Open 2goese opened 8 months ago
Does the led work and does it have good range on the listed distros?
Are there any module parameters to enable the led? I think the modinfo xxx will show you.
Maybe the default txpower is low and that affects the range. I don't know the command to check it.
Does the led work and does it have good range on the listed distros?
Yep no issues at all
Tried modinfo mt76x0u
with output:
filename: /lib/modules/6.5.7-arch1-1/kernel/drivers/net/wireless/mediatek/mt76/mt76x0/mt76x0u.ko.zst
license: GPL
firmware: mediatek/mt7610u.bin
firmware: mediatek/mt7610e.bin
srcversion: 9BE8EDA0752A3110CF9AB7E
alias: usb:v0E8Dp7650d*dc*dsc*dp*icFFisc02ipFFin*
alias: usb:v0E8Dp7630d*dc*dsc*dp*icFFisc02ipFFin*
alias: usb:v2357p0105d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v2357p010Bd*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v2357p0123d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v0DF6p0079d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v7392pC711d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v20F4p806Bd*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v293Cp5702d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v057Cp8502d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v04BBp0951d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v07B8p7610d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v0586p3425d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v2001p3D02d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v2019pAB31d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v0DF6p0075d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v0B05p17DBd*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v0B05p17D1d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v148Fp760Ad*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v148Fp761Ad*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v7392pB711d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v7392pA711d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v0E8Dp7610d*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v13B1p003Ed*dc*dsc*dp*ic*isc*ip*in*
alias: usb:v148Fp7610d*dc*dsc*dp*ic*isc*ip*in*
depends: mt76,mt76x02-lib,mt76-usb,mt76x0-common,mac80211,mt76x02-usb
retpoline: Y
intree: Y
name: mt76x0u
vermagic: 6.5.7-arch1-1 SMP preempt mod_unload
sig_id: PKCS#7
signer: Build time autogenerated kernel key
sig_key: 20:28:E7:B1:AE:78:54:88:41:13:D6:4A:5B:7C:50:6D:79:3E:96:6E
sig_hashalgo: sha512
signature: 30:66:02:31:00:DB:2E:53:13:9F:A3:56:B4:50:9F:65:43:F0:10:EA:
FD:97:8E:48:4D:04:72:7B:8A:F0:7B:6C:6F:6D:2C:40:1D:44:FB:4E:
B7:21:76:7B:EB:1F:4E:53:34:57:B1:B7:E2:02:31:00:DB:4F:6E:09:
AA:68:FB:28:F9:BC:BE:72:1D:74:F3:87:22:60:3E:98:BA:A4:26:ED:
A9:F4:6C:B7:EC:37:C5:03:9B:56:B2:06:51:BB:C5:96:C8:AA:AC:48:
2E:09:10:AC
But still no clue about LED
I don't know the command to check it.
maybe
iwconfig wlan0 | grep Tx-Power
Bit Rate=72.2 Mb/s Tx-Power=2 dBm
Try modinfo on the other modules. Look for any that have parameters. Look for parm.
$ modinfo mt76_usb filename: /lib/modules/6.1.21-v8+/kernel/drivers/net/wireless/mediatek/mt76/mt76-usb.ko.xz license: Dual BSD/GPL author: Lorenzo Bianconi @.***> srcversion: AF367BA27120C1ECD0D8DFD depends: mt76 intree: Y name: mt76_usb vermagic: 6.1.21-v8+ SMP preempt mod_unload modversions aarch64 parm: disable_usb_sg:Disable usb scatter-gather support (bool)
I found it on the same module you just listed, what is disable_usb_sg
parm?
Hi @2goese
I'll talk about the range first. I made a range test using a few adapters a couple of years ago. The document is not on the Main Menu currently as it dated so I need to do a new test. Going into the test I did not realize that the ACHM had the range that it does so the data caught my eye. The test is here:
https://github.com/morrownr/USB-WiFi/blob/main/home/Performance_Comparison.md
This is a relative test in that the specific numbers only mean something when compared to other adapters that were tested using the same conditions. The data that I keyed in on for range is Link Quality and Signal Level. I looked at iperf3 for same class chipset. I need to put a plan together to do an updated. Maybe I can knock that out by the end of the year.
