fsr 4’s algo was enabled on rdna 4 via hardware changes. people will argue that it works on rdna 2 & 3 via a leaked int8 path, often leaving out that:

• it’s not identical in terms of visual quality (even though it does often improve upon 3.X)
• most critically, it doesn’t generally offer the same level of perf uplift to make it worthwhile (though there are a handful of very good cases)

I suppose I’m kind of opinionated about the whole upscaling and mfg ‘meta’ of (mostly) pc gaming today, but I find it sad that this is what end users clamour for, rather than for games to just not run like shit to begin with.

can you elaborate on these issues

what’s wild is that you’d post this in Linux gaming (rather than a generic Linux message board), where technologies like display independent VRR, scaling and HDR become increasingly relevant.

Wayland is the standard today, in that

• The xorg development team moved directly on to it
• It’s the default for major distros and desktop environments

Arguing about this is an exercise in futility.

If you’re not trolling, I’m genuinely sorry that you’re wasting time and effort down in this rabbit hole occupied by shitty pundits preying on young, impressionable minds, claiming they wish to ‘strip the politics away from software’ when they’re busy doing the direct opposite of what they preach.

cool, I’m just gonna go ahead and enjoy my contemporary display technologies and improved security.

also mixed display scaling values and extended colour space support. Like there’s no argument against today. There were breaking changes as first, sure. The software ecosystem has more or less caught up.

can you tell me which device this is with?

edit: nvm i saw you posted it directly into the thread - gfx1103 = 780M iirc

no hurry at all, thanks a bunch!

Someone on the github issue thread has been asking for clinfo output on affected systems, theorising that client issues are caused by issues with vendor detection.

This is a bit of an ask, but if it’s quick to jump forward and back between ROCm releases on your setup, would you be able to pass your clinfo output from 7.2.0 into the ticket linked above? No worries if not

Good news, seems there’s an existing internal ticket covering this defect. It’s an application-side issue pertaining to libProResRAW.so.

I believe we’re shipping a workaround at the ROCm framework side in 7.2.1

discord was always kind of shit; we definitely didn’t need another blackbox in communications back in 2015.

would love it if people moved away to something like stoat (can’t wholeheartedly recommend element / matrix right now), unfortunately I don’t see it happening at any large scale, and maybe that’s not the end of the world.

many of us made a similar move away from reddit to lemmy and other activitypub-powered aggregators back in 2023. it’s small and niche but pretty lovely? we generally have pleasant experiences here.

sort of apples to oranges but maybe the same can be true for people who want to move away from discord for ideological reasons.

We’ll find out. It’s filed to the public issue tracker, which is a nice change of pace. Hope you can stay appraised of the progress from the following link:

https://github.com/ROCm/ROCm/issues/5970

have filed this to the ROCm team, hope to get someone looking into this tomorrow.

I don’t suspect it’s davinci resolve. This is likely due to rocm-opencl.

It’s a long weekend, so I likely won’t hear back from my colleagues until Tuesday at the earliest

Thank you for linking these. The second one suggests the following

Update: Turns out that this is happening because the update of opencl-amd package, the newest ROCm 7.2 drivers appear to not be compatible with DaVinci, I installed rocm-opencl-runtime from the extra repo since it still on 7.1.1 and now its working again… It also doesn’t work with opencl-mesa, it crashes launching a project

I’ll dig through to see what can be done about it.

Can you share the post here? Does it elaborate on what’s failing?

Oh shit, I’m sorry. I first played it back when it was called called geneshift with my colleagues. We were led to this after discovering that GTA2 had a PVP mode, and we just wanted to dick around at lunch time together.

I hope you can get the original 2 working someday! Genuinely had no idea it was unplayable today :(

might as well? My 10 will be arriving soon and I’ll be doing the same. Reverting to stock is also an easy enough process

I wasn’t aware this wasn’t playable (then again, the same applies to the first game), but you may be interested in checking out spiritual sort of reboots like skillshot city?

The graphene community in the past has pointed out Firefox’s incomplete content sandboxing implementation and suggested that other aspects of security are not up to chromiums standard. They pointed out other technical shortcomings as well, though I can’t recall them, I’m not sure how urgent they’d be.

This was several years ago, and I’m not sure if any of this has been addressed, but I wouldn’t like to rely on manifest v3 compliant ad blocking.

I get the impression that Firefox may continue to lag in this regard, and I don’t feel that people like us are made vulnerable by this, though I do worry about people like my parents.

yeah, they do that 😅

where do they find the time?!