Well you have to state why it wasn’t good. It was incredibly region-dependent, but if you live near one of their endpoints the latency wasn’t noticeable and the quality was great, as it was for me.

In the end I got to play a bunch of games for free, and have an extra controller I still use, so there’s that. They made us whole, at least, after they shut down (I even imported my into the breach save game into Steam with Google takeout after)

By some definition. They have always been usable to some degree because I think animators or something use Linux commercially on Nvidia, and for gpgpu they are still top class on linux (nothing comes close)

They haven’t always been the best for gaming or desktop (Wayland) use though, since Intel and AMD opened up their drivers.

Arguably in my experience Nvidia has been far less buggy for the last 30+ years on x11, and with this change they may have finally reached parity on Wayland, haven’t tried it myself.

Depends on what you’re doing with it, but prompt/context processing is a lot faster on Nvidia GPUs than on Apple chips, though if you are using the same prefix all the time it’s a bit better.

The time to first token is a lot faster on datacenter GPUs, especially as context length increases, and consumer GPUs don’t have enough vram.

I guess it’s finally to the point where selfhosters can admit to using k8s and not be bombarded by comments saying it’s overkill, which has happened in the past for:

• Self hosting at all
• Using VMs
• Using containers
• Using docker compose
• Using k8s (⬅️ I guess we are here)
• Using helm charts or whatever ends up replacing this

Anyway, I believe there is a tool also to turn docker compose files into k8s manifests if we want to take this a step further!

Kitty supports images, not sure about alacritty, although there are many competing protocols for image display in a terminal emulator, so it could be that it just doesn’t support a particular program.

I use sunshine and moonlight. It’s designed for games but works far better because of it, as in if it’s good enough for games, the latency will be far better than other RDP protocols.

It doesn’t do clipboard sharing though.

If you’re a tinkerer it’s kind of addicting. I thought I’d give it a try just to see what it was like, and ended up staying up all night customizing it, and now about a month later I don’t really want to go back to KDE (been using KDE for almost 20 years)

If this can handle routing 10g this is a great choice to use as a router. It’s actually quite difficult to find a gateway that’s around this price and ISPs (at least here in Canada, or my part of Canada) are offering internet over 1Gbps at the same price as gigabit, but their routers are awful.

I didn’t measure performance, I was talking about battery life, but no, I didn’t do any benchmarks.

Even Intel has these. I think this patch set goes a bit further and takes into account the silicon lottery differences between cores (according to the patch series)

I’m using the patch set on my framework 7840u and didn’t notice a difference though, though this is really YMMV.

Proxmox is a lot more user friendly than virt-manager (yes I’ve used both, but I just started using proxmox).

Right, I know EFI images are stored in the EFI partition, but with secure boot, only signed images can be executed, so they’d need to steal someone’s signing key to do this.

That’s… Stored in the EFI partition or changeable in userspace?

So I don’t get it, I have my entire boot image in a signed EFI binary, the logo is in there as well. I don’t think I’m susceptible to this, right? I don’t think systemd-boot or the kernel reads an unsigned logo file anywhere. (Using secure boot)

Have you tried playing with the flow control settings on your ubiquiti switch? I was having problems streaming video games until I toggled that setting (forget if it was on or off, though).

You didn’t have to tweak PIPEWIRE_LATENCY or adjust the latency in guitarix? In my setup the latency isn’t great out of the box.

I’ve used Linux for over two decades (red hat to Gentoo to Ubuntu to arch) and I must say it’ll be a tough sell to get me back to an RPM or a debian based distro solely due to how god awfully slow the package managers (dpkg and rpm) are.

Since Docker came along and brought with it the ride of Alpine and APK, it made me realize that system upgrades on a modern processor, fast internet, and an SSD should take seconds, not minutes.

Why’d you create your own dockerfile repo vs just improving/changing the one in the main ooba repo?

A lot of people thought this was the case for VMs and docker as well, and now it seems to be the norm.

As per the link (YouTube) in the other thread, it seems like iGPU + increased allocation of VRAM is better than using the CPU, though it also seems APUs max out at 16GB. Maybe something AMD can improve in the future then…