:22 acsd: eth6: Adjusted channel spec: 0x1008 (8)
Dec 19 10:03:22 acsd: eth6: selected channel spec: 0x1008 (8)
Dec 19 10:03:22 acsd: eth6: txop channel select: Performing CSA on chspec 0x1008
Dec 19 10:29:44 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER
Dec 19 11:03:04 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER
Dec 19 11:34:20 acsd: acs_set_chspec: 0x1007 (7) for reason ACS_CSTIMER
Dec 19 11:34:20 acsd: eth6: selected_chspec is 1007 (7)
Dec 19 11:34:20 acsd: eth6: Adjusted channel spec: 0x1007 (7)
Dec 19 11:34:20 acsd: eth6: selected channel spec: 0x1007 (7)
Dec 19 11:34:20 acsd: eth6: txop channel select: Performing CSA on chspec 0x1007
Dec 19 11:36:25 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER
Dec 19 12:43:03 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER

Things are slightly more complicated on 5GHz than they are on 2.4GHz because of DFS, radar events and 160MHz.
 
Last edited:
This is not quite accurate. When set to Auto acsd is constantly collecting information (noise, glitches, interference, etc.) for all channels (except DFS channels) to calculate a "channel switch candidate score". Then approximately every 30 minutes (ACS_CSTIMER) it will decide whether an alternate channel's score has exceeded a certain threshold to be worth switching channels. This can be seen in the log extract below (I stole this from a different thread as it was a good example :)).

Code:
Dec 19 09:29:45 acsd: acs_set_chspec: 0x100b (11) for reason ACS_CSTIMER
Dec 19 09:29:45 acsd: eth6: selected_chspec is 100b (11)
Dec 19 09:29:45 acsd: eth6: Adjusted channel spec: 0x100b (11)
Dec 19 09:29:45 acsd: eth6: selected channel spec: 0x100b (11)
Dec 19 09:29:45 acsd: eth6: txop channel select: Performing CSA on chspec 0x100b
Dec 19 10:03:22 acsd: acs_set_chspec: 0x1008 (8) for reason ACS_CSTIMER
Dec 19 10:03:22 acsd: eth6: selected_chspec is 1008 (8)
Dec 19 10:03:22 acsd: eth6: Adjusted channel spec: 0x1008 (8)
Dec 19 10:03:22 acsd: eth6: selected channel spec: 0x1008 (8)
Dec 19 10:03:22 acsd: eth6: txop channel select: Performing CSA on chspec 0x1008
Dec 19 10:29:44 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER
Dec 19 11:03:04 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER
Dec 19 11:34:20 acsd: acs_set_chspec: 0x1007 (7) for reason ACS_CSTIMER
Dec 19 11:34:20 acsd: eth6: selected_chspec is 1007 (7)
Dec 19 11:34:20 acsd: eth6: Adjusted channel spec: 0x1007 (7)
Dec 19 11:34:20 acsd: eth6: selected channel spec: 0x1007 (7)
Dec 19 11:34:20 acsd: eth6: txop channel select: Performing CSA on chspec 0x1007
Dec 19 11:36:25 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER
Dec 19 12:43:03 acsd: acs_set_chspec: 0xeb32 (48/160) for reason ACS_CSTIMER

Things are slightly more complicated on 5GHz than they are on 2.4GHz because of DFS, radar events and 160MHz.

Mine never changes unless I reboot or bounce the radios. Maybe due to being an older router. Quite possibly a new feature of AX (I'm on AC). I need to stop assuming everyone is as outdated as me 😄

So I guess no need to bounce it, as long as the algorithm works well and does in fact adjust channels as needed.
 
Yes, I've done that and I'll keep a check on it over the next few days. I'll soon know if there is a problem as my Canon network printer will either run slow when printing or the scanner will return an error message. I only have a few devices that use the 2.4 GHz band now, fortunately (printer/sat nav/ Amazon Kindle/ Honeywell evohome controller and one of my security cameras), but I'm glad that this hasn't turned out to be a hardware issue, even though I thought it was initially, hence the reason I sent my other router back for a refund.

As @ColinTaylor points out, your AX router should scan and adjust channels as needed when set to auto, so disregard my point about rebooting or bouncing anything.
 
