Swap is used to defrag ram on linux. Could be related to that. In any case this is pretty normal. I have 43Gi available and it is used 700Mi of swap.

I don’t think it should have a significant effect on performance. On the other hand gamescope is broken on my system right now. There are probably plus and minuses of both.

Great. now how long till we can game without nesting the game inside gamescope?

I am using a 4080 on Plasma 6 wayland. It works pretty well. DLSS works. I haven’t tried frame generation, but I think support was added recently. Occasionally I have a wake problem, but I believe that is bios related. I am encountering a gamescope blackscreen freezing issue.

Only stuttering I am experiencing is a Steam overlay bug and Bluetooth interference sometimes.

Compositor choice and how recent of release will effect your experience as they are all independent implementations and still improving.

I wouldn’t expect that a 20 series card to be a specifically buggier experience over the 40 and 30 series though.

Did you happen to enable Bluetooth LE-Audio mode for the Headphones in an app? If so you need to turn it off.

You probably just had a program(s) use a lot of ram and that pushed some processes to swap. When the system needs it, it will be moved back. I’d +1 on setting up zram. Can be setup easily on most distros.

My fingers don’t speak it is the problem.

Worst is when installing a new distro(usually in a vm ) and it defaults to nano and for some weird reason no vi of any sort is installed. I hated nano. Last time I intentionally used something like nano was the 90s with pine I think.

I suppose that is true. Intel seems to think so as well as their low power n100 is about the performance of a 1500x.

Sure, not much per gen, but if you compare say a 1700x vs the current 9700x, you are roughly looking at a 3x improvement in single and multicore performance increase.

My community college(1997) had a Suse linux computer lab that I learned on. It was mostly used as a networking/server and programming platform.

Loki was the leading porting developer at the time.

Until risc-v is at least as performant as top of the line 2 year old hardware it isn’t going to be of interest to most end users. Right now it is mostly hobbyist hardware.

I also think a lot of trust if being put into it that is going to be misplaced. Just because the ISA is open doesn’t mean anything about the developed hardware.

It isn’t as simple as just compiling. Large programs like games then need to be tested to make sure the code doesn’t have bugs on ARM. Developers often use assembly to optimize performance, so those portions would need to be rewritten as well. And Apple has been the only large install of performant ARM consumer hardware on anything laptop or desktop windows. So, there hasn’t been a strong install base to even encourage many developers to port their stuff to windows on ARM.

I could see developers using both the NVK and M1 drivers depending on which best suits their needs for hardware similarity. It is also interesting that both are not super opensource friendly hardware manufacturers. Good hardware, less so on openness.

Yeah. Only systems that can be interpreted in real time are viable. Not sure how recent we are talking either. On top of that, interpretation will be inherently worse on battery life.

A-shell

Hadn’t heard of a-shell. On my iPad I occasionally use iSH for bittorent, among other uses. I do agree it is on the slower side, but that is in part because of Apple’s rules about JITs and this forces iSH to use an interpreter. I do think an interpreter of ARM ISA would potentially be faster as the ISA is fixed width and x86 is not, but I don’t know if that would be margin of error faster or not.

Thanks for sharing.

Edit: I may have came across that before, but iSH is more flexible, if much slower.

Same here. With the exception of the explicit sync, which will hopefully be resolved this week, I have been running Plasma 6 wayland since February. And honestly when I tried the X11 version it had more issues.

Either way you’d have to look at the compositor as that is what handles input. I haven’t used Weston, so I don’t know where to start.

SDDM uses kwin_wayland. Plasma store the setting for that in $(HOME)/.config/kcminputrc I believe that is used by a different part that is not used by SDDM. Best suggestion is to submit a feature request. Having proper input support would go along with power management as a needed feature for SDDM on wayland.

Not sure if your distro version has a new enough version of systemd, but newer versions have a systemd-oomd service for that. It may not be enabled by default. On older versions you could try early-oom which is not part of systemd. OOM stands for Out-Of-Memory.