About the LED: I need to ask the devs but I do not think LED support is in the modern Linux wifi stack. I can't think of any wifi 5 and later usb wifi adapters where the in-kernel driver is supporting an LED, if the adapter has one. Let me do some testing and checking. There is a manual way to start LED support for the mt7612u chipset via debugging. Alfa, with its WiFi 6e adapters that it released this year has LEDs that work but the LEDs are not driver controlled. The LEDs are actually cool. If there is power, the LED is on but they slightly recessed the LEDs and made the light level very low. You can see it but it is not bright enough to bother you if you are trying to sleep.
My experience with the ACHM is a good one. It has proved to be a very dependable adapter that has no thermal problems and huge uptime. If the LED worked, I'd be looking for a way to turn it off so that but if it is important to you, maybe a solution can be found... maybe something like for the mt7612u but we would need to talk to the devs.
Nick
I am looking for a parm related to the led, not the scatter gather. It sounds like there isn't one from what Nick said so I don't know how other distros control the led.
Maybe run this command on archlinux and a distro that has better range.
$ iw wlp0s20f3 info Interface wlp0s20f3 txpower 22.00 dBm
See if the default txpower is different.
Also check the country between the one that works and archlinux: $ iw reg get
@morrownr
Thanks for the information.
About the range test. Does it still valid with the non supported OS? For me range seems abnormal here in Arch.
Regarding the LED. I would appreciate if you could find a solution, I like it blinking.
The reason why I choose this adapter is because of pentesting. Even though driver is quite old, still on USB2. How do you compare this chipset with MT7612u and MT7921au?
@bjlockie
$ iw adapter info
Uhhh... nope wrong command. I think it's iwconfig
$ iw reg get
This one i got a bunch of help command spreading all over my screen
Thanks for the suggestions
@2goese
About the range test. Does it still valid with the non supported OS?
Should still be valid. Non-supported in many cases just means resellers don't want to be bothered with paying for support to learn the details of several different distros and it is a valid issue. I have built up a small lab with test systems over the years, I have several distros on the systems so I can test the drivers I work on with multiple distros. Different ranges with different distros is not something that has caught my attention.
For me range seems abnormal here in Arch.
Tell me what you are looking at that is telling me the range in different in Arch.
Uhhh... nope wrong command. I think it's iwconfig
iwconfig has been depreciated for years, Please use iw. What are you trying to do?
$ iw reg get
Here is what I get:
$ iw reg get
global
country US: DFS-FCC
(902 - 904 @ 2), (N/A, 30), (N/A)
(904 - 920 @ 16), (N/A, 30), (N/A)
(920 - 928 @ 8), (N/A, 30), (N/A)
(2400 - 2472 @ 40), (N/A, 30), (N/A)
(5150 - 5250 @ 80), (N/A, 23), (N/A), AUTO-BW
(5250 - 5350 @ 80), (N/A, 24), (0 ms), DFS, AUTO-BW
(5470 - 5730 @ 160), (N/A, 24), (0 ms), DFS
(5730 - 5850 @ 80), (N/A, 30), (N/A), AUTO-BW
(5850 - 5895 @ 40), (N/A, 27), (N/A), NO-OUTDOOR, AUTO-BW, PASSIVE-SCAN
(5925 - 7125 @ 320), (N/A, 12), (N/A), NO-OUTDOOR, PASSIVE-SCAN
(57240 - 71000 @ 2160), (N/A, 40), (N/A)
$ uname -a
One of mine is:
$ iw reg get
global
country US: DFS-FCC
(902 - 904 @ 2), (N/A, 30), (N/A)
(904 - 920 @ 16), (N/A, 30), (N/A)
(920 - 928 @ 8), (N/A, 30), (N/A)
(2400 - 2472 @ 40), (N/A, 30), (N/A)
(5150 - 5250 @ 80), (N/A, 23), (N/A), AUTO-BW
(5250 - 5350 @ 80), (N/A, 24), (0 ms), DFS, AUTO-BW
(5470 - 5730 @ 160), (N/A, 24), (0 ms), DFS
(5730 - 5850 @ 80), (N/A, 30), (N/A), AUTO-BW
(5850 - 5895 @ 40), (N/A, 27), (N/A), NO-OUTDOOR, AUTO-BW, PASSIVE-SCAN
(5925 - 7125 @ 320), (N/A, 12), (N/A), NO-OUTDOOR, PASSIVE-SCAN
(57240 - 71000 @ 2160), (N/A, 40), (N/A)
Replying with my 2nd account
@morrownr
For the range, I just feels something different on Windows graph bar than in Linux nmcli
graph bar.
