802.11be the cheap way.... <$40
- Thread starter Tech Junky
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
There's some delay in getting it working as the modules needed to fully activate it are still in the works. I've been working on other stuff and haven't gotten back to it yet. Last I checked there has been some movement on the modules though but it's a bit of a pain to enable them vs the easy way of them being included in the kernel release.
I have to compile the kernel and add the specific updates for further testing when I find the time. The box sees the adapter right now but won't connect as a client or setup as an AP without the updates. Windows on the other hand does work as a client but there are driver issues there as well.
@thigobr
Look at the heat sink on that thing though.....
I've used other module sin the past that had those and they absolutely burn up if you try to modify them and run hot even at idle. While the 9274 is the "AP" chipset being used with most off the shelf options. $80 is cheap but, the HS worries me.
Look at the heat sink on that thing though.....
I've used other module sin the past that had those and they absolutely burn up if you try to modify them and run hot even at idle. While the 9274 is the "AP" chipset being used with most off the shelf options. $80 is cheap but, the HS worries me.
This is a thread right in my wheel house. Been an avid 802.11n Atheros 9380/9390 user. Did you measure net TCP throughput yet? At 6 GHz how much worse is penetration of walls vs. 2.4 GHz 802.11n at say 100mW and using some 3 dBi omni antennas? Did you re-use dual-band 2.4/5.8 GHz antennas?
No testing has been done since the module needs additional drivers in Linux to perform as an AP. I'm still working on it but fighting other battles ATM for more important things like switching 5G providers.. .well, just how it's billed and better performance. I ditched TMHI for $55/mo for TMO voice SIM and built my own "modem" instead that gets 4-5X the DL speed and is cheaper @ $48/mo and can be used with my phone by moving the SIM which cuts $15/mo in phone costs as well. So, makes up for the expense of the modem setup I came up with and will pay it off within a year or so in the savings. Plus the TMO SIM has an additional 40GB/mo of HS data which is more than the current phone plan @ 10GB/mo.
As to prior testing though with the AC1200 card I had good speeds but only 866mbps as AC maxes at that. With an AX AP though and AX411 wifi adapter I could hit 1.7gbps when combining the two bands into a single pipe.
With the current WIFI-7 cards they're a bit of a pain but, most of it's the newness with a single driver release and the other issue is my laptop likes to spazz out occasionally with the touchpad which usually resolves with a suspend/resume but, the driver for wifi complains about a power fault and BSODs occasionally.
My current AP though which lacks 6ghz does cover things corner to corner in 1300sq ft w/ full speeds on 5ghz. So, I don't think 6ghz will be much difference though max speeds would likely be in the same room w/o obstructions. I still think though it will beat the AP with the 320mhz of bandwidth vs 160mhz.
BE is just a bit of a toddler at this point still since it's not certified and lacking 100% drivers to enable features still in some cases. There's a couple of QCA options out and they're listed above but, client vs AP series makes a big difference in costs / heat / performance. The higher end options I found were in the $500 neighborhood for an M2 but, then you have cheap AIO options like TPL that sell for about the same but likely have worse specs inside.
It's just a waiting game at this point and $40 isn't that big of a gamble considering it can still be used as a client option if it doesn't get enabled for AP use.
Well, looks like the kernel just added support for the adapter as noted in the changes file.
I would agree if it were off the shelf equipment. Router oems tend to have a ton of bugs and break more than they fix usually. With Linux and hostapd though you don't have all of the other crap to deal with that gets piled on top to make it pretty and easy to use.
Finally taking some time to play around with this a bit more since it's been enabled.
hostapd / needs to be compiled vs just installed to enable BE .... bit of a nightmare to deal with at this point of the game
NMCLI / NMTUI - got it into AP mode a bit easier but, needs some tweaking.
- first attempt brought it up in WIFI4 (N?)
- currently in WIFI5 (AC) / speed is meh at the moment
- speed is likely due to it broadcasting 20mhz ATM / needs to be tweaked LR is hovering around ~200mbps
- need to enable AX/E / BE somehow through NMCLI
- seems to be a quirk as well causing it to disco for some reason
hostapd / needs to be compiled vs just installed to enable BE .... bit of a nightmare to deal with at this point of the game
NMCLI / NMTUI - got it into AP mode a bit easier but, needs some tweaking.
- first attempt brought it up in WIFI4 (N?)
