Hacker News Reader

Comment by tester457 | Hacker News Reader

The author considers the bug reporting to be a benefit, it's a contrast from this post last month "One game, by one man, on six platforms: The good, the bad and the ugly" [0].

> The Ugly

> Linux accounts for less than 1% of the total players, so from a business perspective, it is not worth it. And the platform has the highest percentage of players who run into platform-specific issues - the time used for customer support only makes the economy worse. Also, I don’t own a Steam Deck, so adding support is not always easy: Valve does not provide a SteamOS image to test. So Steam Deck-specific code (like controller support, starting the game in full-screen, making the UI bigger, etc) is not tested end-to-end. I’ve used HoloISO with limited success.

> TL;DR: Even worse economy than Mac but less annoyance perhaps

[0]: https://news.ycombinator.com/item?id=37862606

View on HN

> And the platform has the highest percentage of players who run into platform-specific issues

Except that in the case of the OP's posted article, their Linux bugs were few:

"Do you know how many of these 400 bug reports were actually platform-specific? 3. Literally only 3 things were problems that came out just on Linux."

> The author considers the bug reporting to be a benefit, it's a contrast from this post last month

>> Linux accounts for less than 1% of the total players, so from a business perspective, it is not worth it.

These actually don't need to contradict. The current post specifically suggests that at a cursory glance it looks like a bad decision to support linux users and instead triage those bugs but that in reality the bugs they report are more universal. We can call this a colloquial paradox because the simple reasonable assumption is in direct opposition to the nuanced one.

Let's call it Koderski's paradox since that's the Reddit user's name.

> Let's call it Koderski's paradox since that's the Reddit user's name.

I'm in favor of this name.

Now we just gotta make a wikipedia article lol

It's almost self defeating. Seems like developing games for linux is harder.

less people play games on it, so it get's less dev attention.. so quality suffers.. so less people play games on it.

I'm not blaming game devs, linux devs or users for this nor am i blaming Microsoft etc.

It just is. Efforts like Wine and Steamdeck go above and beyond to maybe changing this.

> Seems like developing games for linux is harder.

This is not because of some property of Linux, but mostly because of game developers' unfamiliarity with it. Sure there's some issues with window servers but if you're building a game using any mainstream engine or even simpler frameworks like SDL that shouldn't be much of an actual issue.

> It's almost self defeating. Seems like developing games for linux is harder.

Except that the parent's point is refuted by the OP's article (that brought us all here to discuss):

"Do you know how many of these 400 bug reports were actually platform-specific? 3. Literally only 3 things were problems that came out just on Linux."

From talking with some indie devs, the worst platform is Mac, the rest depends on one's preferences and experience - some people feel more comfortable on linux.

Developing ANY user-facing application for Linux is harder. Linus himself explained why.

https://www.youtube.com/watch?v=Pzl1B7nB9Kc

TLDW: Linus speaking in 2014, lamenting that there isn't binary compatibility between distros, or even between major releases of a given distro.

This was before the Snap project though. (Somewhat related: I hadn't realised until today that Snap isn't fully Free and Open Source.)

https://en.wikipedia.org/wiki/Snap_(software)

Snap is yet another Canonical NIH copy of others' solutions. There's already Flatpak as well as AppImage or just-ship-your-damned-dependencies before that. Lack of binary compatibily of random libs shipped by the distros is not a real problem in practice - base libraries (libc, opengl/vulkan, libasound/pulse) etc all have backwards compatible interfaces and the rest you can and should just include with your binaries. This has been known as far back as Loki back at the start of the millenium but somehow there are still people spreading FUD.

I believe Valve is working on a SteamOS 3 image for general purpose PCs and confirmed it publicly, but we don’t know the release date.

This is great news, have to say i love they keep taking whacks at it.

Just remember its not out of altruism. It's very beneficial to Valve to be able to eliminate their Windows dependency. I will say, it is good when incentives align for the benefit of the consumer.

I don't think many people expect companies to do things out of altruism, but, as you say, we are happy when our interests and theirs align.

That’s the point of markets.

