Latest 30 Comments

Lutris has always been the worst option for people wanting to game on Linux, honestly. It was a great project, that didn't keep up with the times and honestly had a god awful UI and UX.

I used Lutris until I discovered Heroic and haven't looked back. Heroic isn't perfect, but it's developed by people who listen to user feedback and actually made it a pretty good experience (and I believe even have a partnership with GOG!). Lutris dev in my experience look at github issues and around online always seemed a bit caustic to me so I didn't engage much there either.

In a way, sad it's gone this path as it's one less option, but I think there was a track record that showed this could eventually happen.

Hopefully someone will fork it or some more projects pop-up, variety is the spice of life after all.

View this comment - View article - View full comments - Report this comment

My two cents is that the feedback is a bit harsh.

When you think about it, what is the overlap between those objecting on AI for copyright reasons and those downloading roms or pirated games, seems to me that’s not a strong argument?

The ai models in question have been trained on existing works but nothing precludes using models trained on permissible works, these are already being developed (maybe available), so let’s push for better training ethics instead?

The energy argument is an argument for renewable energy, there is enough to power the world, I can find some links, we should push for energy ethics?

Code quality standards and checks don’t change just because of who or what wrote the code so are folks unhappy with the bugginess of the project?

Otherwise I think there is a reasonable argument to make that the dev needs more coding resource than the community provides. Think people can disagree with that and support the project but just because ai is used in a process doesn’t mean vibe coded slop, the dev clearly understands the codebase so can safeguard code quality.

View this comment - View article - View full comments - Report this comment

To those uninstalling Lutris, come over to bottles! Its okay-ish, needs work, but I've gotten to play several games, use their archive backup function to move and restore games offline as well. (And I hope doesn't also have the AI problem)

View this comment - View article - View full comments - Report this comment

Oh no, the download button and shell script wrapper can no longer be trusted to handle my login credentials, if only there was something I could do about it, like, say, just not use it.

On a more serious note, I have not found myself using Lutris or tools like it (Heroic) in ages. As a library to launch titles from they don't work for me because the design does not suit my taste (trying really hard not to be mean to people who like to have their entire screen being taken up by rounded corners here). As a place to download the installers through they don't work for me because my accounts keep being logged out. As a tool to handle the configuration to make a game actually playable they don't work for me because the scripts, if present at all, keep being outdated or specific to some hard- or software I don't have -> simply don't work. As a tool to manage Linux native games they add no value for me as I still have to manually reinstall dependencies after a SteamOS update. And finally, even the add to Steam button doesn't work reliably.

More trouble than it's worth. Also grrrrr, AI!

View this comment - View article - View full comments - Report this comment

Hope this project sinks completely because of this.

View this comment - View article - View full comments - Report this comment

Quoting: scaine"I'm doing a thing people hate, so instead of not doing that, I'll continue to do it, but hide it better"

That's a bold position to take in any project, let alone a FOSS project. And just because he can use it in non-slop manner (maybe? hopefully?) it doesn't absolve him of the ethical concerns many of us have about genAI.

You can use in a non-slop way, carefully reviewing the generated code and making adjustments and refactors. Something that he can do as an experienced developer. It would be different if he was blindly accepting the AI generated code.

View this comment - View article - View full comments - Report this comment

Goodbye Lutris, there are plenty of alternatives to what you do, who either don't use AI slop, or at the least aren't ***holes about it.

View this comment - View article - View full comments - Report this comment

He said some "fairly reasonable" things, that even if I didn't fully agree with, I also didn't think was completely wrong...until those last few sentences when he went "All In" with the: "Anyway, I'm just a scumbag and proud of it!".

Luckily, it's been a while since I last used "Lutris" and never on my latest computer. It looks like that won't change anytime soon.

View this comment - View article - View full comments - Report this comment

Uninstalled.
The whole thing was going down hill anyway after they reworked the wine version management system a la auto downloading versions without my consent. F that sh1t.

View this comment - View article - View full comments - Report this comment

​This d7vk is very impressive with d3d3 titles and others in my case:

MotoRacer 2 DX3



StarWars Shadows of Empire DX5



Radeon ARK Demo DX7



​😀

View this comment - View article - View full comments - Report this comment

Play-by-appointment: mechanisms that reward returning to the game (e.g. daily quests) will get a PEGI 7.

