[pduan]

I never understood why people on Linux always fight to have games on the platform. From the dev perspective, Linux is probably a sub 5% market share so it's never worth the effort compared to releasing new features/content.

Why fight this losing trend when you could easily dual boot or even have a dedicated gaming rig?

[CyberShadow]

> I never understood why people on Linux always fight to have games on the platform.

Because it is important to challenge monopolies least they abuse their position of power. With DirectX, XNA etc., Microsoft pretty much holds the monopoly as the PC gaming OS. Microsoft has made bold moves against individual freedom and privacy in Windows 10, and I'm sure more will follow.

And what better platform to compete against Windows' monopoly on PC gaming than the most popular free OS? I think that in the long run, continuing to create and display demand for PC games on Linux will only help gaming and even computing in general.

[NEDM64]

Sorry, but you're being naive.

Rarely there are standards bodies that aren't controlled by companies and their interests.

If they allowed OpenGL to flourish, it would ultimately be controlled by either Nvidia, ATI, PowerVR or someone else in the interest group to suit their hardware. Somebody has to pay the engineers.

Look at W3C, if is practically controlled by Google, and they only do what suits their needs, the way they want.

Look at Bluetooth, Apple is controlling the thing, and now they are releasing a newer version that's selfishly taylored for Apple to sell expensive headphones probably with patented methods, which they will use to protect themselves in court in case someone goes on a patent war against them.

[CyberShadow]

You are talking about something else.

> If they allowed OpenGL to flourish, it would ultimately be controlled by either Nvidia, ATI, PowerVR or someone else in the interest group to suit their hardware.

As opposed to DirectX, which is not implementable on other systems without large reverse-engineering efforts, because the technology depends on proprietary and encumbered middle-ware. Does a former specification for DirectX even exist (aside the documentation for Microsoft's implementation of the API)? Or it is one and the same, is it not? Where is the DirectX equivalent of MESA?

W3C, Bluetooth etc. have nothing to do with PC gaming. Why are you bringing this up? But since you mentioned W3C, what would you prefer - proprietary technology controlled by a single entity, such as Adobe Flash, or an open standard, such as HTML5?

An open platform allows free choice of technologies to use it with. A closed platform restricts those choices. A monopoly in control of a closed platform has even more freedom to restrict freedom of choice, and push forward with platform lock-in.

[fsloth]

"An open platform allows free choice of technologies to use it with."

Which are also made by companies just like Microsoft.

It's really hard to combine gaming enthusiasm (with the exception of retro gaming) with love of the 'free choice' and not be delusional.

The main reason: the GPU. There is very little value in having an open specification, if the implementation sucks. And the complexity of the graphics stack is such that nobody but huge monoliths that can focus lots of paid personnel to implement the GPU, implement the drivers, fix the drivers, fix the drivers, fix the drivers...

It's grueling, backbraking work to make a complete hardware based graphics stack.

Yes, there is MESA - so if software rendering suffices one can use more open options.

We can argue what do we mean by "gaming" of course. Solitaire, chess, tuxracer - yeah, I agree, free is very doable.

GTA v, Witcher 3 ( triple aaa titles) - you need a BigOrg to mind your graphics stack, of which there are only handfull on earth, of which Microsoft is one, and not worse as the others (if we limit discussion to games).

[shmerl]

> Yes, there is MESA - so if software rendering suffices one can use more open options.

Mesa has hardware accelerated implementation of OpenGL. You don't need "big org" to do it. But it surely helps if they put resources into that work.

> of which Microsoft is one, and not worse as the others (

It's worse than many others because their goal is not to improve the graphics stack, but to lock everyone into using MS systems.

[fsloth]

>> Yes, there is MESA - so if software rendering suffices one can use more open options. >Mesa has hardware accelerated implementation of OpenGL. You don't need "big org" to do it. But it surely helps if they put resources into that work.

The hardware specific driver is the more laborious thing to get correct. Well, OpenGL as well, but the driver is what I was after, not the specific API on top of it, and if the driver is buggy the OpenGL implementation on top can only try to patch things only so much.

>> of which Microsoft is one, and not worse as the others ( > It's worse than many others because their goal is not to improve the graphics stack, but to lock everyone into using MS systems.

I don't think there is any way MS can lock everyone into using their system.

I would claim MS has driven innovation in hardware accelerated realtime graphics by providing a stable and tested API at a time when OpenGL implementations were buggy and broken due to the messy work Khronos did with the standard and the way the implementations were done.

[CyberShadow]

Sorry, you are talking about something else too.