- currently in WIFI5 (AC) / speed is meh at the moment
- speed is likely due to it broadcasting 20mhz ATM / needs to be tweaked LR is hovering around ~200mbps
- need to enable AX/E / BE somehow through NMCLI
- seems to be a quirk as well causing it to disco for some reason
Code:
nmcli con show --active
NAME UUID TYPE DEVICE
AMD 9786185d-7a3b-4d29-8955-4e2acf918e32 wifi wlp92s0
nmcli device wifi
IN-USE BSSID SSID MODE CHAN RATE SIGNAL BARS SECURITY
* BC:F4:D4:A1:FC:15 AMD Infra 36 0 Mbit/s 0 WPA1 WPA2Check defconfig « hostapd - hostap - hostapd/wpa_supplicant if you are missing
CONFIG_IEEE80211AX=y
CONFIG_IEEE80211BE=y
CONFIG_SAE=y
CONFIG_SAE_PK=y
or similar. You also need to patch the neighbor-detection to not fall back to 20 MHz channels if traffic is detected.
Here is a buildable Arch Linux package with suitable patches (I assume your devices are falling back to 802.11n 20 MHz, needs noscan=1 at the top of hostapd.conf): Index of /files/hostapd-custom/
In your hostapd.conf you will need to observe hw_mode=a as well as ieee80211ax=1 and likely a bunch of others, check hostapd.conf « hostapd - hostap - hostapd/wpa_supplicant for the entire list.
You also have to check if you aren't limited in bandwidth and output power by crda/regdb which contain a signed database. Run "iw reg get" to check the output. For testing purposes, it could look like this:
global
country DE: DFS-UNSET
(1 - 2999 @ 40), (N/A, 24), (N/A)
(3000 - 7000 @ 160), (N/A, 24), (N/A)
(50000 - 80000 @ 2160), (N/A, 24), (N/A)
phy#0
country DE: DFS-UNSET
(1 - 2999 @ 40), (N/A, 24), (N/A)
(3000 - 7000 @ 160), (N/A, 24), (N/A)
(50000 - 80000 @ 2160), (N/A, 24), (N/A)
But before you try to get a patched regdb + crda running, see if you can change the country code to something that has more bandwidth.
Wifi used to be easy in the 802.11g days, but in the 2020s things got really complicated.
CONFIG_IEEE80211AX=y
CONFIG_IEEE80211BE=y
CONFIG_SAE=y
CONFIG_SAE_PK=y
or similar. You also need to patch the neighbor-detection to not fall back to 20 MHz channels if traffic is detected.
Here is a buildable Arch Linux package with suitable patches (I assume your devices are falling back to 802.11n 20 MHz, needs noscan=1 at the top of hostapd.conf): Index of /files/hostapd-custom/
In your hostapd.conf you will need to observe hw_mode=a as well as ieee80211ax=1 and likely a bunch of others, check hostapd.conf « hostapd - hostap - hostapd/wpa_supplicant for the entire list.
You also have to check if you aren't limited in bandwidth and output power by crda/regdb which contain a signed database. Run "iw reg get" to check the output. For testing purposes, it could look like this:
global
country DE: DFS-UNSET
(1 - 2999 @ 40), (N/A, 24), (N/A)
(3000 - 7000 @ 160), (N/A, 24), (N/A)
(50000 - 80000 @ 2160), (N/A, 24), (N/A)
phy#0
country DE: DFS-UNSET
(1 - 2999 @ 40), (N/A, 24), (N/A)
(3000 - 7000 @ 160), (N/A, 24), (N/A)
(50000 - 80000 @ 2160), (N/A, 24), (N/A)
But before you try to get a patched regdb + crda running, see if you can change the country code to something that has more bandwidth.
Wifi used to be easy in the 802.11g days, but in the 2020s things got really complicated.
Last edited:
@Stephan
Good places to look. I'm running Ubuntu but should be able to cobble together something with source. Patches in the other hand throw me for a loop on how to import them. As for country US typically has unlimited options for bandwidth. I used to run AP mode on an AC card so, not foreign territory. Just not used to compiling things.
Good places to look. I'm running Ubuntu but should be able to cobble together something with source. Patches in the other hand throw me for a loop on how to import them. As for country US typically has unlimited options for bandwidth. I used to run AP mode on an AC card so, not foreign territory. Just not used to compiling things.
Ok I edited that post enough. Btw ieee80211be=1 seems to me not even finalized. Wikipedia says end of 2024. So lots of items that can break and you are right with the qualifier "at this point of the game". Including driver bugs i.e. so you can't even enable 802.11be. Check your regulatory settings anyway. I left Ubuntu over 10 years ago, when I wanted a kernel patch but couldn't figure out how to proceed "cleanly". Went to Arch Linux and their PKGBUILD was so intuitive to me, I never looked back. When I look at the table at IEEE 802.11be - Wikipedia I can see that 802.11be usurps the entire 2.4 GHz 80 MHz of spectrum. Wow. Can't even fall back to lower channel widths like 40 or 20. Unless fall-back means back to 802.11ax. Also recommend to ping something at rate of 10 times per second over the air link if you think you're good. Just for an hour. To see if there are drop-outs because something crashes and restarts.
