GL-MT6000 Flint 2 Should I keep it? WiFi Woes

[Kit]

I have recently purchased the Flint 2 (at full market price) and need to decide whether to return it before the return windows closes. I confess that I purchased it before having read about the various WiFi issues in this forum! Are people confident that a fully satisfactory firmware is possible to resolve the issues, or is it possible that there is an inhererent hardware issue which cannot be solved?

I am very surprised that such a router could have been put on the open market without adequate testing that it actually met the claimed specifications.

Any thoughts would be appreciated - should I keep it?

[admon]

I can only speak for myself: I’m not sure that it’s not a hardware problem - that’s because I don’t know anything about hardware.

However, as no new chips have been developed here, but existing systems have been put together to form a product, it should all be fixable. On the whole, my device is no more trouble than I would expect for domestic use (as Wi-Fi is simply garbage technology)

Would I return the device? I have no idea.

[danbratu]

@Kit

well trailed

“well trailed” ? i believe it was quite opposite, hence all the issues. pre-ordered 2 of it, sold both one week ago. i don’t have time to be a beta-tester for a product supposed to be stable since at least one month ago.

[admon]

@danbratu

i don’t have time to be a beta-tester for a product supposed to be stable since at least one month ago.

That’s what you get when you choose a device based on OpenWrt.
It’s not like an apple product - it’s more like some nerdy unix computer.

[danbratu]

are you sure about this? i do run openwrt on a bunch of omada APs, ubi APs, r6s gw, flin1 x 2, slate ax (soldius versions on glinet devices) plus few other devices without any issues. please stop being glinet’s advocate. admit that flint2 is a major failure and move forward

[admon]

@danbratu

are you sure about this?

All my experiences with OpenWrt show me that it is always a matter of “tinkering”. It follows the OpenSource and Linux idea, you have to bear that in mind.

@danbratu

please stop being glinet’s advocate. admit that flint2 is a major failure and move forward

I like the Flint2 and get on well with it. It does its job. Whether I have 1300 Mbit/s in Wi-Fi or 550 MBit/s is completely irrelevant to me.

The device has exactly one task for me: getting Internet via PPPoE, putting Mullvad VPN on it and distributing the whole thing in the home network. No creepy “file sharing”, “printer via USB”, “SQM” or any other stuff. Sometimes I use ZeroTier between 2 GL-devices so I can access my Jellyfin media server from abroad - but that’s all.

[danbratu]

than it should be advertised this way and not with all the extra features. i am using glinet products same way but general consumer buys it also based upon advertised features.
i’ll ask you a simple question: apart from 4.5.2 firmware (which is the only one kind of stable) why the rest of the versions are advertised as stable when they have so many issues? glinet doesn’t sell their items only to linux geeks they’re selling it and target any general consumer without knowledge of linux etc hence the only firmware marked as stable should be a stable one not sometimes stable. anyway, please just ignore my future posts and please stop replying my posts as it is very clear to me you are somehow affiliated to this company and you act as their advocate (as anyone can clearly see it)

[admon]

@danbratu

apart from 4.5.2 firmware (which is the only one kind of stable) why the rest of the versions are advertised as stable when they have so many issues?

I am not aware of “big issues” tbh. Yeah, 160 MHz band seems to be some kind of defective right now - but all other features work for me. Goodcloud has hiccups, fine, but what “big issues” do you talk about?

The main problem is that Wi-Fi is a terrible technology, and it depends on the receiver and sender to get a good connection. As I said, 160 MHz is broken for me using my Intel(R) Wi-Fi 6 AX201 160MHz - but after I changed it to 80 MHz it’s fine.

Would I call the firmware “unstable” because of that? Nah. Are there issues? Yep.
But none of them are terrible or make the product unusable. At the end it’s still IT and IT works like this. There are bugs, there are issues … that’s totally normal.