My comment was about monopolies and how they creep into other fields through lock-in and proprietary technology which restrict choice. You have warped my comment into one advocating for a fully open platform stack, a strawman. At this point, why not also argue that the games are open source too? I mean, that would be nice, but it wasn't what I was saying.

> Which are also made by companies just like Microsoft.

And that's fine, as long as it doesn't come with strings attached such as "This technology shall only run on Microsoft Windows" or "All users of this technology, even open-source projects, must pay royalty fees".

[fsloth]

Sorry, - strawman was not intended. As a person with some experience in the graphics industry, the way I see it: if one uses hardware accelerated real time graphics capability, one is always effectively locking some parts of the rendering codebase in to some specific hardware platform. No matter what the API used to program the platform, you are effectively locked in to very few vendors. There really is no rational reason to highlight Microsoft as especially bad player in this market - on the contrary (as long as they are not the only game in town, of course).

[viraptor]

> The main reason: the GPU. There is very little value in having an open specification, if the implementation sucks

I don't know, maybe we should try. Currently the implementations suck, because there's no specification, not the other way around.

> And the complexity of the graphics stack is such that nobody but huge monoliths that can focus lots of paid personnel to implement the GPU, implement the drivers, fix the drivers, fix the drivers, fix the drivers...

Substitute GPU / graphics stack with general purpose OS. Sounds plausible, yet there are lots of people working on that huge monolith for free and getting paid.

[frik]

MS actively has been trying to negated PC gaming to get consumer to XBox (closed all MS Game Studios, stopped all franchises like Age ofvEmpire and FlightSimulator series (everyone is angry about the Farmville style cheap F2P knockoffs), closed MS Windows Live, stopped DirectXdotNet and XNA, changed completely the direction of DirectX with DX10 and yet again with DX12, and the infamous privacy issues aka telemetry that half of the gamers will never upgrade to Win10 including myself). macOS, SteamOS/Linux, PS4 and WiiU/NX will get a lot more gamers - and that will continue until MS fires its third CEO sooner than later. Win10 UniversalRuntime/Store games are implemented in such a bad way it's just laughable and XBoxOne has little marketshare compared to PS4. And Win32 game market share is still huge, also thank to Steam (Valve is very much against Win10!), compared to the WinStore which has little to offer and compared to Android and iOS the WinStore on WinPhone smells and feels like it is already and legacy platform.

[emp_zealoth]

As far as I know, all the antiprivacy bullshit was introduced in Win7 too

[shmerl]

So, you think it's better not to have open standards and let lock-in and monopolies run rampant? Hard, doesn't mean it shouldn't be done.

[kordless]

I agree with you, but I also think asking leading question sucks. Just say what you mean!

Open Standards are the MOST CRITICAL thing we can be considering right now. If we can agree on standards that can adapt to other standards, using AI most likely, then we can get on with being able to trust each other again.

In order to trust each other, we have to be honest about our intent. Personally, I think the use case of games on insecure hardware is probably fine. To a point.

[shmerl]

It was a rhetorical question :) I don't think anyone in their mind would say that we should pursue lock-in rather than open standards. Except crooked monopolists.

[fsloth]

"XNA"

XNA is deprecated: http://www.gamasutra.com/view/news/185894/Its_official_XNA_i...

I find it hard to believe gaming capability would be a major motivator in the PC market. They're more of an item in the PC enthusiast segment - which is not the majority of the market.

If one looks at what sort of hardware is sold in systems - the majority is not really well equipped for high end gaming[0][1].

Therefore I conclude in the general PC market does not hold games that important.

My view of the hardware spec of the average desktop or laptop GPU is anecdotal but I would guess it's fair to say they are sub NVidia's 960M in performance (which I have on my years old 'gaming' laptop and which is kinda slow).

If someone has better statistics I'm all ears!

[0] http://www.statista.com/statistics/272595/global-shipments-f...

[1] http://www.kitguru.net/components/graphic-cards/anton-shilov...

[CyberShadow]

> I find it hard to believe gaming capability would be a major motivator in the PC market.

One of the most frequently cited reasons I hear about for why home users and enthusiasts are not using Linux is availability of games. It's true that this does not speak of the entire market, but it's not an insignificant part. It is also a market segment that is more realistic to dislodge than, e.g. corporate use.

> the majority is not really well equipped for high end gaming[0][1].

But "gaming" != "high-end gaming"...

[nurettin]

Not sure what you mean by "fighting", perhaps you mean expressing opinions and needs online which would mean pretty much everybody is fighting here.

I like to stay on linux because it is a great development environment and as soon as I am done with a game or get psyched for implementing an idea during yhe game, I am able to switch immediately back to whatever I was doing before I started playing.

