Quoting: CatKillerSo you need a Linux test machine, and you need to fix bugs in your game for running on Linux exactly as you would if you had a native build,

The process of bugfixing a native Linux build vs one running through a compatibility layer and/or VM is not the same, no.
And neither are the bugs you will encounter. Things that break between Windows/Linux are different than things that break between Windows/Linux-Pretending-To-Be-Windows.
The first actually has way more possible breaking points, though a lot does depend on your engine used.

When you debug something through Proton, you need a Linux machine capable of starting your game (even a VM will typically do, if the bug you're hunting is not GPU-related). That's it.
All your development will still be done on Windows, you don't need any familiarity with the platform other than "copy files to target machine".
I'd barely even call it debugging, honestly. It's make change, rebuild, launch, test if it works now.
It is rather inefficient, but also very simple and any dev can pull it off, even if they've never used Linux.

When you debug a native build (no matter the platform, actually), you first of all need a complete development environment on the target platform for all the usual development tools like running a debugger (best a debugging build if that's something your stack supports).
All of this is much more involved and requires familiarity with the platform at hand.
What it gives you is a way better toolset to find and fix bugs.

So, native development is actually way more powerful and efficient. I'd always do it if given the possibility.
The advantage of fixing bugs specifically for Proton is really only that you need no "intimidating" familiarity with Linux.
But that advantage is incredibly huge to those who (unfortunately) think Linux would be difficult to use for them due to lack first hand experience.
I don't know many devs that wouldn't prefer working on Linux after they have gotten used to it (which doesn't even take that much time). But I also know a lot of developers that will just never even try because they are too afraid of having to work with an unfamiliar environment.

I feel like all of this has a lot more to do with psychology than it does with actual attributes of the different platforms. But in the end, it is what it is, and the result is that working with Proton is a lot more friendly to Windows-devs than doing and maintaining a native Linux build.

View this comment - View article - View full comments

Quoting: CatKillerNonsense.

If it works, great; you're providing no more support than ProtonDB does. If it doesn't work then you're either saying "works on Windows, screw those guys", so no support, or you need to figure out why it doesn't work. And developers don't want to test in the first place.

Speaking as someone with 15+ years of software development (including a few years with games), you don't sound like you've ever developed anything in your life other than a fever and some kind of attitude issue against developers.

When you officially support Proton, then yes, that means you'll have to figure out why something doesn't work in Proton and can't just say "screw those guys". Duh?
The vast majority of those cases are from those I already listed (with file-related issues and codecs probably taking the biggest chunk), with solutions in most cases coming down to very small changes (such as not relying on some Windows-path-shenanigans or changing the encoding of a video to play on Proton) that any dev can do without having to dive deeper into Linux.
It is exactly as I said: It still does not require you to be familiar with Linux or develop that familiarity in order to fix an issue.
And as time goes on and both Wine and Proton keep getting better, those cases are becoming less and less.

I've seen more than a few devs that don't even do official Proton support still fix something in their game in order to make it work on Proton after someone reported an issue - without ever actually touching Linux.

View this comment - View article - View full comments

Quoting: CatKillerBut if a developer is "supporting Proton" rather than "supporting Windows and hoping for extra money from Linux users" then they need a Linux testing pipeline anyway for their Windows build in Proton. Proton saves them hardly anything, and is just a mechanism to be able to say "Wine won’t ever be something that we guarantee will work" when they aren't doing that testing.

Not true.
For Proton, they just need something that can run their game via Proton. That is decently easy to find a guide for, set up and - more importantly - does not require any dev to have knowledge about the platform.
It's more or less a "run and test if it works" kind of deal.

Which is in no way comparable to actually having to work on issues in code specific to the platform, which can be a lot of work even if you are familiar with it.
While (good) cross-platform engines can save you a lot of that work, there still remains some, especially with some topics like files, security, codecs, etc.
Sure, some games - especially smaller indie games - won't be affected by these topics, but the bigger the game, the more likely that it would be affected.

Doing Proton only saves developers a lot of time - otherwise, we wouldn't see a surprising amount of devs doing tests eg on the Steam Deck themselves.

Quoting: mr-victory

Quoting: TheSHEEEPa native client FOR A MULTIPLAYER GAME WITH SELF-MADE CLIENT-SIDE CHEAT PREVENTION

AFAIK Mac client doesn't have this anti cheat.

That would be extremely weird, I have a hard time believing it.

