Honest review of Asus WRT 5.0 (3.0.0.6 Branch)

[L&LD]

That is not my experience. But go ahead and believe (partial) specs.

 

[Darkje]

I really had a good experience. I have ax86u pro. But the update to .6 branch is really old now. So went back to merlin

 

[Dedel66]

@Darkje
3.0.0.6.102_33308 for ax88u is from 12.01.2024 so NOT very old...
I think Asus is working on 3006.xxx and there will be updates for ax86u pro too.

But I'm also with Merlin WRT as long as reports about 3006 don't sound really positive...

 

[Darkje]

@Dedel66 for ax 88u pro it is new aswell as the GT ax6000. But i dont have that model right. So they release a beta to test. Andere leave users with all kind of safety issues for nearlly a year

 

[glens]

Those aren't similar. The RT-AX88U Pro is superior.

That is not my experience. But go ahead and believe (partial) specs.

Just curious if something new has developed which I've missed by not camping in the forums recently.

Apart from surmised longer-support / line-up-existence, or possible appearance preference, what is the nature of your experience where the AX88 Pro exhibits any functional superiority over its fraternal twin?

 

[L&LD]

The RT-AX88U Pro is a noticeably more stable, responsive, and slightly more performant router vs the older GT-AX6000.

My guess is the same as usual: newer SDK, better RF design, more optimized firmware. Two extra years of experience can bring a lot to the table.

 

[RMerlin]

My guess is the same as usual: newer SDK

Both routers use the exact same SDK (5.04axhnd), with the same driver code.

 

[ColinTaylor]

Just curious if something new has developed which I've missed by not camping in the forums recently.

Apart from surmised longer-support / line-up-existence, or possible appearance preference, what is the nature of your experience where the AX88 Pro exhibits any functional superiority over its fraternal twin?

Didn't we already have this discussion here, and the one (not two) year difference in release FCC approval dates? Well, 13 months if you want to be pedantic.

 

[L&LD]

TUF-AX6000 vs GT-AX6000;which one is the better wifi router?

Hello, I'm currently living in South Korea for a while, and I'm looking for high-end Wifi routers for my studio apartment. I'm planning to place 2 routers, since the studio is about 1780 sqft. (165 sqm.) I'm planning to go for Asus since the products support Aimesh. The GT-AX6000 and the...

www.snbforums.com

TUF-AX6000 vs GT-AX6000;which one is the better wifi router?

Hello, I'm currently living in South Korea for a while, and I'm looking for high-end Wifi routers for my studio apartment. I'm planning to place 2 routers, since the studio is about 1780 sqft. (165 sqm.) I'm planning to go for Asus since the products support Aimesh. The GT-AX6000 and the...

www.snbforums.com

 

[glens]

Didn't we already have this discussion here, ...

Exactly what I'd had in mind, though this time the statement was more emphatic. Hence my seeking clarification.

The RT-AX88U Pro is a noticeably more stable, responsive, and slightly more performant router vs the older GT-AX6000.

That sounds like you've had one in comparative service. Is this the case? The RT is truly all those things vs. the GT? Aside from indication there has been a revised switch (I believe it was that) in the RT since start of production, it still seems to me the greatest difference between them is an alternate board for the router form factor.

 

[Amplifier]

Have you installed the latest FW 3.0.0.6.102_33308 for the RT-AX88U Pro? I just installed today and I am having a lot of Web UI inconsistencies and some kind of data corruption, I make an update, Apply the change and the Web UI shows like nothing was changed, only a Reboot will fix in some cases, in other not even a reboot

My apologies for the delay. I updated on about 24th of January 2024. So far, only this new issue. It's an annoyance, since it does create additional steps for me whenever I reboot. I'll have to ponder on ways to improve the way I deal with it.
Can't complain in general, since I don't have any issues with connections or performance.

Nitpicking.
In Wireless > General: With Smart Connect ON, Target Wake Time for 5GHz doesn't stay enabled.
Not sure if it's just a cosmetic issue on the frontend or there is an issue with an actual operation on the backend. Same toggle didn't have an issue with a Smart Connect OFF.