For example my neighbours network who lived approximately 7m away from my house got 3 bars on Windows, 1 on Linux. The connection is unreliable though, sometimes it got disconnected too. Should i trust the display bar?
About iwconfig
and iw
(I'm not really familiar with it)
I usually like to view status of my adapter and also not forgetting ifconfig
for IP and MAC. Also lot of wifi hacking tutorial still use those commands.
~ iw reg get
global
country 00: DFS-UNSET
(2402 - 2472 @ 40), (6, 20), (N/A)
(2457 - 2482 @ 20), (6, 20), (N/A), AUTO-BW, PASSIVE-SCAN
(2474 - 2494 @ 20), (6, 20), (N/A), NO-OFDM, PASSIVE-SCAN
(5170 - 5250 @ 80), (6, 20), (N/A), AUTO-BW, PASSIVE-SCAN
(5250 - 5330 @ 80), (6, 20), (0 ms), DFS, AUTO-BW, PASSIVE-SCAN
(5490 - 5730 @ 160), (6, 20), (0 ms), DFS, PASSIVE-SCAN
(5735 - 5835 @ 80), (6, 20), (N/A), PASSIVE-SCAN
(57240 - 63720 @ 2160), (N/A, 0), (N/A)
Is there something wrong?
For example my neighbours network who lived approximately 7m away from my house got 3 bars on Windows, 1 on Linux.
This is a test where you are at your neighbors house with your laptop connecting to your wifi router? I'm trying to figure out the exact situation here, Did you check to see what channel was in use with Windows and what channel was in use with Linux? Did you do any throughput testing.
A handy utility that I use to help when testing is wavemon:
On Debian the installation would be:
$ sudo apt install wavemon
To run:
$ wavemon
It would also be handy if you did a speed test with both Windows and Linux to see where we are. Something like:
I usually like to view status of my adapter and also not forgetting ifconfig for IP and MAC. Also lot of wifi hacking tutorial still use those commands.
I know that a lot of tutorials use iwconfig but many tutorials just recycle old stuff to save effort. Here us the linux.org informational link:
https://wireless.wiki.kernel.org/en/users/documentation/iw
There are a lot of good tutorials about 'iw' available. The most often use is probably the following command:
$ iw dev
which shows the wifi interfaces.
Is there something wrong?
Possibly so. You appear to be running global defaults which is usually not desirable. Were you connected to the internet when you ran $ iw reg get ?
I may need to round you up a tutorial regarding the regdb. It is possible that Arch is messing this up.
Also, what is your country if you don't mind telling us?
I mean I'm trying to connect to my neighbours 2.4Ghz wifi from my house, their channel probably set to auto so I couldn't guess. On my router I set the 5Ghz to use channel 149 the adapter shows max 13 dBm Tx-Power and Link Quality=45/70 Signal level=-65 dBm. 2.4Ghz on channel 1 shows max 2 dBm Tx-Power and Link Quality=70/70 Signal level=-31 dBm. I even disabled the MAC randomization for stable link.
Wavemon is indeed a handy one, thanks for the suggestion.
Sometimes after connecting to a network, say I want to scan the network again with nmcli device wifi list
it's slow. It only shows 1 network which is the one I'm connected to, other networks just disappear. This condition is random though, so I'm not sure if this bug or something else, the temporary solution for this is just restarting the NetworkManager service.
For speed test on Linux: Minimal 16.91 Mbps, Average 18.37 Mbps, Maximum 21.56 Mbps. Really bad for 5Ghz maybe because my router is not that good or maybe my ISP doesn't give me enough bandwidth. For 2.4Ghz: Same as above with exception of Maximum is 19.88 Mbps.
Same thing happens on Windows. Only the range seems better, for example on Linux 5Ghz network I got only 2-3 bars on Windows the bar maxed out.