Your suggested alternative of waiting for a reboot and restoring all the files and editors is just uncomfortable and I am surprised you couldn't figure this out.

[ekimekim]

a) A few years ago it was a sub 1% market share and no games bothered. Now it's a sub 5% (disclaimer: I'm making these numbers up) and i'd say approx 50% of indie games and 10% of AAA bother. We fight because we're slowly winning.

b) A dual boot setup is awfully annoying. I keep a ton of things open in a "i'm in the middle of this" state, and shutting everything down/losing that state in order to reboot is a massive cost, and it means in practice that I only ever boot into windows to play games if I plan on playing a specific game for several hours at least - it means I can't play a game casually. Over time, I've started avoiding windows only games altogether. As for not having a dedicated gaming rig: Space, mainly. And cost, secondly. I could potentially set something up, but I'd need to share monitors/keyboard for practical reasons, which is a ton of setup I don't really have time to get right. Other solutions like a windows VM, same reason.

[clw8]

I completely feel you on b. I always have a set of various folders opened and even with shortcuts it's a pain in the ass to reopen them all (because what's open depends on what I'm doing).

[nitrogen]

It didn't have to be this way. Linux and UNIX desktop environments used to excel at session management and restoration, but if you can even find the option anymore in your desktop environment of choice, it rarely keeps all of the state from all of the apps you would want.

[rhubarbquid]

a) It never hurts to let the developers know you want it

b) In theory, you should be able to hibernate Linux and boot in to Windows

> I'd need to share monitors/keyboard for practical reasons, which is a ton of setup I don't really have time to get right

I use a fairly cheap KVM, which isn't problem free, but is less hassle than swapping cables between my macbook and my PC (used for nothing but games). Even swapping the cables between 2 computers isn't that huge of a hassle ;-)

[mvkel]

It's still sub-1% market share. Would you design a site that only works on IE6?

[khedoros]

That's not an apt comparison.

- IE6 is strictly, inherently an obsolete platform, running on obsolete OSes and superseded by later versions of the same browser

- No one's talking about making the games work only on Linux.

[c22]

No, but I might design one that also works on IE6.

[eon1]

That sounds like an interesting challenge, actually..

[rtpg]

Mac used to be the same story, but now I would bet that at least 10% of indie game sales are on Mac.

If you're using tools like Unity, then shipping to Linux is not nearly the headache it used to be.

I play a lot of games, but thanks to this shift I no longer have to run Windows (can't play overwatch but I can live without it for now). I, for one, am glad I don't have to double my purchases.

[rhubarbquid]

> Mac used to be the same story, but now I would bet that at least 10% of indie game sales are on Mac.

One of the big things that changed was that Macs got a lot more popular and became a bigger market.

I think the main reason we don't see more games for Linux is that it's just not a big enough market (yet?)

[shmerl]

Actually we do see way more games for Linux. If you paid attention, their amount increased a lot in the past several years. From major engines, to major distributors, Linux is now in the gaming mainstream. Only legacy publishers mostly lag behind (the likes of EA and Blizzard). But they are always followers, not innovators.

[rhubarbquid]

I think we agree, I meant "more than we could" not "more than there used to be" ;-)

[wyager]

Almost half my 150 steam games are on mac, and I don't specifically target any type of game.

[wtbob]

> Why fight this losing trend when you could easily dual boot or even have a dedicated gaming rig?

Because I don't want to pay Microsoft. Because I don't want Microsoft to collect my data. Because I care about my freedom. What better reason is there?

[sfifs]

Sure.

But then folks like you need to be willing to pay the game developer whatever it costs them to hire an engineering bench to maintain a linux port - ideally with significantly higher profitability than the windows version to make it worth the effort.

[wtbob]

> But then folks like you need to be willing to pay the game developer whatever it costs them to hire an engineering bench to maintain a linux port - ideally with significantly higher profitability than the windows version to make it worth the effort.

I've no problem paying for Linux games. At one point I spent quite a bit on Humble Bundles (before they got boring), and I always paid twice what the averages for Windows & (then) OS X were.

[Erwin]

A Humble Bundle is a last-ditch effort to sell an old game -- appealing to hoarders for the company to gain some publicity (as some just give all to chaity). The last "2K" bundle had $360 worth of games -- for $8.97 and only a small proportion of that comes back to the company.

If you want to support Linux as a platform, I recommend you donate at higher e.g. $50 or $100 tiers to Kickstarter projects that have Linux stretch goals and be vocal about your donation.

Paying 5% of the sale price years after the game has come out doesn't strike me like fruitful.

