Opinions / Experimental data on Arkane Linux


optimized to maximize performance on modern hardware by targeting the x86-64-v3 CPU microarchitecture level.

For a distro claiming an

experimental mindset

… I’d expect benchmarks front and centre to buttress the performance claims, but I don’t see any.

Has anyone tried it? Are there benchmarks anywhere?

I have seen this distro used in the wild in a protondb report, so it presumably has more than one user.

I don’t expect such optimisations to make much — if any — perceptible difference in most workloads, but I’d be happy to be shown wrong.

This reminds me of CachyOS.



Since there’s a general trend in this direction - Tumbleweed already announced, Fedora is planing, Arch is planing (but there are already ALHP, CachyOS, this distro …), and Gentoo always existed - there are existing benchmarks. So you can just search for x86-64-v3 Linux benchmarks in general.

Afair there’s an indicated uplift of roughly 5 to 15%, with some software being slightly slower and some benefiting greatly.


there are some benchmarks linked here:

summary: show varying results, from no change to rather significant speed ups.

and Arch Devs are openly discussing going for x86-64-v3 optimization, too:

I am also running core system packages in optimized variants for x86-64-v4 (Zen4 here which supports AVX512) , namely the kernel, mesa and wine-staging. It does not hurt performance, but there are some cases where I get more FPS in games with the optimizations vs when I don’t use them (minimal but noticable).
If Arch offers a way to get all packages in optimized variants, I would definitely use them - does not hurt, but can improve things.

Edit: And a link for those not knowing what x86-64-v3 v4 mean:


I think you pasted the wrong link. This leads to an unrelated topic on “Python and xslt”.

thanks, missed the 3 in the end

1 Like