I’ll seriously consider EU V when it comes out, but it’s not even officially announced yet. Until then, I’m trying to whittle down my unplayed games. Hopefully I’ll get through a bunch before I find more interesting games.

Yeah, whatever works. I do the same, but I’ve seen some setups where PC works on both. It really depends on where things are.

Or a long cable. It just depends on how things are located. Steam Link could also work, depending on how good your wifi is and if your TV supports it.

Couldn’t you just use Steam Big Picture + Controller for TV play?

gift-wrapped install

The ones I’m aware of are:

• Bazzite - similar to SteamOS, but based on Fedora Atomic
• Garuda - based on Arch
• Nobara - based on Fedora

Bazzite is going to work differently from pretty much everything else, so I can only really recommend it if it’s being treated like a console. Otherwise there’s just way too much learning to figure out how to properly make changes.

Garuda being Arch based is a liability. I don’t recommend any Arch spins until you’re comfortable with both Arch and Linux, which certainly doesn’t describe your average new Linux user. They’re going to mess around with the AUR and they’re going to break their system.

Nobara is probably the best option here, but it’s maintained by one person who promises to keep working on it. It’s also heavily modified, so getting help from the Fedora community may be iffy.

For Debian, you can get 90% of the benefit of the above by installing whatever release is in testing (trixie in this case, I don’t recommend testing directly). Switch only if you run into issues nobody else seems to have.

Use what you want of course.

Never go 100%, there’s a 1% chance they’ve actually put some effort into it.

Bullseye

Bookworm?

EEVDF allocator, sched_ext

Schedulers really don’t matter much for gaming workloads.

Ntsync is a bit more exciting, I’ll give you that. Looks like it’s in testing, so you can update to trixie to get it. Running on the testing branch is usually fine, and it’ll be pretty up to date until release freeze time, at which point it’s best to sit tight until the new testing branch settles down (a point release or two is usually plenty).

manual tweaking

Tweaking is usually not worth it. The most I really do is up sending like the mmap limit if a game is struggling.

comparatively niche game-related problems.

I’m not really sure what types of issues you’re referring to. Either you’ll have OS questions, in which case the generic help is ideal, or you’ll have game specific issues, in which case most distros will be extremely similar (e.g. proton db for game specific workarounds).

For people coming to Linux, I’ll usually recommend sticking with Steam verified or playable games as well, since those should largely just work. Yeah, you might be able to eek out a few FPS or, more likely, framerate stability with some tweaks, but that’s honestly not worth it to most.

If you want to tweak, you’re probably also the type to want the “advanced” distros like Arch and will ignore my advice anyway. And that’s fine, I use something else as well, but I’m going to stick to a more predictable experience for new users.

And the crazy thing is, I agree with your description of games I like, yet my list is almost exactly opposite yours. I just finished Zelda: Echoes of Wisdom last night, I loved GTA and RDR, and I didn’t particularly like Skyrim and only played about 15 min of Dragon Age.

My favorite games are really hard to describe without lumping in a bunch of games I don’t like.

That’s one of the few I missed recently, which is too bad.

Wow, that’s impressive. Or maybe I have a problem of buying way too many games.

Did you actually install EGS, or are you using Heroic?

Lots of good points here.

I’m lucky to work in an org with a product team that does a fantastic job of detailing the problems they want to solve and an idea of the workflow a user should take. They discuss that w/ our design team and architect, who basically come to a conclusion about how the problems get solved in a way that works well for both users and developers.

If you ask our customers, they’ll say they want A and B (in many cases charts and graphs), when neither would solve the actual problems they have (e.g. optimizing cost), and what they need instead is to provide a little more data (i.e. their upstream and downstream costs) so we can produce better recommendations (i.e. reduce X and increase Y to get similar results for less cost). We’re really good at simulating things, our customers are really bad at it, yet customers ask for features that let them try to simulate things instead of us providing that value.

That goes double for games, since people are a lot more emotionally invested in games that workplace software. If the player is complaining about “bad controls,” the solution may be some kind of indicator of off-screen enemies (i.e. improve time to react), which has nothing to do with the controls themselves.

Day-to-day software does work on ARM, games don’t. I can run macOS or Linux on ARM, but that’s not going to encourage game developers to port their software to that architecture.

Apple doesn’t particularly care about gaming on macOS, if they did, they’d be working directly with game developers to get games working there. They have made some attempts to improve APIs and whatnot games would need, but that’s also relevant for other software Apple seems to care about more (CAD, Adobe products, etc). Gaming just isn’t something they really care about.

Proton is the only reason the steam deck is as good as it is

Absolutely. I actually didn’t make a Steam account until they ported their client to Linux, and back then they didn’t have any form of compatibility for Windows games. It wasn’t until 2018 (~5 years after initial Linux launch) that they released Proton. That’s a big part of why Steam Machines failed, they were released 3 years before Proton was a thing so game selection was extremely limited. Valve hoped devs would create native ports for Linux, but that never happened, and it’s why Proton is a thing at all, and why the Steam Deck launched years after the initial Proton launch (they wanted a large library of games).

