github.com

semperverus, to linux in LACT: Linux AMDGPU Controller for overclocking and fan curve control
@semperverus@lemmy.world avatar

Sad to see that it’s GTK-based. Cool concept overall though, I may give it a shot at some point to compare to CoreCTRL

isVeryLoud, (edited )

As a GNOME user, I’m happy to see it’s GTK, but idk why it’s not Libadwaita.

We’re never gonna get one cohesive UI on Linux, it’s always gonna look out of place somewhere.

If the backend is not tied to the UI, it might be easy to reuse the work for a Qt6 front-end.

Lodra, to linux in kando: 🥧 The Cross-Platform Pie Menu.
@Lodra@programming.dev avatar

Has anyone here tried Kando yet? Looks like a very nice tool. How does it compare to typical keyboard shortcuts?

Chewy7324,

This project is currently in a very early stage of development. Kando is not yet a functional menu but rather a prototype which demonstrates the feasibility of the concept.

Since Kando is still in early development, it might be a good idea to look at the Gnome Extension Fly-Pie. It’s from the same developer and it looks like Kando will be similar.

Railison, to linux in kando: 🥧 The Cross-Platform Pie Menu.

Getting Sims 1.0 nostalgia

oscardejarjayes, to linux in LACT: Linux AMDGPU Controller for overclocking and fan curve control

CoreCTRL, but written in Rust? Based. I’ll try it out when I have the time.

TerkErJerbs, to linux in LACT: Linux AMDGPU Controller for overclocking and fan curve control

Dope, thank you for posting. Been using ‘Core CTRL’ for quite awhile but Imma give this a rip. For some reason the former tool never could control the fan curve for my GPU (all the other fans in the box worked fine with it) so this might be profit.

gary_host_laptop, to linux in kando: 🥧 The Cross-Platform Pie Menu.
@gary_host_laptop@lemmy.ml avatar

I wonder what’s the difference between this and Fly Pie, like, why did he stop developing the first and started the latter.

min_fapper,

He says so in the readme

Chewy7324,

I have been working on Fly-Pie for more than 3 years now and I am very happy with the result. However, I have always wanted to create a similar application for the desktop in general. This is why I started this project.

WheelcharArtist, (edited ) to linux in LACT: Linux AMDGPU Controller for overclocking and fan curve control

is there any advantage over corectrl or is it simply another tool?

Chewy7324, (edited )

The big advantage for me is that lact runs as a (systemd) daemon. This is more convenient for me than having to autostart CoreCtrl.

A disadvantage of the daemon is that it can’t be packaged on flathub.

Enable and start the service (otherwise you won’t be able to change any settings):
sudo systemctl enable --now lactd
You can now use the GUI to change settings and view information.

LACT has an API over an unix socket.

github.com/ilya-zlobintsev/LACT/blob/…/API.md

NixDev, to linux in sigoden/argc-completions: Autocompletion for any shell and any command.

Have you used this? Is there any benefits over bash-completion?

Chewy7324,

I’ve not used it yet, but I found bash-completion to be lacking quite often. Completion is one reason I’m using fish atm.

But from the looks of it it’s exactly what bash/zsh is missing for me.

cashews_best_nut,

Have you tried ZShell? I’m consistently amazed at the number of great plugins that get made for it.

Chewy7324,

I’ve used zsh for it’s support for posix sh and have my config. But I find fish to be faster with the features I want and it has those features ootb.

Maybe I’ll give zsh another try.

TheButtonJustSpins, to linux in kando: 🥧 The Cross-Platform Pie Menu.

I’m currently eating pie for breakfast, so this confused me at first.

SpaceNoodle,

Is cheesecake a pie?

NixDev,

I vote yes

TheButtonJustSpins,

Agreed

radioactiveradio, to linux in Shadow Cast v0.6.1: GPU Accelerated Screen Recorder - Now with Wayland Support

Wow you’re fast, beautiful goofy picture guy.

sonymegadrive,

<3

dbx12, to programmer_humor in Someone has started answering to the github stalebot with memes

The stalebot is most times useless. The only scenario where I can see use of it is a maintainer waiting for the reporter to add information. But closing issues because no maintainer checked on them? That’s garbage and discourages bug reports.

kevincox,
@kevincox@lemmy.ml avatar

But they get scared because their program has 500 bugs! Close them and now your program only has 10 bugs! Problem solved.

/s

dbx12,

absolute galaxy brain moment

Potatos_are_not_friends, to programmer_humor in Someone has started answering to the github stalebot with memes

After a extremely long week, I sometimes participate in open source. I have to deal with malicious commits. I have to follow up on issues from misguided individuals who are actually looking for tech support. I have to guide new contributors to how this massive repo works and to submit tests. I have to negotiate with the core team and these convos can often last months/years.