I consider this too low. I know how hard my child (8 yo) would be crying if I wouldn't allow to fetch the daily reward.

View this comment - View article - View full comments - Report this comment

Quoting: pbHerein lies the problem with the society. "I'm just one person using the AI, even if I stop, it won't solve the bigger problem". Now substitute "using the AI" with "littering", "eating meat", "speeding", "parking in illegal spots", "not voting", "preordering games", "buying from companies known to exploit their employees" etc. etc. you get the point. We will be our own undoing.

Quoting: SzkodnixI partially agree with Lutris dev that those are indeed big companies just being a**holes and developing AI tools in an unethical way, and that hatred towards AI indeed starts getting out of hand with people complaining about AI in general at this point (including less popular models not developed by huge companies and some LLMs run locally). Guys, hate companies doing evil and stupid, not the tools that are neither good nor bad, as those are JUST TOOLS that you either use or not use. If those tools are useful or not in this form, that's a separate topic :D

But there's a different problem that he did not mention: licensing. And for now determining under what license is certain code, especially the one generated by LLMs, is a mess. Trust is one thing (at least any notice would be appreciated), but the second thing is for this developer or other users of that code not to get into licensing mess.

Idk man I’m not a big fan of weapons, do they have a right to exist? Questionable at best for 99% of them (like i can see why a huntsrifle exist, sure, but not a tomahawk missle). I agree it’s how we use the tools but if the tool is 1000% garbage in every way why use the tool if all it does is make you dumber?

LLMs have some fields were they are probably a good addition and help (image recognition and OCR come to mind) but guessing code not so much.

View this comment - View article - View full comments - Report this comment

If these mechanisms punish players for not returning (e.g. by losing content or reducing progress) they will become PEGI 12.

Could pokémon become a pegi 12 game because of all those one time only missable distributions ?

That would be neat tbh. I'm getting sick and tired of not being able to simply restart a decade later without having to hack those things in. And i'm not sure it's possible for the switch games. I dropped them at the time, if they ever make a good one down the line and i want to go back to them... I probably won't, but it do feel like fomo at its best to retain the player base tbh.
And with the speculation over TCG that adds to it, it ain't good.

View this comment - View article - View full comments - Report this comment

Whether or not I use Claude is not going to change society

If you then push that code in a GitHub repository, it definitely will, as AI is trained on GitHub repositories.

View this comment - View article - View full comments - Report this comment

Oh, this is finally blowing up now? I actually skipped the entire section where the Mathieu glazed AI in The Linux Destination podcast. It was pretty long.

Look, I get it, people use AI. I personally don't mind if he generates codes, checks it, rewrite it into something that fits with the rest of the code in the project. Others' lines might differ, but as long as the program didn't become slop, I am willing to tolerate it.

But this:

Anyway, I was suspecting that this "issue" might come up so I've removed the Claude co-authorship from the commits a few days ago. So good luck figuring out what's generated and what is not.

Is absolutely the wrong way to handle it.

Although I do find it rather ironic that Lutris, in the same podcast, is claimed to be a platform for game preservation, and yet he uses AI, which Myrient (a game preservation site) has cited to be a reason why they're shutting down.

View this comment - View article - View full comments - Report this comment

Not just that but copyright becomes an issue. Who actually owns the generated code? And now it's being hidden, how can anyone tell?

For reference: https://zomglol.wtf/@jamie/116059523957674208

View this comment - View article - View full comments - Report this comment

I partially agree with Lutris dev that those are indeed big companies just being a**holes and developing AI tools in an unethical way, and that hatred towards AI indeed starts getting out of hand with people complaining about AI in general at this point (including less popular models not developed by huge companies and some LLMs run locally). Guys, hate companies doing evil and stupid, not the tools that are neither good nor bad, as those are JUST TOOLS that you either use or not use. If those tools are useful or not in this form, that's a separate topic :D

But there's a different problem that he did not mention: licensing. And for now determining under what license is certain code, especially the one generated by LLMs, is a mess. Trust is one thing (at least any notice would be appreciated), but the second thing is for this developer or other users of that code not to get into licensing mess.

View this comment - View article - View full comments - Report this comment

Liam is completely right.

But I also think, that the Lutris developer, strycore, has valid points I can absolutely relate.

Classic dilemma. We will see...

But the most important statement is the one from GlorousEggroll there:

there is no real problematic issue reported here other than opinion.