About iw
I'm still trying to get used to it.
Were you connected to the internet when you ran $ iw reg get ?
Yes I did.
Also, what is your country if you don't mind telling us?
I'm from Indonesia.
Link Quality=45/70 Signal level=-65 dBm.
This is a marginal situation. You may be able to get a usable connection and maybe not. It won't be a fast connection.
Link Quality=70/70 Signal level=-31 dBm.
This is an outstanding situation where you should have no problem with the connection or speed as long as congestion is not too bad.
Really bad for 5Ghz maybe because my router is not that good or maybe my ISP doesn't give me enough bandwidth.
This speed check on the internet gives me some ideas but internet speed checks are not optimal for what we are going. If you have a RasPi4B, we could setup a iperf3 server, it is easy, and we could check via your lan.
Speaking of your router: Are you familiar with how to access the settings and how to change the settings?
Same thing happens on Windows. Only the range seems better, for example on Linux 5Ghz network I got only 2-3 bars on Windows the bar maxed out.
We need to determine if this is a perceived difference or a real difference. Hopefully we can do that.
About iw I'm still trying to get used to it.
You will get used to it.
Not showing a country code with $ iw reg get is a little concerning and could be a problem. Let's mess with it:
$ iw reg set ID
Then run:
$ iw reg get
...and post the results.
I need to look in the regdb to see what is allowed in ID.
What Linux distro and version are you using?
If you have a RasPi4B, we could setup a iperf3 server, it is easy, and we could check via your lan.
I don't have a RasPI :( I also don't have any experience with LAN :( My interest is mostly on wireless, but maybe in the future.
Speaking of your router: Are you familiar with how to access the settings and how to change the settings?
A little bit only some that I’m already familiar with, different brand have different settings.
We need to determine if this is a perceived difference or a real difference. Hopefully we can do that.
I'll wait then, because it's the main issue aside from the LED.
You will get used to it.
Hopefully.
~ sudo iw reg set ID
~ iw reg get
global
country 00: DFS-UNSET
(2402 - 2472 @ 40), (6, 20), (N/A)
(2457 - 2482 @ 20), (6, 20), (N/A), AUTO-BW, PASSIVE-SCAN
(2474 - 2494 @ 20), (6, 20), (N/A), NO-OFDM, PASSIVE-SCAN
(5170 - 5250 @ 80), (6, 20), (N/A), AUTO-BW, PASSIVE-SCAN
(5250 - 5330 @ 80), (6, 20), (0 ms), DFS, AUTO-BW, PASSIVE-SCAN
(5490 - 5730 @ 160), (6, 20), (0 ms), DFS, PASSIVE-SCAN
(5735 - 5835 @ 80), (6, 20), (N/A), PASSIVE-SCAN
(57240 - 63720 @ 2160), (N/A, 0), (N/A)
Nope nothing's changed.
What Linux distro and version are you using?
~ uname -r
6.5.7-arch1-1
My interest is mostly on wireless, but maybe in the future.
I'm not trying to turn you into a lan administrator, just trying to get more accurate test results but we are good.
Something does seem messed up but it is not clear what.
No offense.
I manage to change the country code after installing wireless-regdb
then reboot, here's the results:
~ iw reg get
global
country CN: DFS-FCC
(2400 - 2483 @ 40), (N/A, 20), (N/A)
(5150 - 5350 @ 80), (N/A, 20), (0 ms), DFS, AUTO-BW
(5725 - 5850 @ 80), (N/A, 33), (N/A)
(57240 - 59400 @ 2160), (N/A, 28), (N/A)
(59400 - 63720 @ 2160), (N/A, 44), (N/A)
(63720 - 65880 @ 2160), (N/A, 28), (N/A)
phy#1
country CN: DFS-FCC
(2400 - 2483 @ 40), (N/A, 20), (N/A)
(5150 - 5350 @ 80), (N/A, 20), (0 ms), DFS, AUTO-BW
(5725 - 5850 @ 80), (N/A, 33), (N/A)
(57240 - 59400 @ 2160), (N/A, 28), (N/A)
(59400 - 63720 @ 2160), (N/A, 44), (N/A)
(63720 - 65880 @ 2160), (N/A, 28), (N/A)
~ iw reg set ID
~ iw reg get
global
country 98: DFS-UNSET (ALFA AWUS036ACHM)
(2400 - 2483 @ 40), (N/A, 20), (N/A), NO-OUTDOOR
(5150 - 5350 @ 80), (N/A, 20), (0 ms), NO-OUTDOOR, DFS
(5725 - 5825 @ 80), (N/A, 23), (N/A), NO-OUTDOOR
phy#1
country CN: DFS-FCC (TP-LINK WN722N V1)
(2400 - 2483 @ 40), (N/A, 20), (N/A)
(5150 - 5350 @ 80), (N/A, 20), (0 ms), DFS, AUTO-BW
(5725 - 5850 @ 80), (N/A, 33), (N/A)
(57240 - 59400 @ 2160), (N/A, 28), (N/A)
(59400 - 63720 @ 2160), (N/A, 44), (N/A)
(63720 - 65880 @ 2160), (N/A, 28), (N/A)
country CN: DFS-FCC (TP-LINK WN722N V1)
Canada?