Well, WiFi alliance certified BE so.... I run extended pings anyway for other reasons already. NM does seem to have dropouts ATM but, I just did some basic setup to get it into AP mode to prove it works at a basic level so far and need to tweak it a bit if possible before dealing with patches and such with hostapd. Just picking my fights right now with the OS starting with the low hanging fruit.
@Stephan
So, someone on another forum pointed at patching the firmware....
Went there and grabbed the most recent option and have the files dated 1/15/2024 >>> /lib/firmware/ath12k/WCN7850/hw2.0/
My current versions were from looks like quite awhile ago....
After FW update
Now a couple of things stand out... the FW is now using newer info 12/6/2023 and there's now a board-2 file like on prior QCA cards but.... is it loading or using the old method m3/amss instead.... Maybe the next push tot he FW will include the patch. I checked for the 6.8-RC2 and it's out but, failed testing and don't feel like building / compiling it as nothing was mentioned for anything I'm using or wanting.
I guess it's time to monitor stability for awhile before bothering with digging into it deeper to get the BW up to where it should be which likely isn't something to be done with NM but hostapd for more control over the connection params.
So, someone on another forum pointed at patching the firmware....
Went there and grabbed the most recent option and have the files dated 1/15/2024 >>> /lib/firmware/ath12k/WCN7850/hw2.0/
My current versions were from looks like quite awhile ago....
Code:
[ 5.813755] ath12k_pci 0000:5c:00.0: BAR 0 [mem 0xa1a00000-0xa1bfffff 64bit]: assigned
[ 5.813778] ath12k_pci 0000:5c:00.0: enabling device (0000 -> 0002)
[ 5.814356] ath12k_pci 0000:5c:00.0: MSI vectors: 16
[ 5.814362] ath12k_pci 0000:5c:00.0: Hardware name: wcn7850 hw2.0
[ 6.388609] ath12k_pci 0000:5c:00.0: chip_id 0x2 chip_family 0x4 board_id 0xff soc_id 0x40170200
[ 6.388613] ath12k_pci 0000:5c:00.0: fw_version 0x1005bc12 fw_build_timestamp 2022-10-11 12:13 fw_build_id QC_IMAGE_VERSION_STRING=WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4
[ 6.924974] ath12k_pci 0000:5c:00.0: WBRF is not supported
[ 7.460368] ath12k_pci 0000:5c:00.0 wlp92s0: renamed from wlan0
Code:
[ 6.205039] ath12k_pci 0000:5c:00.0: fw_version 0x100301e1 fw_build_timestamp 2023-12-06 04:05 fw_build_id QC_IMAGE_VERSION_STRING=WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
[ 6.212227] ath12k_pci 0000:5c:00.0: failed to fetch board data for bus=pci,qmi-chip-id=2,qmi-board-id=255 from ath12k/WCN7850/hw2.0/board-2.binI guess it's time to monitor stability for awhile before bothering with digging into it deeper to get the BW up to where it should be which likely isn't something to be done with NM but hostapd for more control over the connection params.
Look here: Arch Linux - linux-firmware 20240115.9b6d0b08-1 (any) Never old (also: never without a few regressions). If you ride such recent hardware also need to ride a rolling distro. And if you don't trust me, trust Chris Titus:
Cool idea. I've read that this trick isn't a great idea though. I never could figure out why. The implication was that performance was crap. I would like to set up something like this eventually, the router/AP boxes are overpriced at an insane level for the newer wifi standards. Even 6E would be really nice to jump to, mostly for the new spectrum. I only have a few devices that can use it, but I'm still running AC because I can't bring myself to pay for the newer versions. lol. I only need basic AP functionality, VLAN would be nice, routing and firewalling is handled on Opnsense.
It also helps that most of the gear around the house is close to an ethernet port, so most of the traffic on my network is wired.
It also helps that most of the gear around the house is close to an ethernet port, so most of the traffic on my network is wired.
Look at the markup of 10X the cost of an adapter vs stuff on the shelf.
When it comes down to it though the experience is the same for the most part for single or dual clients. The issue beyond that comes down to RF scheduling for multiple clients.
I used a QNAP 1200 AC card for years w/o issue and didn't have to deal with the junk FW something off the shelf would have had released often. all of the off the shelf gear runs the same SW hostapd to run the wifi. So, why not take back the control and not allow them to profit off SW they get for free? Of course you don't get the slick GUI or pretty enclosure but, you also don't pay the premium and can get better performance relatively easy.