Apply the change and the Web UI shows like nothing was changed, only a Reboot will fix in some cases, in other not even a reboot

Web UI related issues I troubleshoot with basic principles in mind, fresh browser profile (Firefox, Chromium based, etc.), staying close to defaults as possible. Keeping in mind, Hard Refresh for a browser page, once in a while.

 

[glens]

In Wireless > General: With Smart Connect ON, Target Wake Time for 5GHz doesn't stay enabled.
Not sure if it's just a cosmetic issue on the frontend or there is an issue with an actual operation on the backend. Same toggle didn't have an issue with a Smart Connect OFF.

Relatively simple investigation involved.

Disable "Smart Connect". ssh in and issue command "nvram show > /tmp/twt.disabled". Or ".enabled" if that was the state of TWT.

Alter the state of TWT.

ssh back in if necessary and issue the complementary command (appropriate filename extension).

Enable both TWT and Smart Connect (perhaps sequentially), ssh back in if necessary, and do "nvram show > /tmp/twt.smart".

Now you've got 3 text files you can compare.

"diff /tmp/twt.*abled"

See which variables pertain. Say it's 'wl?_twt' so issue "sed -i '/_twt/!d' /tmp/twt.*"

Then "for i in /tmp/twt.*; do echo $i:; cat $i; echo; done"

It will be completely evident what's going on.

You can delete the text files if you wish, but they're innocuous, exist only in RAM, and will disappear on their own at the next reboot (what, a year from now? hopefully).

 

[Amplifier]

Relatively simple investigation involved.

Disable "Smart Connect". ssh in and issue command "nvram show > /tmp/twt.disabled". Or ".enabled" if that was the state of TWT.

Alter the state of TWT.

ssh back in if necessary and issue the complementary command (appropriate filename extension).

Enable both TWT and Smart Connect (perhaps sequentially), ssh back in if necessary, and do "nvram show > /tmp/twt.smart".

Now you've got 3 text files you can compare.

"diff /tmp/twt.*abled"

See which variables pertain. Say it's 'wl?_twt' so issue "sed -i '/_twt/!d' /tmp/twt.*"

Then "for i in /tmp/twt.*; do echo $i:; cat $i; echo; done"

It will be completely evident what's going on.

You can delete the text files if you wish, but they're innocuous, exist only in RAM, and will disappear on their own at the next reboot (what, a year from now? hopefully).

Though I wasn't bothered by the issue with TWT, I appreciate your reply very much. Was always curious to try SSH in general. Never had a reason to. Maybe I'll find some time during the week to give it a try.

 

[Amplifier]

Greetings @glens
I knew this would be interesting, and it was.

diff and scp were not found. Couldn't figure out how to operate in the router, so I just copied files to USB, then played with them locally.

(like this, but more files with changes to compare)

nvram show > /mnt/USB/twt.default
nvram show > /mnt/USB/twt.enabled

Compared files with diff on my pc (e.g. but with more files)

diff --unified twt.default twt.enabled

Not sure I follow how sed operates,

sed -i '/_twt/!d' twt.*

Copy to buffer to use for the next command? (Terminal didn't give output.)

for i in twt.*; do echo $i:; cat $i; echo; done

Output was roughly like so.

twt.default:
wl0_twt=0
wl1_twt=0
wl_twt=0

twt.enabled
wl0_twt=1
wl1_twt=0
wl_twt=1

As far as I got. If I don't touch Smart Connect (initial setup choose SC and don't disable it), then enable TWT on 5GHz, CLI will say that TWT is enabled=1 but the web GUI still says disabled. Forced refresh and logging in, in private window.
If I disable Smart Connect, even if I re-enable it after, everything works. As if enabling 2 sets of SSID (even if they are identically the same) opens something up to not cause TWT to "get stuck" in "disabled" in WEB GUI.