Android is quite different though, since the user has very limited access to the actual hardware and largely interacts with the system through approved apps. Whether it uses Linux or some other kernel is largely irrelevant for the functioning of the device.

When I say “Linux,” I mean the rest of the userland as well, whether that’s GNU or something like busybox. There are certain expectations about user interaction that SteamOS satisfies that Android really doesn’t.

I’m 100% fine with Steam as-is, the DRM is optional and you can install anything else you like. If that’s what takes over, I’ll be ecstatic.

I’m often wrong about what I don’t like much of the time. For years, I thought I hated rogue-likes, and then I found some I really liked. Likewise for other genres.

I know what games I don’t like, and what games I’ll probably like and not like, but only after I see some gameplay, and even then I’m not all that accurate. And sometimes my opinion of a game will change after playing it.

Precisely. Valve builds stuff largely to drive some kind of idea forward so hopefully other devs will copy it and people will buy more games. They also want them to make money, but they make so much more through Steam than any of their games that it’s not really worth the investment. Counterstrike (their biggest cash cow) makes ~$1B, Steam makes ~$10B (estimates of course). Most of their games are way below either numbers in total revenue ever.

it never quite reached the console players’ mindset

Exactly. Console players want to plug it into a TV and start playing. PC gamers are happy to tinker a bit. Steam Machines might fill that gap if Valve ever successfully launches one, but consoles still provide a good experience for a lot of people so it’ll be hard to shake loyalists.

So yeah, your assessment is spot on.

Yeah, not a fan of macOS, but I really wish Apple cared more because it could totally drive forward ARM compatibility in games.

Agreed.

Hopefully other platforms start caring about Linux because I really don’t want Gabe to be replaced by a “profit all the things” CEO and for Linux gaming to become a locked-down platform.

And probably break some face bones.

I don’t think that was the real issue though. People didn’t want it because a ton of games didn’t work, manufacturers weren’t excited because there wasn’t an existing market, and Valve wasn’t really invested. They basically tried to pawn most of the risk off onto hardware manufacturers.

The Steam Deck took the opposite direction, they invested themselves in pushing hardware, which meant they had more incentive to get games to be compatible, and the result is creating a market that other manufacturers could actually quantify. They took pretty much all of the risk themselves, and later manufacturers decided to jump on board.

? How so?

Valve didn’t prove Steam Machines, didn’t really market them, and game support wasn’t there, they just launched it with a promise to fix stuff.

They did the opposite with the Steam Deck, they proved the concept with their own hardware, they marketed them heavily, and they had a ton of games ready to go at launch.

They learned from the mistakes made with the Steam Machine.

Sifu? If so, yes, it was released more than 12 months ago.

Just finished it in one sitting, what a ride.

Most Linux users, including gamers, don’t really benefit from improvements to Linux since most of it is drivers for hardware they don’t have. Most userland software can be installed via flatpak or PPA (or other form of additional repository for your distro) if you really need something newer. But my understanding is that people (esp gamers) get annoyed more by stuff changing than missing out on new stuff.

The whole point of recommending a stable distro is to give the best chance of the person finding the help they need, as well as things not breaking randomly, and you get that with stable release distros. If the user knows enough to disregard that, they know what distro would be a better fit anyway.

In that specific case, yeah, maybe try a different distro. SteamOS will still be a worse option since Valve doesn’t have any published update cadence.

But still stick to a major distro, like Fedora or Linux Mint. It’s unlikely you’ll actually run into issues on Debian though…

Yes, all of that is for old systems, like 6+ years old, as was stated earlier. Intel has been clear about needing a recent-ish CPU (Intel 10th gen or AMD 3000 series) with resizeable bar support enabled. So if you’re seeing terrible performance, options are:

• upgrade CPU - a 5600X runs ~$120 and works with many older boards
• hack around missing option for resizable bar support
• wait for Intel to address the problem
• buy a different GPU - RX 6700 XT is a good deal used

If you’re building a new system or upgrading from an APU, the B580 is a phenomenal deal. If your system is 6+ years old, you’ll probably want to upgrade your CPU anyway.

Absolutely! I remember having ATI embedded graphics on my motherboard and it was more annoying than Nvidia’s drivers. Nvidia didn’t really change since then, it’s just that AMD submitted their driver to the kernel, so newer software tends to work better with it.

dGPUs on laptops have always sucked on Linux, this isn’t new, nor is it necessarily a problem specific to Nvidia. Graphics switching on Linux just isn’t smooth, which is why I haven’t bought a laptop with a dGPU since switching to Linux. I hear it works, I just don’t see the point. Get a cheap laptop with an AMD APU and you can play casual games on it, and then build a cheap desktop PC with the savings and your experience will be much better. It’ll probably cost a bit more at the start, but your laptop will last much longer and you can upgrade the desktop more cheaply.