global country 98: DFS-UNSET (ALFA AWUS036ACHM)
So the ACHM is set to global default and the &22 is set to Canada. Yup. Something is messed up.
I'm pondering this. In this day in time, this setup should be automatic but something is not right. The 722 may be hardcoded in eeprom to Canada and we cannot do much about that. The ACHM does not appear to be hardcoded, which is good, but it is not getting the proper info from the kernel.
Are you locked into using Arch or would you consider an alternative distro?
Canada?
It's China probably set to default, I could use ID though but it won't last until reboot.
So the ACHM is set to global default
This I had different result than the other one:
~ iw reg get
global
country ID: DFS-JP (ALFA AWUS036ACHM)
(2400 - 2483 @ 40), (N/A, 26), (N/A), NO-OUTDOOR
(5150 - 5350 @ 80), (N/A, 23), (N/A), NO-OUTDOOR
(5725 - 5825 @ 80), (N/A, 23), (N/A), NO-OUTDOOR
Are you locked into using Arch or would you consider an alternative distro?
Probably, I just recently migrated from Windows and I'am still learning the way Arch works, maybe if using a virtual machine.
It's China probably set to default, I could use ID though but it won't last until reboot.
Yes. Canada is CA so I had that wrong. Why we are seeing CN is not clear but it may be a that CN is set in eeprom. A lot of things have been messed up over the years due to a lot of floundering over the years regarding exactly how to handle the laws regarding wifi over the world.
country ID: DFS-JP (ALFA AWUS036ACHM)
Your ACHM appears to be working the way it should.
You can download and check the regdb to make sure what you are seeing is the correct info for ID:
https://kernel.googlesource.com/pub/scm/linux/kernel/git/sforshee/wireless-regdb/
You not have needed to download the wireless file you downloaded to set the country code because things don't work that way anymore and have not for a few years. It seems the ACHM is working as it should and the 722 is locked into an incorrect country code.
I'am still learning the way Arch works...
I'm probably going to catch the wrath of Arch users when I say this but learning Linux with Arch is not what I would suggest to the those starting out on Linux. I have numerous distros installed in my lab and to test "Arch" I use Manjaro which is based on Arch. If you were to ask me what you would suggestion as a distro to start with:
https://linuxmint.com/download_lmde.php
LMDE 6 is based on Debian 12. It is more than just a good starter distro... I use it on my main dev box. It is extremely stable and easy to use.
Sorry for the extra discussion. To get back on track, where are we? What problem are we trying to solve?
A lot of things have been messed up over the years due to a lot of floundering over the years regarding exactly how to handle the laws regarding wifi over the world.
Let me clarify: This comment is not intended just for Linux. It is intended for Wireless in general and applies to Windows and all other operating systems. A lot of modernization has had to take place with wireless over the last several years and there are products out there that will cause problems.
check the regdb to make sure what you are seeing is the correct info for ID:
Why should I care about the regulatory domain thing, it shouldn't affect the perfomance right?
LMDE 6 is based on Debian 12. It is more than just a good starter distro... I use it on my main dev box. It is extremely stable and easy to use.