Mine never changes unless I reboot or bounce the radios. Maybe due to being an older router. Quite possibly a new feature of AX (I'm on AC). I need to stop assuming everyone is as outdated as me 😄

So I guess no need to bounce it, as long as the algorithm works well and does in fact adjust channels as needed.
The same basic behaviour was present on my old RT-AC68U running John's fork. It was not obvious though because the RT-AC68U really liked sticking to the channel it was on (and then I started using a fixed channel). I thought for a long time that once it had chosen a channel at bootup it never changed, but that was not the case.
 
The same basic behaviour was present on my old RT-AC68U running John's fork. It was not obvious though because the RT-AC68U really liked sticking to the channel it was on (and then I started using a fixed channel). I thought for a long time that once it had chosen a channel at bootup it never changed, but that was not the case.

Odd, never seen any logs like that in my AC1900 but may not have it turned up high enough, or maybe it depends on the firmware, I'm running regular merlin. My Ubiquiti AP lets me select it as an option and it does change from time to time but never seen the Asus change. Probably only does it if there is something really severe I guess.
 
If you are using 40mhz channel, set it to 20 only. While it may be getting you a few extra megs right now, it is harming your neighbors and probably will end up reducing your performance at other times (in other words, it will be inconsistent, and a consistent 75 is better than a varying 5 to 90).

Kind of like the 1/6/11 rules - 20/40 MHz channels in 2.4 is another one...

The secondary channel has always been opportunistic since 802.11n was released - it's only used if it passes the clear channel assessment process first, otherwise it's primary/control channel only...

There was one of the devs for Google Fiber that worked on their gateways that did a study using info from their gateways at a scale that only Google could do, and their findings were...

Better to enable 40MHz channels than to not - and the takeaway there was that the control channel is always present, the secondary channel is available to be used, mostly because not everyone is using the airtime _at the same time_ - so doing 20MHz only was leaving potential bandwidth on the table...

@drinkingbird - you can respond to me privately, or look at the sticky I've posted some time back...
 
Might want to redact your IP address from those Speedtest pictures.
What could he be exposing himself to by publishing those IP addresses in public?
 
Kind of like the 1/6/11 rules - 20/40 MHz channels in 2.4 is another one...

The secondary channel has always been opportunistic since 802.11n was released - it's only used if it passes the clear channel assessment process first, otherwise it's primary/control channel only...

There was one of the devs for Google Fiber that worked on their gateways that did a study using info from their gateways at a scale that only Google could do, and their findings were...

Better to enable 40MHz channels than to not - and the takeaway there was that the control channel is always present, the secondary channel is available to be used, mostly because not everyone is using the airtime _at the same time_ - so doing 20MHz only was leaving potential bandwidth on the table...

@drinkingbird - you can respond to me privately, or look at the sticky I've posted some time back...

I still feel the 40mhz rule applies - differing opinions don't have to be kept private.

Sometimes theoretical testing and real world are vastly different (more than sometimes in my experience).

I've got two examples.

During COVID, when everyone fled NYC, a friend stayed in a huge (several hundred unit) new apartment complex outside of the city. Router was an older 2.4ghz only N150 that they owned. No matter how much tweaking and tuning I did, could not even sustain 10M, often falling to 1 or below or completely timing out on a speed test. Looking at inSSIDer there were dozens of very strong 2.4ghz networks and probably half of them were showing 40mhz. Sure, a few of those people may have set it that way intentionally to try and get more speed, but many were just on the default auto setting from the ISP (most likely all or most had the ISP's router) and obviously the band was not "clear" for 40mhz use, but those routers were picking it anyway. Most people didn't notice as the ISP router was dual band, but a few there were complaining that their older chromecast or other IOT stuff was not working.

In my local neighborhood, have been helping neighbors try to get better signal to their Blink cameras. The cameras are outside of windows and the AP inside. Signal strength shows excellent on both ends but throughput often can't keep up. Their TP-Link AP happily goes into 40mhz mode if I don't lock it to 20. That one may be valid since indoors there are several neighboring networks but not ridiculous, and the signals on most are relatively weak. Personally from looking at them, I would still think it is too saturated for 40mhz, but the router doesn't. However outside, where the cameras are mounted, those 40mhz networks are quite strong, along with 15-20 20mhz ones. This obviously impacts the cameras and is part of the issue. It also impacts me, my outdoor AP is 2.4ghz only, and over the past 5 years or so I've gone from being able to push 75 no problem to 20-30. So obviously there is congestion, but plenty of 40mhz out there (and believe me, I know who most of them are based on their network name and they're running the ISP router and don't even know it has a GUI, much less logged in and changed anything).