And contributing to open-source is one of the few things that give me pleasure, even if it’s a extremely thankless job.

But I’m tired man.

I’m not dealing with low-quality memers who are providing zero value. Nor should we encourage it.

Anders429,

I would argue that in this case the maintainers are in the wrong for not even responding to the issue, not the reporter responding with memes.

db0, (edited )
@db0@lemmy.dbzer0.com avatar

I do FOSS as well, but I’d rather people have fun punting the stalebot than just keep repeating “this issue still exists”. I will probably get a chuckle out of it.

Aatube, to programmer_humor in Someone has started answering to the github stalebot with memes
@Aatube@kbin.social avatar

They shouldn’t even be using the probot, it’s deprecated, unmaintained and thus potentially vulnerable

Deebster,
@Deebster@programming.dev avatar

Also just the whole concept is wrong and encourages “me too” spam just to keep the thing from timing out and not being fixed.

Aatube, (edited )
@Aatube@kbin.social avatar

I actually see a legitimate use case for it and helped add the actions version in a project where I'm a collaborator.

Quite a bit, certain bugs disappear after an update without us targeting it (partially because the logs get fudged a bit after going through dependencies, so sometimes multiple bugs have the same cause or it's actually a dependency issue that got fixed) and sometimes we forget about old feature requests.

The stale reminder doubles as a reminder for us to (re)consider working on the issue. When we know something probably isn't gonna get fixed suddenly, we apply a label to the issue. For enhancements that we'll definitely work on soon™, we apply help wanted. We've configured the action to ignore both. We also patrol notifications from stale to see if something shouldn't go stale. This is a medium-sized project so we can handle patrolling and IMO this helps us quite a bit.

Deebster, (edited )
@Deebster@programming.dev avatar

Fair enough; I didn’t consider artifacts like logs and traces. I suppose a stale marker might prompt the original reporter to retest and supply fresh ones (or confirm it’s fixed in the dependency case).

In an ideal world I suppose we’d have automated tests for all bug reports but that’s obviously never going to happen!

canadaduane, to linux in Shadow Cast v0.6.1: GPU Accelerated Screen Recorder - Now with Wayland Support
@canadaduane@lemmy.ca avatar

Is it possible to get this to work with OBS studio? I see the author mentions OBS as an “Alternative Project” but it seems ideal to have these pieces work together.

Strit,
@Strit@lemmy.linuxuserspace.show avatar

OBS already does screen recording on Wayland…

julianh,

Is it GPU accelerated?

FaeDrifter, to linux in GitHub - Acly/krita-ai-diffusion: Streamlined interface for generating images with AI in Krita. Inpaint and outpaint with optional text prompt, no tweaking required.

This is suuuper cool, but looks like having linux+amdgpu limits me to the cloud option.

I supposed this is bc we don’t have a DirectML equivalent yet.

pantherfarber,

I got it to work yesterday. Have to go into the python venv it installs, remove torch and install it the way it describes on the comfyui GitHub.

wim,

There have been some efforts to run pytorch and StableDiffusion on ROCm. Not sure if that could be combined with this.

warmaster,

Crap, I was hoping to try it. I wonder if AMD will announce something in their FOSS / AI event.

wewbull, (edited )

It works today. Only problem I have is the memory management is pretty poor, and it’s pretty easy to run out of vram.

Rx7600 8GB + 5900X Rocm 5.7.1 Pytorch 2.1

wim, (edited )

Interesting! Got any links that explain how to set it up?

I just got a laptop with an RX 6700M 10GB ans am eager to try it :)

wewbull,

Not really. I’ve had to do quite a bit of experimentation.

My setup that I’ve settled on:

  • Rocm system libraries from Arch Linux
  • PyTorch nightly for Rocm pip installed into a venv (see instructions on pytorch homepage)
  • Set HSA_OVERRIDE_GFX_VERSION to 11.0.0. This is just for the RX7600 and it tells it to use the RX7900 code as the pytorch version hasn’t been compiled with 7600 support.
  • Start software.
wim,

Thanks!

byteseb,

Hmm, that’s weird. I was able to run Stable Diffusion locally with Linux + RX6600.

Probably because I used Easy Diffusion. At first, I couldn’t get the GPU acceleration to work, and I was constantly running out of RAM (Not using VRAM), so my system always froze and crashed.

Turns out it was a ROCM bug, that I don’t know if it’s fixed by now, but I remember “fixing it” by setting an environment variable to a previous version.

Then, it all worked really good. Took between 30 seconds to 2 minutes to make an image.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • localhost
  • All magazines
  • Loading…
    Loading the web debug toolbar…
    Attempt #