This for me isn't much of a concern.
Getting self-signed certificates regenerate on reboot is annoying.

syslog:

RT-AX88U_Pro: start https:1024
RT-AX88U_Pro: start httpd:80
httpd: reload cert and clean all files
httpd: Save SSL certificate...1024
httpd: Generating SSL certificate...1024
Mastiff: init
httpd: Succeed to init SSL certificate...1024
httpd: Succeed to init SSL certificate...80

Good day to you!

 

[glens]

Not sure I follow how sed operates,

sed -i '/_twt/!d' twt.*

Copy to buffer to use for the next command? (Terminal didn't give output.)

There'd be no feedback for that command unless it failed ("no news" is good news). The "-i" means edit the files in place and the "'/_twt/!d'" means delete any line which does not contain the character string "_twt". So the files then contain /only/ such lines (less to weed through).

 

[Amplifier]

I admire how your explanation is concise, it's like regex in English.

I had a rough idea what commands and options do, some bells and whistles (plugins) in zsh, help me better understand what different options do (to a degree).

Regex is a bit tricky to search for online, sometimes.
I still had to look up what ! and d individually do, for better comprehension, in hopes it creates a sticky note in my mind, to have less wasted space in there. (Who knows, might save a life one day.)
So.. kind of ! - "does not contain" and d - "string/line".

Appreciate you breaking it down for me

ta ta, later gator... I mean glens

edit
I'm guessing maybe why they don't really use (I think) exclamation point in road signs for warning, because it would put some programmers in danger translating it as "does not contain" ¯\_(ツ)_/¯ /s

 

[glens]

! is a negator. So the programmers would take the implied urgency the exactly wrong way! Hahaha.

Honestly, although I've rather lost track of exactly what was your dilemma, I do recall in working through my suggestion that I couldn't remember whether it was !/regex/ (as in a TI calculator [and didn't work]) or /regex/! (as in an HP better calculator [RPN]).

 

[Amplifier]

! is a negator. So the programmers would take the implied urgency the exactly wrong way! Hahaha.

Honestly, although I've rather lost track of exactly what was your dilemma, I do recall in working through my suggestion that I couldn't remember whether it was !/regex/ (as in a TI calculator [and didn't work]) or /regex/! (as in an HP better calculator [RPN]).

TI calculator... Could have never imagined that was part of forming your suggestion.
Amazing.
Like watching, from top of a mountain, deep within, a water of life, streaming through glens. Glens lush and green when not forming a stream.
Or so I got from https://www.vocabulary.com/dictionary/glen

No dilemma. You just introduced me on how to diff router.settings
I received your commandments in lines, and I followed you.

Kind of like running diff world.default world.beware_scribes_add_remove
Compare and you will know them by their fruits.

No idea what causes GUI settings to not match CLI.
Who and why put that enmity between them, no idea.
I'll leave that to the wizards.

 

[glens]

I welcome your accolades. So few folk "get me"...

GUI is a tradeoff for limited (in)convenience. Command-line is where it's at if one wants to get there more easily and quickly; daisy-chaining things in the *nix philosophy. Still have to type either way, just getting right to it is usually better and almost always much faster.

My taskbar currently shows three browser windows (each with multiple tabs) and 24 terminal windows. Time to do some housekeeping...

 

[glens]

Regarding 'time to do some housekeeping'... her thoughts exactly!

OE

This little bit of discussion reminded me of my P100 with, I think, 4 Mb RAM running Slackware with an early 2.x Linux kernel, before tabbed browser windows (Netscape). Somewhere I've got a screenshot indicating 96 netscape windows open, and the system was highly responsive! This was about the time Y2K was becoming a concern. The 32-bit seconds-counter overflow was actually more of a concern, but it wasn't until Windows98 itself was nearly done-for before anyone discovered a counter Windows used in W95 overflowed in like 39 days and caused a system lockup. This tells me that few people had left (were able to leave) such a system running long enough to identify the issue!

Meanwhile I'd been using an old headless 386-25 system for network (dialup and LAN) management and was truly pissed when a prolonged power outage ran out my UPS and failed the computer at somewhere near a couple years uptime! Life was simpler "back in the day".