@danbratu

it is very clear to me you are somehow affiliated to this company and you act as their advocate (as anyone can clearly see it)

Nope. I like the products and I like that normal people are somehow involved in feature requests, and you can talk to the support directly by just mentioning one of the GL.iNet staff here. That’s much better support than with many other manufactures.

I like the fact that it is a living product. That has advantages and disadvantages. I can understand if someone calls it “unstable” - I don’t feel that way.

Transparency note: I don’t feel obliged to provide you this information - but since you think I’m connected to GL.iNet, I can tell you frankly: not at all. I received 2 routers from GL.iNet as a gift. One XE3000, so that I can look into the problem with roaming and Wi-Fi channels within Germany, and a Flint2 because they accidentally sent it to me as well. I did not receive any other payments, discounts or other bonuses. I bought the Flint2, I’m currently using myself, as an Early Bird. My Flint and Opal1200 are both fullprice products I bought myself

[Almahadeus]

@admon

Wi-Fi is simply garbage technology

Can you also please stop repeating this false assertion? You keep going on about how WiFi is such a terrible thing. The 2.4GHz SSID on my 7 years old tiny Mango router reaches out to two houses down the road with no problem whatsoever. There is absolutely nothing wrong with WiFi technology and the problems are strictly with this router’s design, software or hardware.

[danbratu]

just ignore him. he’s just admitted he’s somehow affiliated (he will never admit he’s actually on their payroll) to glinet so whatever he posts, his credibility is near zero. obviously stating that wifi is garbage is a non-sense (not as stable as hardwired, of course, but reliable when using proper tested and working as expected hardware - all my APs are rock solid with zero issues, including 2 x flint1 4.2.1 firmware). also stating that glinet products are mainly addressed to linux geeks is another non-sense (glinet doesn’t advertise it such as).

[admon]

@Almahadeus

You keep going on about how WiFi is such a terrible thing. The 2.4GHz SSID on my 7 years old tiny Mango router reaches out to two houses down the road with no problem whatsoever.

Wi-Fi is much more complicated than just range. 2.4 GHz is indeed pretty stable because it’s an more or less abandon standard. It‘s plenty of years old.

The main problem with Wi-Fi is that it depends highly on the devices and the general environment. That‘s nothing you would find with wired networks.

As I said in my first entry: Yep, there are issues with Flint2 now - but it’s under investigation. So just wait. Complaining won’t help anyway. Instead you should provide some diagnostical information: Your settings, your logs, your device configuration and so on.

@danbratu

he will never admit he’s actually on their payroll

Damn, now you got me. Time to ask for an raise.

[Kit]

Many thanks for the various feedback Very interesting to hear the views. On balance I have decided to return it while I can. I paid the equivalent of 200 US dollars and for that I expect it to be very close to perfect. Perhaps I expect too much!

[Almahadeus]

@admon

Yep, there are issues with Flint2 now - but it’s under investigation.

GL.iNET (and indeed their beta testers) have then failed in their most basic of tasks and should stop selling this unit until the issues have been completely sorted out (and that is coming from a GL.iNET fan and current user as well). I have said it before and I will say it again: they keep on pushing new devices to the market chasing profit with no proper testing while forgetting to polish their firmware on the pre-existing routers which have been losing support left, right and centre (my x2 Brume 1 routers are just an example). Broken features were things like multiWAN, VPN policies, shared storage, tailscale, DFS channels…etc which you have argued might not be essential features for all users but problems with wireless radio is just taking the mickey.

[admon]

@Almahadeus

they keep on pushing new devices to the market chasing profit with no proper testing

Yup, I agree. Better testing would be more helpful. Same for pushing all firmware features to all devices instead of having a plenty of versions.

That might be the con of not developing everything by your own.

[slesar]

Felling like kiskstarter backer with new technology product but still not stable technology products

[danbratu]

bro you don’t deserve a rise… quite opposite, if you’d be one of my employees i’d consider firing u.