[shmerl]

> But then folks like you need to be willing to pay the game developer whatever it costs them to hire an engineering bench to maintain a linux port

Many Linux users do. Especially with backing crowdfunded games which pledge Linux support.

[shmerl]

I have a gaming rig. And it runs Linux. Hopefully this makes things more clear for you. And I don't see any losing trend here. Amount of Linux games continues growing, same as Linux gaming market itself. So trend is clearly winning. But some developers need more pushing.

[chucksmash]

Because they a) have Linux and b) want to play games? What's not to understand?

Enjoyment of a game is not contingent on the slice of the OS market share pie chart I'm in.

[simulacrum]

In addition to all the other replies (because we have computers with powerful GPUs that run Linux, because many other games run on Linux, because all the modern games engines support Linux, etc), in this case one of the devs previous games also runs on Linux: http://store.steampowered.com/app/242110/. That seems like reason enough to at least ask the question of whether they have any Linux plans.

[vamur]

Maybe people don't want the hassle with dual booting and a $100 Windows license to play games. Instead, they can save $100-$150 more and get 480 or 1060. And thanks to 2500 games on Steam there are plenty of games on Linux now.

And Linux marketshare and number of games are still rising, so it's anything but a losing trend.

[vanattab]

I hear what your saying and agree. But in GP's defense the developer did talk about a linux release but then went quite.

[mpd]

Because as a developer, it's often easier to develop for Linux than other major platforms. You often get Linux compatability for "free". It's worth asking for for this reason.

[yladiz]

That's not exactly true. With some newer tools, it's possible to export to an OpenGL Linux executable, but in other tools, especially older ones, it's still so specialized for DirectX that's it's not always possible especially with AAA games, and companies don't have much incentive to make Linux versions because the market share is significantly lower than Windows. The only reason that it's more easily possible with game platforms like Sony and Nintendo's hardware offerings (MS uses a special flavor DirectX for Xbox) is that the two aforementioned companies plus a lot of other game dev companies pour a lot of money into developer resources that Linux distros doesn't have.

[shmerl]

You mean to say they put resources into lock-in to tax cross platform development. Those who care about the progress of the industry put resources into shared technologies like Vulkan, which makes cross platform development more affordable.

[yladiz]

Yes, some do development in OpenGL, but why would a big company put resources in the development of something unless it has tangible benefits for the company? E.g. why would a big AAA company like EA put effort into changing their tooling to support OpenGL when they make millions from consoles and PC? It would only make sense to develop for non-DirectX or change tooling if the cost of changing is outweighed by the benefit. It's the same reason why a team like Github took 2 years to change from Rails 2.3 to 3 even though 4 had already been released for a while by that point.

I'm not saying it's right, and I wouldn't mind seeing better gaming support on Linux and Mac, but I'm saying it's reality.

[shmerl]

Asking common sense questions about practices of legacy publishers is a futile endeavor. I stopped asking such questions for a while already. Those who want to innovate, do it, and find it useful for them. Size of the budget has nothing to do with interest to release for Linux and have wider cross platform reach. If anything, big publishers have more resources to do it. At the same time, innovators happen to be smaller studios, who actually are expanding Linux gaming market, while legacy publishers don't pay attention.

[JesseMeyer]

For anything seriously graphical, you're wrong.

[shmerl]

For anything seriously graphical there is Vulkan which is basically the same where it's supported.

[yoklov]

This is not a good argument, Vulkan is both brand new and much more difficult to support than other graphics APIs.

Also Vulkan support doesn't fix busted drivers, wonky fullscreen, OS integration, etc.

It also doesn't fix audio, or HID support, which is both huge headaches on linux.

[shmerl]

So what if it's new? Technology shouldn't stand in one place. MS got so scared of it, that they rushed to push their NIH lock-in alternative (DX12). It demonstrates they understand the strong potential of competition here.

Drivers can be busted anywhere (including Windows). But with Vulkan they are reduced to hardware layer which is ironed out rather fast normally. Not sure what you mean about OS integration (sounds too vague). Fullscreen, input and etc. are mostly issues related to X11, which should be cleaner with Wayland usage. The later takes longer than it should to get adopted, but all major DEs are already close to it.

Sound - never had any sound problems on Linux for a while. Are you sure you aren't measuring it by experiences of some early broken Pulse releases?

[zamalek]

> never had any sound problems on Linux for a while.

User experience is not the same as developer experience. Debian (where Pulse seems mostly standard) is not the only distro. Vulkan is expensive, time consuming and difficult to develop against - for a game that already had multiple launch delays.

[yoklov]

This is true for certain types of software, but it is neither easier nor free to develop games for linux, by any stretch.

[hackits]

Calling BS here.