I hear you on the install script being scary. I currently don’t have any other distro-agnostic way I know to handle the install, especially for a system service like this. Bash is universal 
With any online install script, it’s good practice to review it before running it. As far as the app goes, the code is open, and CI builds the binaries with a checksum that is also available in the same release assets.
FInally i can give this a try on a linux,
I already found 1 issue. It reports my system having 62GB RAM, but it actually got 64GB. OS itself is reporting 66,7GB available under system monitor within linux os. But under settings > system > about shows 64GB RAM.
It only reports what really is available for Ram, your missing 2 GB are reserved for the iGPU.
Thats not a Bug. Mine is reporting 55GB Ram as i set my iGPU to 68GB Vram and the app rounds down. 64GB is just 63.xGB
i do not agree, my iGPU is set to 0,5GB ram so it should still say 64GB available.
SInce both OS is showing 66,7GB and about shows 64GB. only this fan controller thingy is saying 62GB which is wrong and mostlikely a bug.
Do you mean you just migrated to Linux as your main desktop? Windows is beyond broken right now and I would not use it if I didn’t have to work 
Yeah i did. I hopping on and off every year to check, but comming back to windows due to programms i need sonetimes or games that don’t work. But now i am committed to stay Linux 
I’ve read many folks in this community use Linux over windows, and I think one of these days I might just make a secondary drive a version to play around with. I think I’m just too used to windows, its easy and comfortable/convenient to me as I’ve always used it. But I’m not opposed to learning something new. Only issue I think I may encounter is my current wifi7 card is not supported on Linux 
One issue I’ve ran into with the Windows version of the app is that the power control (via RyzenAdj) does not seem to work once I launch a game. I suspect this has something to do with AMD’s Adrenaline stealing power control for itself, but I don’t know for sure. I can’t test if that happens on linux because the feature’s not available yet.
On Linux last night while playing Overwatch on the Framework Ctrl custom-default fan curve (where it caps at 80% fan speed), I encountered the 544mhz CPU issue. As such, it doesn’t appear that my device (with known overheating issues) is a fan of lower fanspeeds for “real” workloads.
I have noticed previously that it would totally shutdown RyzenADJ when I opened battlefield 6. I think it’s more so having to deal with the anti cheat flagging RyzenAdj
Yup, that’s the game I was testing.
@Kemal_Ozturk What temperature sensor are you using to represent dGPU temp? Mangohud and HWInfo64 is saying my dGPU is ~75c, but your app is saying high 90’s and seems to be doing a much better job of tracking with performance throttling. I’ve been spending a lot of time troubleshooting CPU overheat issues, when really it might be my GPU overheating if your temp sensors are to believed.
Edit: looks like this only happens in Linux. I don’t see the same behavior on Windows.
I just tried the linux version on Framework 16: it perfectly works!
Thank you for your job 
Dew it! No better time to try it out. I’m pretty sure people on this thread can suggest some Windows like distro so you don’t feel alienated on your initial hop. My wifi works pretty well, but I’m not sure about the new wifi card. I have the old one that came with my old mainboard.
What was your temps looking like? Capped at 100C or something? I know some BIOS release had that 544mhz issue but I thought they fixed that since then.
I’m just reporting whatever the framework_tool CLI is reporting. And the CLI is talking to the EC directly so I would not be surprised if it’s more accurate.
If it’s more in line with when the throttling happens, it’s safe to say the high 90 ones are the real temps. Even if it’s a sensor issue, the laptop thinks it’s overheating and is throttling down. I doubt it would be a sensor issue though.
Not sure, I wasn’t keeping an eye on it (for once, lol) but I assume it just overheated. I’m not sure any FW16 would overheat in that scenario with the same fan curve, mine seems to have an overheating issue.
I wonder if framework_tool CLI is capturing a different sensor for windows vs linux usage. If I didn’t have such a unique thermal situation right now, I doubt I would’ve noticed. Perhaps I’ll do some intentional testing an put in a bug report if it’s repeatable.
I’ve been doing some research and thinking about Bazzite for me personally to test. But yeah the original wifi 6e cards work, however, I replaced mine with a wifi7 for MLO and from what I have been reading it isn’t supported on Linux. Only one way to find out I suppose…
If you mean wifi 7 is not supported in linux then, that is wrong.
wifi 7 works on linux.
Since i own FW 16 with amd wifi 7 card and it works on most distro out of the box that i have tried.
I think they mean that their specific card isn’t supported, which is believable. I’ve owned many devices that didn’t have linux support for the wifi card, the speakers, or some other random component.