Some of this is that the decision is based on some threshold/algorithm that each company programs in, whether they are using reasonable settings or not depends on the person who decided on the numbers. The other part is that while two APs indoors or far away from each other may believe that there is clear space to use 40, but a device in between them, or outside, etc will be negatively impacted.

I can't think of a time I've fired up a scanner and not seen several 40mhz networks even in densely populated areas with lots of networks. So I still believe the "neighborly" thing to do is stick to 20, and in reality it is probably going to help you get more stability/consistency too.

I couldn't resist a couple good deals on security cameras during Prime Day, just want one aimed at the front door to alert me when a package is out there, and one on my driveway due to a few incidents in the area of unlocked cars getting stuff stolen (I lock my truck, but sometimes you accidentally unlock it, or maybe someone gets desperate and decides to start going for locked ones). I'm sure I'm going to have an aggravating time getting them placed with good enough signal and cursing those 40mhz neighbors.

Maybe if I set up a bunch of APs with 40mhz and really high power for a day or two all their routers will dump to 20mhz mode. They probably aren't even using anything 2.4.
 
Last edited:
What could he be exposing himself to by publishing those IP addresses in public?

Eh, depends how paranoid you are. You annoy someone here, maybe they try to hack you or flood you with data. Or some bot scraping IPs from text and images adds it to the list of ones to try to infect. Given some recent issues with Asus and vulnerabilities, probably not the worst idea to keep your IP (and the brand of router it is associated with) to yourself. Not singling out Asus either, scanning a subnet of IPs will get you plenty of responses, but it is much more valuable to know the brand and model of device (and even potentially firmware version) associated with that IP.
 
Why do you need this channel? You checked, you know it doesn't work well. Use another (or Auto channel). Be happy, it works. That's all.
Well, I want to be able to know for sure that it's not a problem with the router itself, seeing as it's brand new. I am auto-monitoring channel 11 every 10 minutes at the moment (which is currently working normally), but as soon as the problem kicks in again, I'm going to plug the internet cable into my ISP's FRITZ!Box router and see what results that gives. If that exhibits similar behaviour then it will eliminate the problem as being the router, but if it works fine, then there must be another explanation.

On the redacting issue, I'm now going to remove some of the images I posted.
 
Well, I want to be able to know for sure that it's not a problem with the router itself, seeing as it's brand new. I am auto-monitoring channel 11 every 10 minutes at the moment (which is currently working normally), but as soon as the problem kicks in again, I'm going to plug the internet cable into my ISP's FRITZ!Box router and see what results that gives. If that exhibits similar behaviour then it will eliminate the problem as being the router, but if it works fine, then there must be another explanation.

On the redacting issue, I'm now going to remove some of the images I posted.

It would be very unlikely that the radio is only dead on that channel/slice of spectrum. The likelihood goes to near 0 when it has happened on two different routers.

You don't have your ISP router also set to channel 11 do you? That would certainly explain your issue if they are nearby. You can't simply move the internet wire, you'd need to only have one powered on at a time.

It is possible that their router will deal with whatever bluetooth, baby monitor, cordless phone, etc is interfering a bit better than the asus. It may also be a bit worse. I can't see there being a drastic difference though.
 
@drinkingbird Interesting read for those of us that don't know this as well, thank you.

One take away for me - it's more proof the marketing people at ISPs have more influence than the poor guys on the helpdesk.

Default router config is all about maximum speed in isolation because "hey - you're the customer and you're all we care about". The fact that most customers don't live in an isolated bubble and this just translates to support headaches and a reputation for providing a crap ISP router is secondary.

(Or in some cases just another sales opportunity - "Bad signal you say? Here, sign up for our mesh network devices!")
 
If you are in Europe, might want to try Ch13. No problems with my IOTs. I have 500mbps fiber up/down. Tested from bedroom, I think it is not too bad considering the range and walls. Ch6 and Ch11 are very saturated in these blocks of apartments.

IMG_0472.jpeg
IMG_0473.jpeg
 
Similar threads
Thread starter Title Forum Replies Date
A Starting OpenVPN with TAP Client seems to crash RT-AX86U Asuswrt-Merlin 22

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top