Right now since I'm still tinkering with it and testing the basics for stability it's not 100% in terms of performance but, it's already doing AP mode when it was suggested that it wouldn't / couldn't. I noticed the RF isn't too great with the cheap antennas I have on the card at the moment but, I have better ones from the AC card that will boost the RF to a satisfactory level again. Current speeds though are adequate for my 5G / LTE ISP setup though during some hours would only be a fraction. The main temptation though would be the ~5gbps BE internally when moving files around w/o needing to hook up an Ethernet cable to get those speeds. Instead of Ethernet though I've been using TB instead which gets me 1.5GB/s or 3X more than Ethernet. The option for higher Ethernet speeds w/ a laptop gets costly since even a 10GE dongle would run ~$150 + upgrade the server $75 vs a TB cable which was under $25.
Since updating the FW though it does seem more stable as my phone stayed connected for quite a bit longer before switching back to the AP. Not seeing as much fluff in syslog either since the FW update.
For $40 for the card it's a decent deal once I put some more time into building out the functionality to make it really perform well as an AP. Since it's internalized inside the server the options for configuration are limitless and monitoring traffic / gathering statistics is easier vs an external AP. I would really like it if there were some other options to tinker with but, either they're junk or $500+ for the controllers that are geared towards use in the AP market. The point though is to spend less and get the same performance. There's just a bit of untangling of things to get it up to speed.
So, I've been using one of the adapters in my laptop and using the Gigabyte driver x.x.x.909 and it presents some issues that have been naggingly annoying for the past few months now. I went on the hunt for updates for this apparently niche card and finally found an updated version on MSI's site.
www.msi.com
3.0.0.1078
Just installed the updated driver and will have to see if this improves things. Mainly I have an issue with my trackpad that requires a quick suspend / resume to purge the issue. Well, the WIFI driver normally works for maybe 4-5 times of this before crapping out and requiring a full reboot to resume network connectivity. Sometimes though it hangs on a DMA driver BSOD though upon reboot. It's just a PITA sometimes and maybe the newer version will reduce my frustrations. When using the Intel AX411 this wasn't an issue so, it leads me to think it's a quirk of the 909 driver.
Seeing as thought he laptop is Intel I might reconsider using the BE200 instead.... Though there are some potential drawbacks to this idea.
On the AP front there's been some movement on rolling in the ath12k modules
I just updated to kernel 6.8 a little while ago and haven't had a chance to check things out when it comes to the WIFI card / AP mode yet.
MSI MEG Z790 ACE MAX, E-ATX - Intel 13th/12th Gen - 24+1+2 Phases, DDR5, PCIe 5.0, 2.5G LAN, Wi-Fi 7
The MEG Z790 ACE MAX furnished premium aesthetics with its sleek black finish and silver embellishes. Designed to unlock full potential by elite hardware with premium durability, the MEG Z790 ACE MAX is an outstanding E-ATX platform for extreme settings.
Just installed the updated driver and will have to see if this improves things. Mainly I have an issue with my trackpad that requires a quick suspend / resume to purge the issue. Well, the WIFI driver normally works for maybe 4-5 times of this before crapping out and requiring a full reboot to resume network connectivity. Sometimes though it hangs on a DMA driver BSOD though upon reboot. It's just a PITA sometimes and maybe the newer version will reduce my frustrations. When using the Intel AX411 this wasn't an issue so, it leads me to think it's a quirk of the 909 driver.
Seeing as thought he laptop is Intel I might reconsider using the BE200 instead.... Though there are some potential drawbacks to this idea.
On the AP front there's been some movement on rolling in the ath12k modules
Code:
sudo dmesg | grep ath12k
[ 4.106677] ath12k_pci 0000:5c:00.0: BAR 0 [mem 0xa1a00000-0xa1bfffff 64bit]: assigned
[ 4.106699] ath12k_pci 0000:5c:00.0: enabling device (0000 -> 0002)
[ 4.106946] ath12k_pci 0000:5c:00.0: MSI vectors: 16
[ 4.106951] ath12k_pci 0000:5c:00.0: Hardware name: wcn7850 hw2.0
[ 5.087576] ath12k_pci 0000:5c:00.0: chip_id 0x2 chip_family 0x4 board_id 0xff soc_id 0x40170200
[ 5.087580] ath12k_pci 0000:5c:00.0: fw_version 0x100301e1 fw_build_timestamp 2023-12-06 04:05 fw_build_id QC_IMAGE_VERSION_STRING=WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
[ 5.095111] ath12k_pci 0000:5c:00.0: failed to fetch board data for bus=pci,qmi-chip-id=2,qmi-board-id=255 from ath12k/WCN7850/hw2.0/board-2.binLooks like some more movement on the driver side. A few specific to WCN7850 and more for ath12k in general unlocking some additional options to support AP mode.