[xize11]

I disagree that it is the beta testers fully at fault, sometimes also a beta tester hits the wall or a bottleneck of reachability even if it was not direct in a personal attack, but it was brought more ‘lightly’, in other times it just didn’t came up and it was configurated under their normal configuration.

I’m honest but since the mwan3 introduction i’ve had nothing than problems with connections and dropping connections, then kmwan came and other issues got introduced.

Slowly advanced users got kinda felt gaslighted and not always got heard correctly they just want to vote for stability as result it either got stashed for later or it got bandaided, and the community wanted more features pushed.

With other words you have two groups of people you need to keep satisfied at the same time and demand, and thats not realistic.

Also theres still kinda a backlog from sdk 3 and older bugs and newer features, like that the gl ui is still not responsive with the vlans, if you read the 2023 features topic, alot is still on their todo list or under consideration, i think its still a long overhead people tend to forget, but some of these features still come from the era of gl sdk 3.

What people sometimes also forget is that gl-inet also has a small team of developers , i think its really a demand issue of trying to make both groups happy which does not always work if the base is not solid due the overhead still.

I mean i can relate to this, ive been programming in java alot in the past, and abstraction, also abstract rules like having stronger ruled version control, polymorphic code is really important to make the code base as extensive as possible, and when vlans was not key for this design, or custom networking then it opens the door of many more problems and it will take time to make a code design working for something it was not intended to.

^ and besides that, then they also have to port OpenWrt designs, like iptables to nfttables, i can see already a issue because iptables can wild card domains, nftables cannot.

So i can see it from both sides beta testers to developers, sometimes it should be just a: sorry, no.

But on the otherside, it can also be a marketing issue or challenge, OpenWrt is often for people with atleast a little network/tinkering skill, but the routers are advertised like normal routers, it can be also a audience issue it targets, this makes times to understand a bug also more complex and tasking and in other situations a fix got worse for other people, i still think its not wrong to have a bug tracker and a format so this behaviour can be in more control.

For now its still a tinker router for me but thats maybe also because i go advanced

For the wifi its just a unfortunate situation, but i don’t see it as a beta testers or devs fault perse, you cannot always forsee everything and every situation, filogic tree did not even existed then in OpenWrt i believe this name transition took place in the middle of the testing😅

And theres much more going on with mediateks drivers and routers, the ax6s also have all sorts of issues but its a different chip, but it was stable for me in the past.

And yes i was beta tester , and no im not here for personal attacks or flames, i speak for myself, maybe i don’t always agree, that doesn’t mean i appreciate the effort and hard work from gl-inet and their fellow beta testers

[danbratu]

agree to you up to a point: there’s nothing to appreciate or to thank at/to a beta tester or a developer (you are getting paid for it - or getting free products). an end user pays real money for a product advertised as being good enough to hit the market therefore a customer will expect to get what he/she pays. it is that simple. if a product is faulty, don’t sell it (atm pppoe, wifi, vpn policies etc - all being basic needs - are faulty so…)! period

[xize11]

True and you are not wrong, thats why i think the problem is also how it has been advertised/marketed👍

I’ve also bought a second one myself, but without the expectation it will be a full product (comming from a flint 1 and before a mango v2 ).

For me it always has different quircks, like i also sometimes have on routers with vanilla OpenWrt (older tp-link 1043nds also had these times i always wifi disconnected).

But i totally understand how someone looks at it for their first router

[Almahadeus]

The beta testers here are all now sounding like they are cosying up to the company and coming up with all sorts of excuses to justify its poor performance and strategy. When I was a beta tester for the GL-AXT1800, I was one of the first to point out to the significant flaws in its repeater function and handling WiFi channels and DFS problems but my constructive criticism was only falling on deaf ear before I seem to have fallen out of favour with the developers. It is a big shame that they have gone for blind profit before developing the fantastic product that we were all hoping for.