Three ways to get someone to do something:

* make them love you

* pay them

* make them fear you

Markets incentivize behavior that would otherwise require altruism or fear.

> Just remember its not out of altruism.

I’ve always been curious what point is being made when people say this. Care to expand on it?

At least in my case, the point being made when I say "Just remember it's not out of altruism" is to remember that it's not out of altruism.

Why is it important to be reminded of that?

If it's based on Arch (insane rolling release crap idea) then I won't be using it.

I've been using Gentoo Linux (which is also a rolling-release distro) since the early 2000s (maybe 2002?).

Even taking into account the libpng upgrade disaster in the early-to-mid 2000s, it's the most sane, most stable, least-trouble-free distro I've used.

I've wanted to find some other good distro, but all the ones I've tried either

a) Only ship ancient packages in their official repositories, requiring me to go to unofficial sources to get oh so many up-to-date versions of things.

b) Inevitably do something pants-on-head retarded as part of normal operation (nearly always during upgrades) and require me to bust out my Linux sysadmin skills to unwedge the system

or (sometimes) both "a)" and "b)".

You don't like rolling-release distros? That's fine, more power to you. They've been working fine for me for (oh dear god, I'm so old) ~20 years.

Gentoo stable (aka not ~arch) might be rolling release but can also be surprisingly conservative, often getting new versions even after Debian stable.

Gentoo testing (aka ~arch) does come with occasional surprises (although most common is some package update failing to build) but still worth it compared to having to deal with big updates or even reinstalls.

Yeah, agreed on both counts.

I run testing on my desktop, and stable-with-some-exceptions on my servers and router.

For the past couple of decades, I've been periodically looking for other equally-no-surprises distros so that I can stop burning so much power on building from source, and, man, I don't think I'm ever going to find one.

Of course it will be, that's what the Decks run now. They don't update in the same way or pace as regular Arch, though. pacman is basically disabled. You're encouraged to do everything through flatpak and appimage, and then the occasional blessed system update comes through for the other parts.

Using a non rolling release distro as a desktop makes 0 sense. You're holding off on upgrades, forcing 1 gigantic upgrade at once that will never be tested. Do you have a dev or staging environment for your desktop upgrades? Why not do small incremental updates all the time?

Sounds like the difference is whether it's about platform-specific issues.

> Do you know how many of these 400 bug reports were actually platform-specific? 3. Literally only 3 things were problems that came out just on Linux. The rest of them were affecting everyone

> And the platform has the highest percentage of players who run into platform-specific issues - the time used for customer support only makes the economy worse.

Also:

> Fixed that and added a debug log there to just make sure it worked well, and after reviewing unrelated reports from windows I found that it would hit these players too, they just didn't report that.

Wouldn't surprise me if a lot of the unreported bugs were Windows-specific and they're just not seeing reports of them, or not managing to find the cause because the reports are so nonspecific.

One is built with webgl and typescript, the other with Godot.

It shouldn't surprise anyone that a tech built for a specific purpose (video games) with Linux as the first class platform, performs better and is more stable than arguably the second most complicated piece of software engineering the humanity has ever devised - the browser.

The more complicated your stack is, the more problems you will have. The Linux specific bugs also tend to increase (and become nigh unsolvable) with Unity. Godot is much leaner, and tends to wield this fact as its strength. As an unrelated example, there are multiple community plugins for implementing different physics engines in Godot, when such a thing is an exercise in frustration in Unity or Unreal, for varying reasons.

I don't think that they should release for a platform they can't test.

So no one should release on linux? Because no one can test every linux distribution.

There's a difference between

- marking it as officially supported and asking the full price when it has never been run on a special hardware device running a custom OS (the steam deck that the person you're responding to mentioned)

and

- saying "we've tried on Ubuntu 24.04 default install with AMD drivers as well as Fedora 82, but ymmv obviously we know you love to tweak your setups just like us <3"

They probably never said they explicitely support steam deck. It's probably steam deck users that are interested in it, and valve doesn't provide a testing platform.

Reply on news.ycombinator.com