Gentoo. It will be the last distro you hop to. Because it’s whatever you want it to be. Don’t be afraid. It even has a special command and portage repo to install all the support files and ancient libraries from 2002 your old games need in one shot.

You can snag a binary kernel, browser and some compilers now too if you don’t want to deal with that. It’s not much more difficult than Arch nowadays.

The Black Mesa HL remake took 16 years to release (2004-2020).

Pydio and Seafile are alternatives I’ve tried. Pydio was pretty fast too. I agree with you on Nextcloud, I want to like it but I inevitably start having issues and it’s slow even after tuning. It just tries to do too much and shouldn’t be that complex to spin up a file server.

Don’t reboot it, just patch!

Thanks!

As I understand it, it bind-mounts the /dev/nvidia devices and the CUDA toolkit binaries inside the container, giving it direct access just as if it was running on the host. It’s not virtualized, just running under a different namespace so the VRAM is still being managed by the host driver. I would think the same restrictions exist in containers that would apply for running CUDA applications normally on the host. Personally I’ve had up to 4 containers run GPU processes at the same time on 1 card.

And yes, Nvidia hosts it’s own GPU accelerated container images for PyTorch, Tensorflow and a bunch of others on the NGC. They also have images with the full CUDA SDK on their dockerhub.

I find I can’t get into them if I physically play them on PC. One look and I get the “potato graphics” feeling and start increasing AA, texture packs, etc. Now if I stream it to my TV or phone, it’s like “oh okay this is normal for this thing” and it tricks my brain to not caring.

No just LTO. Right now only Ubuntu, Fedora and SUSE Tumbleweed turn it on by default.

I’ve rebased a few of my containers with SUSE and noticed some improved load times on my web services as well. I don’t run anything demanding either, just bored. It’s like half a second improvement lol.

I try to find ways to make my setup more bulletproof or faster whenever I get the itch. As an example, I recently switched to OpenSUSE and Podman to take advantage of the LTO optimized packages and rootless containers.

I tried to run my online life through self hosting but I found a lot of the services weren’t reliable or capable enough to get real work done. So I went from 30 containers to about 7 and have a lot less to tinker with.

They have a Dockerfile that enables hardware accelerated transcoding. It’s not available on their Dockerhub unfortunately. Works great with Nvidia.

In addition to what the other person said, it can perceptually identify videos which makes tagging your library a breeze.

Great! A relatively recent development is dist-kernel and has greatly simplified kernel installation.

I’ve been running a server with the same Gentoo install since 2016, still stable as ever. Using OpenSUSE Tumbleweed on the desktop though, really liking it so far.