View this comment - View article - View full comments

"For all the reasons described above, Wine won’t ever be something that we guarantee will work, but also for all the reasons described above, we’d really like to make it work."

This, to me, makes sense.

Quoting: AsciiWolfThis mindset is one of the reasons why I am still not sure whether I am happy about the popularity of Valve Proton or not...

The alternative would be practically no games working on Linux, at all.
Or only through the horrible hassle of manually fiddling around with Wine.

Linux is never going to amount to such a large player base that releasing a native client FOR A MULTIPLAYER GAME WITH SELF-MADE CLIENT-SIDE CHEAT PREVENTION would make business sense to developers.
For single-player games as well as devs using an out-of-the-box anticheat solution that does support Linux, the situation could be different depending on their circumstances (engine used, cross-platform experience, etc.).

Once Proton enters the picture, the question completely changes, as there is now a third (and fourth) option between native client and no Linux support at all.
Does Proton lead to less native versions? Of course. Well, at least I'm fairly certain it does.
Is that a bad thing? Not really.

You don't really have much of an advantage from a native Linux version compared to Proton. Very often, that version is developed once and then badly maintained, if at all. It might simply cease to function as fixed precompiled binaries tend to do on Linux as the kernel/distro itself advances. Which is one of the great strengths of Linux - it doesn't need to carry that extreme burden of having to support decades of legacy-code and programs.

The performance difference seems completely negligible to me - if it exists at all for a given care, I don't care if I could run a game at 200 fps or "just" 180 if all I need is 60-144 depending on the display.

The lack of official support, eg a game breaks Proton compatibility that it used to have but didn't officially support and then you can't play it anymore and devs don't care to fix it... I have heard of such cases, but in ~5 years of gaming exclusively on Linux by now, I don't recall this ever actually happening to me.
And if it did, I'd probably move on to one of the other hundreds of games in my library ;)

These things are all so minor compared to being able to play the vast majority of games on a much better OS.
So I'm definitely extremely happy Proton is as popular as it is and may it continue to be so for a very long time.

View this comment - View article - View full comments

Quoting: MadWolfor as EA likes to call them surprise mechanics these games need to be banned

Come on, man.
Where is your sense of pride and accomplishment?!

View this comment - View article - View full comments

"Linux kids saved from exploitative game & business practices"

Great news!

View this comment - View article - View full comments

Quoting: ElamanOpiskelijaThe investment is not worth it anymore. Even the impact of Age of Empires 4 has been marginal, in the context of X-Box Studios games. There won't be a big production coming from big studios anymore.

You have to look at smaller studios, like the ones doing Riftbreaker and Age of Darkness, which by the way are both primarily single-player.

For the "big name failures", you really have to look a bit deeper.
AoE4 failed for some extremely big reasons:
1) Competing with yourself - I think that one is self explanatory.
2) It launched in an absurdly shoddy state with features lacking on all fronts - massive holes in the set of features people come to expect from an RTS title nowadays, especially big name ones.
3) Waste of money during development. Look, I believe those documentary videos are nice... but who the hell plays RTS games for that? The money could've been invested in so many better suited things, it's just a mystery.
4) Crazily slow development & updates past launch.

All of these (yes, even the first) could've been avoided, and easily so, with just a sprinkle of common sense among the higher-ups.
My guess is too many yes-men among the hierarchy.

It is by no means a case of the investment not being worth it.
Of course it isn't if you mess up your own game - but that's true for (almost) all AAA launches.

Quoting: CatKillerNorthgard has this: there's the story mode, then the additional story mode that they added recently, and a whole chunk of per-clan challenges. Plus generic skirmish mode.

Yes!
And that game has been a very welcome change. I recommend it to anyone who has the same complaints as I have about the state of RTS gaming.
But it is also a complete rarity aaaand it's not quite a classic base builder RTS with all its special mechanics.

View this comment - View article - View full comments

I wonder when RTS developers will get around to targeting the (clearly larger) pool of single player RTS gamers again with a title that offers a bit more replayability than "play through campaign once".

It's been a while... and meanwhile, almost every RTS that went for the multiplayer crowd and did not have a big name like StarCraft or AoE behind it just straight failed - without competitors ever seemingly learning from each other's failures.

View this comment - View article - View full comments

View this comment - View article - View full comments

"Freedesktop.org SDK 22.08 (Flatpak runtime) 64 bit 7.61%"

What is that?

View this comment - View article - View full comments