ricklinux wrote:
Not a waste of time and I’m not suggesting trying the rt3090 github instructions over the other one. It’s just information that ultimately you’re going to have to decide where you go with it.
Good initial point. This thread can be helpful to those others who have older systems with Ralink chip-sets. Although …
ricklinux wrote:
There aren’t too many options with that particular chip on Arch at least.
And probably not many Ralink chip-sets still around, though the brand was quite ubiquitous at the start of the past decade. After your MediaTek comment I checked the date of that company’s acquisition of Ralink: 5 May 2011. No wonder when I clicked a link for the latest Ralink chip-sets I reached instead a sign-up page for the MediaTek newsletter. Perhaps MediaTek was erasing competition.
ricklinux wrote:
I don’t know about other Linux versions. As far as PopOS goes it’s a solid distro.
I happened upon instructions regarding RT3090 and RT3290 similar to GitHub’s ranging from Linux Mint to OpenSUSE forums. Back in 2012 a fellow installing Arch described my exact problem but it was during the Live USB. Upon installation the problem was gone. That supports my supposition that continuing improvements with the kernel are driving the problems with this old chip-set. For example, this post from 2011:
Apparently the acer_wmi module is loading at startup which is interfering with the wireless, making it think the switch is constantly off.
Removing and blacklisting this package has resulted in me being able to connect to wireless networks
Followed by a “final note” that drives my current approach:
... compiling RT3090 from source and installing apparently doesn't work no matter what I try.
The built-in rt2800pci module actually works the best here. Link is limited to Wireless-G speeds however (54Mbps).
At least now, though, I can say it works, and is stable.
This person also had an ACER but rt2800pci troubles cover other brands from HP to Lenovo and many more. The Ralink problem appears exclusively with Linux. By 2014 users appeared either unable to solve RT3090 and RT3290 failures and partial failures, or only able to achieve partial fixes.