Debian is totally fine, why do you need a rapid update cycle? Everything you need is packaged with Steam. If for some reason you need something newer, you can always use whatever release is in testing at the time (use that release name, not “testing” itself) and you’ll get newer packages with minimal risk of stability issues (a lot of people run testing).

There’s really nothing special about newer packages for gaming. Once it’s working, Debian will keep it that way.

I personally use openSUSE Tumbleweed because I like newer packages for other reasons (I use it for software development) and hate release upgrades because they take forever, but tons of people use stable distros without issue.

If you want some bells and whistles out of the box, I hear Bazzite is good. But any distro will work fine with Steam, and I’d assume Heroic and other launchers should also work fine on any distro they’re packaged for.

There is and will be no Z2 Steam Deck."

That’s too bad, but perhaps it’s not a game changing amount of uplift yet.

That said, they didn’t say anything about other products, so I wonder if we’ll see it in another form factor, like a standalone VR headset.

I highly recommend not using SteamOS on your PC, unless it’s literally used as a console. Use any major distro instead.

Nvidia is fine and has always been fine, it just hasn’t been ideal. I’ve used Nvidia GPUs on Linux for >10 years, and it has worked well.

The main issues are (and have always been):

• kernel/driver mismatch - pretty much only an issue on rolling releases
• late support for new Linux graphics stack features - Wayland recently, but that’s a bit better now

If you use a release based distro and don’t need to be on the bleeding edge (describes pretty much everyone), Nvidia is fine.

I switched to AMD a couple years ago because they offered better value and I needed an upgrade anyway, Linux compatibility was a nice value add.

Decent, but I think it’s a little odd:

• collectibles require backtracking in levels, yet they also record time, so I’m not sure if they’re recommending speed or exploration; levels are pretty linear
• you can buy lives/extra hearts in the store (why?)
• some things can be fiddly

I think Yooka-Laylee is way better, but it’s still fun.

Ok, I just reviewed them quickly, and I see this:

• Nvidia announced 5000 series - no third party reviews, and the official numbers are sus due to different settings; looks like they’re relying on AI for perf improvements, so we’ll see what that ends up looking like
• AMD announced 9000 series CPUs and GPUs (GPUs through the press only) - CPUs are higher end (not applicable here) and GPUs have no details

So all I see are dubious claims by Nvidia about products nowhere near Intel’s lineup with the cheapest one going for $550 (>2x higher than Intel’s top end) and the most expensive going for $2000. If those are interesting to you, you aren’t the target market for Intel’s GPUs, and if Nvidia’s are too expensive and you’re unsure about Intel’s GPU, you’ll wait to see what options AMD launches with, and there’s a lot of room between Intel and Nvidia.

So I’m not exactly sure what the new releases change that you’re claiming.

Reboot. I never played the original.

their worlds are getting so big that Rockstar just can’t make enough content to fill them the way they used to

I really hope GTA VI is smaller than GTA V. I really don’t see the point of a bigger world if it doesn’t add to the fun. But maybe it makes sense for the online version, since players have more places to go, but surely they could have added online-exclusive areas instead (maybe they did? I never played online).

A bigger world isn’t a selling point for me, it’s more of a liability.

I’ve been playing a bunch of shorter games, such as:

• Bomb Rush Cyberfunk
• Firewatch
• City of Beats
• Kao the Kangaroo

And I’m still playing a few games while my kids watch.

I felt GTA IV’s “emptiness” was totally fitting for the story though. Niko was all alone in a massive city, so it makes sense that he’d only know a handful of places to go. It felt like a big city you could easily get lost in, with a few major roads that helped you figure out where you were.

GTA V, however, had no such excuse. The MCs were all natives to the area, all involved in the criminal underworld in different ways, and yet there’s still nothing to do. I expected Franklin to do something gang or car theft related other than a handful of missions, Michael to have more heists and equipment hookups, and Trevor to have some fun drug-related minigames. But no, the side content for each was underwhelming.

Honestly, I found most of Los Santos in GTA V pretty boring. Everything interesting is super spread out, and I don’t know what these “things to do” people keep talking about are, and the minigames that exist are honestly pretty boring. GTA SA Los Santos is fantastic, with fun things around every corner. GTA IV Liberty City is the same. But GTA V just doesn’t have much to do for how big it is. I much preferred messing around in SA or IV than V.

I honestly only finished it because I was tired of coming back to it expecting a different experience due to the hype it got. Maybe the issue is that I never played online (never cared), or maybe I’m just not the target for it.

That said, I completely respect this person for going so in-depth. I don’t care enough to install the game, but I might boot it up for 15 min to find a couple of these spots if I had it installed already.

Exactly. The type of person buying a $250 GPU is likely to be running a recent budget CPU (i.e. first build, pre-builts, or entry into new CPU socket), not an older, higher-end CPU that’s still relevant.