If you dont have a real problem, dont open an ISSUE on an ISSUE tracker.

View this comment - View article - View full comments - Report this comment

Herein lies the problem with the society. "I'm just one person using the AI, even if I stop, it won't solve the bigger problem". Now substitute "using the AI" with "littering", "eating meat", "speeding", "parking in illegal spots", "not voting", "preordering games", "buying from companies known to exploit their employees" etc. etc. you get the point. We will be our own undoing.

View this comment - View article - View full comments - Report this comment

"I'm doing a thing people hate, so instead of not doing that, I'll continue to do it, but hide it better"

That's a bold position to take in any project, let alone a FOSS project. And just because he can use it in non-slop manner (maybe? hopefully?) it doesn't absolve him of the ethical concerns many of us have about genAI.

View this comment - View article - View full comments - Report this comment

Quoting: elmapuli saw this movie before, we dont have an great track record of backward compatibility and we cant know if its fixed now, we have to wait 10 years from now and if the things made now still work, that means we have backward compatibility... for things made years ago, but current stuff might still have issues, as new libraries are used that may not have the same track record.

in any case if wine can be stable then there is some stable basis somewhere, until it doesnt because any distro can break stuff on purpose, but the ones who do would be punished by the lack of users... i mean everyone loves snaps right?

in any case, even if an game work, the mods for it may not work, if an mod rely on changing something on a binary to work, and the linux version has an different binary because its native, we cant expet this mod to work unless we run the windows version of the game on wine insted of native, so yeah, wine/proton will still be used even in a future where all the games support linux natively.

The backward compatibility issue for native Linux software has largely been solved, but not just by the modern kernel architecture. While it's true that the kernel handling DRM/KMS and memory management (instead of X11) created a vastly more stable and secure hardware baseline, older Linux software usually broke due to user-space library rot (e.g., missing outdated versions of glibc or libssl).

The real solution we have today is containerization and modern dependency handling. Technologies like Flatpak and the Steam Linux Runtime solve this by running applications in isolated environments that provide the exact legacy libraries they expect, preventing host system updates from breaking them. Furthermore, the ecosystem now heavily utilizes translation layers, shim libraries, and wrappers to seamlessly route deprecated API calls through modern, performant backends. For example, Xwayland translates legacy X11 calls for Wayland compositors, PipeWire provides drop-in shims for legacy PulseAudio/JACK apps, and tools like Zink or DXVK translate older OpenGL/Direct3D instructions into Vulkan. Add in user-space standardizations like SDL3, Wayland, and PipeWire, and the software stack is more robust than ever.

You are entirely right about mods, though. Because memory hooks and DLL injectors are built to target Windows binaries, highly modded games often run better using the Windows version through Proton rather than the native Linux port.

Quoting: elmapulanyway, do we still need an audio interface for professional usage? or pipewire does the job?

Regarding your audio question: PipeWire absolutely handles professional, low-latency routing (it natively implements the JACK API). However, it is a software routing daemon. You still need a physical audio interface for hardware inputs like XLR microphones or instruments.

View this comment - View article - View full comments - Report this comment

Stop, stop, we went sufficiently back 😄

View this comment - View article - View full comments - Report this comment

I think the lootboxes wouldn't be issue at all if you always get things in value of the price of opening lootbox. This is how plenty of mystery boxes are working (with exception of trading cards, as the amount of some cards is scarce by design). If you just have same amount of all the items of same rarity tier and have set value for them it would not be gambling as you will always get the value you paid for.

View this comment - View article - View full comments - Report this comment

Quoting: Stella

But I think we can do better with a native solution.

I hard disagree with this statement. Pretty much all existing Linux games suffer from various problems ranging from complete inplayability over control and gameplay/visual issues as well as crashes. Running games over Proton is vastly preferable to Native most of the time because the Windows API is much more stable than the Linux API. I myself have had so many issues with native ports that the Proton versions never have, that I've given up on them completely

In the end it's all a matter of support.
The fact that Proton works better is not just because of Windows API being stable, it's because Valve and the community are doing the heavy work of maintaining Proton and related programs so it keeps working well across distros and evolving versions of the ecosystem.
If Unity and game developers are being serious in supporting native Linux, they'll need to commit the same way as Valve does with Proton.

View this comment - View article - View full comments - Report this comment

Quoting: Linux_RocksStill not about to weep for Valve if they lose. lol

This is such a strange attitude to bring to a Linux forum. I don't get it.