You're getting off-topic here
Please focus on the Range and LED. I need solution
Why should I care about the regulatory domain thing, it shouldn't affect the perfomance right?
It can have everything to do with performance. With the wrong regulatory information or simply the worldwide default, you almost certainly get will get less than optimal performance. Without that sorted out, we will have a hard time working toward good performance.
You're getting off-topic here
I'm trying to help. I know a lot about many distros but I am not an expert on Arch. It might be a good idea for you to post on an Arch forum.
Please focus on the Range and LED. I need solution
Try a different distro. This performance could be an Arch problem. Arch needs to be eliminated as a source of the problem.
I can't solve the LED problem as the driver simply does not support the LED. I can ask the devs why this is and if it may change.
The optimal choice for me is either CN or ID, but sometimes ID do get a buffering while streaming on 1080p, I haven't got any problem with CN, gonna try Singapore this time because it's the closest to my country.
It might be a good idea for you to post on an Arch forum.
I think you should scroll to the top, already posted there.
Arch needs to be eliminated as a source of the problem.
This should be a last resort.
Based on a previous test I did on Windows, it might show different range on Arch but in fact the speed test are exactly the same on both OS.
I can't solve the LED problem as the driver simply does not support the LED. I can ask the devs why this is and if it may change.
Well it's good to know that, I'm waiting for the fix.
I think you should scroll to the top, already posted there.
This site gets about 18,000 hits per week. I simply do not have time to reread messages for each issue.
The optimal choice for me is either CN or ID, but sometimes ID do get a buffering while streaming on 1080p, I haven't got any problem with CN, gonna try Singapore this time because it's the closest to my country.
If Arch is working correctly, you should not actually be able to change the country code and it be used by the kernel. If it is actually being used, I currently do not have an explanation and can only say that something seems wrong... unless you are using a very very old version of Arch.
This should be a last resort.
Your call. I've only been using Linux since 1994. What do I know?
Based on a previous test I did on Windows, it might show different range on Arch but in fact the speed test are exactly the same on both OS, don't know why.
This suggests that the range is actually the same or close to the same despite what the numbers may seem to indicate. I've seen this many times. This also suggests that wifi router/AP settings could be part or all of the performance problem. To change the settings, you will have to discover how to enter the wifi router/AP interface and see what settings are available. Here is some general guide I have posted regarding the settings:
Recommended WiFi Router/ Access Point Settings
Note: These are general recommendations, some of which may not apply to your specific situation.
Security: Set WPA2-AES or WPA2/WPA3 mixed or WPA3. Do not set WPA2 mixed mode or WPA or TKIP.
Channel width for 2.4 GHz: Set 20 MHz fixed width. Do not use 40 MHz or 20/40 automatic.
Channels for 2.4 GHz: Set channel 1 or 6 or 11 depending on the congestion at your location. Do not set automatic channel selection. As time passes, if you notice poor performance, recheck congestion and set channel appropriately. The environment around you can and does change over time.
Mode for 2.4 GHz: For best performance, set "N only" if you no longer use B or G capable devices.
Network names: Do not set the 2.4 GHz Network and the 5 GHz Network to the same name. Note: Unfortunately many routers come with both networks set to the same name. You need to be able to control which network that is in use so changing the name of one of the networks is recommended. Since many IoT devices use the 2.4 GHz network, it may be better to change the name of the 5 GHz network.
Channels for 5 GHz: Not all devices are capable of using DFS channels (I'm looking at you Roku.) It may be necessary to set a fixed channel in the range of 36 to 48 or 149 to 165 in order for all of your devices to work on 5 GHz. (For US, other countries may vary.)
Best location for the WiFi router/access point: Near center of apartment or house, at least a couple of feet away from walls, in an elevated location. You may have to test to see what the best location is in your environment.
Check congestion: There are apps available for smart phones that allow you to get an idea of the congestion levels on WiFi channels. The apps generally go by the name of WiFi Analyzer or something similar.
After making and saving changes, reboot the router.
So you're telling about first hand experience with this adapter, yes it's true BUT only on supported OS listed here.
I use Arch linux, unfortunately it's not listed there, I've already explain about the issues here.
Updating the firmware using this guide will only make it worse, it can no longer scan nearby networks.
Am i missing something?