Do you ever change proton version of games if they run for performance reasons, or do you just stick to the default config?

For example, let’s say a game is about a year old, so it would run on an early proton 9. Might changing to GE 10-8 improve performance and battery runtime / FPS?

I started thinking about this when I gave a shot to Wuthering Waves. It runs out of the box, but GPU is a little bit less under pressure with GE 10-8.

What’s your take?

    • cryptTurtle@piefed.social
      link
      fedilink
      English
      arrow-up
      4
      ·
      12 hours ago

      I never notice a performance difference. Just that (for whatever reason) some versions are more stable with certain games

  • Fubarberry@sopuli.xyzM
    link
    fedilink
    English
    arrow-up
    8
    ·
    13 hours ago

    I put everything on Proton Experimental all the time, and only change to an older proton or GE if I’m having a specific compatibility issue.

  • dinckel@lemmy.world
    link
    fedilink
    arrow-up
    3
    ·
    13 hours ago

    Some versions can have an occasional boost or regression in performance, but if you’re going to swap, it’s for compatibility

  • smeg@feddit.uk
    link
    fedilink
    English
    arrow-up
    3
    ·
    14 hours ago

    Default unless it seems broken or protondb recommends otherwise

  • lerky@lemmy.blahaj.zone
    link
    fedilink
    arrow-up
    2
    ·
    13 hours ago

    I have in a few very specific cases, though I’m not sure whether you’d define that as performance or compatibility. Also, the versions I use in those situations are so old that significant differences between them could be reasonably expected. Nowadays we’re at a point where there probably won’t be a noticeable difference.

    Normally I just use the latest GE, in part out of a misguided superstition that newer may equal faster. If issues (usually video codec-related) crop up I test with a handful of older GE versions (7-55, 8-25, and 9-2) that I’ve found to be “magic bullets” of sorts for common issues I’ve run into due to regressions that exist in later builds.

    No harm in doing your own testing though. Well, unless it’s a Denuvo’ed game since you can potentially max out your activation allocations from swapping Proton versions.