I suppose if you only do native gaming, perhaps only buying on GOG and Itch, and so you don't benefit at all from Proton, and don't care about, or for, Valve as a company, that's a valid stance. Even then, you're overlooking the effects of groundswell they've created for Linux by shipping the Steam Deck and promising a Steam Machine, or pushing SteamOS. That groundswell is a big part of why indies still push out Linux builds, from which you benefit.

But I guess if you can overlook all that, AND you don't use Steam ever, then you won't care if Valve are forced to put ID verification in Steam?

But that's a lot of overlooking. And hardly a lol moment.

View this comment - View article - View full comments - Report this comment

Quoting: Stella

But I think we can do better with a native solution.

Pretty much all existing Linux games suffer from various problems ranging from complete inplayability over control and gameplay/visual issues as well as crashes.

I think we play very different games.

View this comment - View article - View full comments - Report this comment

I think the whole endgame for NYAG is not gambling, but using that a vector to enforce these draconic age checks and user monitoring onto Valve's customers through that route globally.

View this comment - View article - View full comments - Report this comment

Quoting: PyronickProton/Wine is an incredible tool, but a properly packaged native game running within this modern architecture is now just as stable as a Windows binary, with zero translation overhead.

i saw this movie before, we dont have an great track record of backward compatibility and we cant know if its fixed now, we have to wait 10 years from now and if the things made now still work, that means we have backward compatibility... for things made years ago, but current stuff might still have issues, as new libraries are used that may not have the same track record.

in any case if wine can be stable then there is some stable basis somewhere, until it doesnt because any distro can break stuff on purpose, but the ones who do would be punished by the lack of users... i mean everyone loves snaps right?

in any case, even if an game work, the mods for it may not work, if an mod rely on changing something on a binary to work, and the linux version has an different binary because its native, we cant expet this mod to work unless we run the windows version of the game on wine insted of native, so yeah, wine/proton will still be used even in a future where all the games support linux natively.

anyway, do we still need an audio interface for professional usage? or pipewire does the job?

View this comment - View article - View full comments - Report this comment

Quoting: GustyGhostI've been playing Linux native games (Unity, no less) that I had bought ten years ago without any issue. What is there left to improve?

i think the devs had to do somethings manually because the engine didnt fully support linux, or if it did the add'ons didnt, so there is that, maybe they can do something about that?

View this comment - View article - View full comments - Report this comment

Quoting: Stella

But I think we can do better with a native solution.

I hard disagree with this statement. Pretty much all existing Linux games suffer from various problems ranging from complete inplayability over control and gameplay/visual issues as well as crashes. Running games over Proton is vastly preferable to Native most of the time because the Windows API is much more stable than the Linux API. I myself have had so many issues with native ports that the Proton versions never have, that I've given up on them completely

You are completely right about the historical mess of native ports, but the "unstable Linux API" (or rather, [ABI](https://en.wikipedia.org/wiki/Application_binary_interface)) argument is based on an outdated architecture. Twenty-five to ten years ago, the ecosystem tried and failed to unify around the "[Linux Standard Base](https://en.wikipedia.org/wiki/Linux_Standard_Base)," which never actually provided a stable target for developers as it shifted goalposts continually. The Linux kernel itself has always been fiercely stable (adhering to the strict "we never break userspace" rule from Linus Torvalds himself), but what broke games in the past was the chaos of fragmented userspace dependencies, such as the glibc mess with conflicting "GNU-isms" across versions, and competing multimedia servers across different distributions.

Today, modern native Linux gaming bypasses that problem entirely through a completely new approach. First, the Steam Linux Runtime ensures games no longer rely on your host operating system's libraries. Instead, they run inside a frozen, containerized environment, completely bypassing the software rot that used to plague Linux ports. Second, hardware translation has finally been standardized. The universal adoption of PipeWire has permanently fixed the decade-long multimedia nightmare, elegantly replacing the deeply fragmented ALSA (ugh), PulseAudio, and JACK stacks with a single, unified audio and video pipeline. Combined with the newly stable SDL3, even older native games are retroactively forced to play nicely with modern Wayland desktop environments without the original developer having to update a single line of code. Proton/Wine is an incredible tool, but a properly packaged native game running within this modern architecture is now just as stable as a Windows binary, with zero translation overhead.

View this comment - View article - View full comments - Report this comment