Rich Geldreich, a former Valve developer, has an interesting blog post about Valve supporting Linux and OpenGL

I am not quite sure what a "legacy publisher" is. If you are referring to the companies currently releasing pretty much 100% of all AAA titles, then...yeah. That's the ones we want to release Linux games right now, no? Except you're happy getting more 8bit retro games looking like the stuff they released back in the mid 80s. Nothing wrong with (some of) them, mind you, but we have 2,500 of these already. What we need to grow as a gaming platform are games appealing to the mass market. "Oh look, Linux can do outdated graphics like we had 30 years ago!!!" isn't going to cut it.

As for these "legacy publishers", they will not change their internal processes for us 1% any time soon. They are used to making Windows games and that's what they will do for the foreseeable future. For the time being we need porting houses to get access to these games, like it or not. It's actually these porting houses that might be able to press middleware developers into providing a Linux port of the stuff that currently doesn't run in it. They ARE what's currently driving Linux game development. We got the major engines already. The battle these days is for the middleware. If you want "legacy publishers" to switch to cross-platform development, we need to become a serious gaming platform FIRST. We're not quite there yet.

Actually one has Linux support indicated.

Right! "Spacious Skies" has.


Last edited by Corben on 6 Jan 2017 at 12:29 am UTC

I am not quite sure what a "legacy publisher" is.

By legacy publishers I mean old media style companies, driven by execs who see profit more important than art. They don't make art - they are certain type of middlemen who profit from others making it. Such companies exist in many art fields. Labels in music, "huge media conglomerates" in films, publishing houses in books, and so on. Games have their analog too, i.e. EA and similar. All those I call legacy publishers. I guess it's hard to define it formally. It's kind of a mindset which combines quite a lot of negative traits, such as purposeful distancing from users of their art, strong mass market bend which sacrifices quality, cutting costs more than necessary to increase profits by reducing support terms and so on. I guess you get the idea whom I'm referring to. Such mindset in my experience usually combines with avoidance of Linux and usage of DRM, though there is no explicit correlation between these. It's more of a common occurrence.

If you are referring to the companies currently releasing pretty much 100% of all AAA titles, then...yeah. That's the ones we want to release Linux games right now, no?

Well, not necessarily. AAA is too ambiguous of a term. Some simply mean big budget games by it, and not all big budget games are made by backwards thinking legacy publishers. Some are made by quite innovative and forward thinking ones, who are interested in Linux too. And in my experience, big budget doesn't necessarily guarantee artistic quality, especially if mass market is the target. Personally I'm interested in more good games, and whether they are big or small budget is really secondary for me. I acknowledge though, that perception of art is to a degree subjective, and good can mean different things for different people.

It's actually these porting houses that might be able to press middleware developers into providing a Linux port of the stuff that currently doesn't run in it. They ARE what's currently driving Linux game development.

I'm not sure they are the biggest drive behind Linux gaming progress. I think engine developers are, and their effort is impacting on a deeper level. But any progress is good, and credit should be given to any good contribution.


Last edited by Shmerl on 5 Jan 2017 at 8:18 am UTC

The fact that Valve created SteamOS to show Microsoft Steam is not dependent on Windows was obvious from the beginning - and yet many people are surprised each time someone mentions it, one example being the famous Dell interview about Alienware Alpha/Steam Machines.

One thing I like about Steam on Linux, including SteamOS - or I should say "Steam on Debian/Ubuntu" - is the fact I don't have to care about libraries. All Steam games just work. I know it's not the case for other distros, but the experience on Debian is extremely pleasant.

It would be nice to have a robust set of libraries for gaming with some umbrella name and a version, which could be targeted by devs and given in requirements the way that Windows versions are specified. Ubuntu is very close, but it makes life more difficult for users of other distros.

That was an interesting read... It sounds like it's calling for an interview, isn't it? (if you do so, Liam, please also ask about the current status of FLOSS drivers)

Now,it isn't like Valve completely abandoned us. I would be happy with just a "back burner" status quo; if it means we can slowly grow. But Valve seems so be doing some stealthy moves, and is apparently working or planning to work on a VR composition system with integrated reprojection and such, IIRC (I didn't look it up, but my point is that they seem to be working on Linux VR).

@Liam, the smiley picker goes over the header bar on my phone, I don't think that's intended :)

The fact that Valve created SteamOS to show Microsoft Steam is not dependent on Windows was obvious from the beginning - and yet many people are surprised each time someone mentions it, one example being the famous Dell interview about Alienware Alpha/Steam Machines.

Well, I hoped that Valve would not just do the least that's required to show independence from Windows but really push the new platform even in the mid to long term. And THAT's what I don't see happening. Just as Geldrich's post hints. That is what I feared and what I'm surprised by this blog post because it names it quite directly.

After all, I am not even sure if Steam on Linux is still this threatening club over MS's head.
For that, way more AAA titles must be available with high-quality ports day one.
Don't see that happening in 2017 despite all optimistic people here.
I heard that before.

EDIT: Still, hope YOU guys are right.

EDIT2: I see that I say "ports", when I just mean Linux as a target platform just from the game dev itself.
Actually I don't want "ports".


Last edited by sub on 5 Jan 2017 at 10:36 am UTC

When the Steam OS project was announced, I was really hyped about it and that Valve will create a great gaming OS. Sadly, that didn't happen. I cannot argue with the facts, that Valve pushed Linux gaming to new heights and took part in creating Vulkan, but throughout the years, Steam OS itself seemed more and more like a garage-project, with only a handful of people working on it, just to barely sustain it. Looking back, it feels like at the beginning, they didn't even have any idea, how the state of drivers and other components were on Linux. Nowadays, you can easily get a distribution with superior performance and better hardware support, you don't even need a rolling one for that. So although I am thankful for everything else they had done, but feel that the Steam OS project was left in the dust and will slowly be buried, without anyone remembering it.


Last edited by tuxintuxedo on 5 Jan 2017 at 10:46 am UTC

For me it was always just about showing Microsoft that there is a way out for Steam. Microsoft wouldn't be able to lock down Windows overnight and Valve showed they could use that time to make Linux a viable gaming platform. Microsoft backed down, Valve put SteamOS on the back burner.

AAA is too ambiguous of a term. Some simply mean big budget games by it, and not all big budget games are made by backwards thinking legacy publishers. Some are made by quite innovative and forward thinking ones, who are interested in Linux too. And in my experience, big budget doesn't necessarily guarantee artistic quality, especially if mass market is the target. Personally I'm interested in more good games, and whether they are big or small budget is really secondary for me. I acknowledge though, that perception of art is to a degree subjective, and good can mean different things for different people.

The hipster is strong in this one. :)

Seriously, I love a good indie title, but if you play a game like Mad Max or Tomb Raider, you'll see where all those millions went. There's just tons and tons of highly polished content, from graphics and animation to audio design, made by large teams of professional artists. The mass market focus and the sometimes glaringly Hollywood-style writing-by-committee (sorry, Ms. Pratchett) does not stop them from being astoundingly beautiful and simply hours and hours of great fun.

Of course this is all subjective but assuming a game must be "less good" if it's "mainstream" is just silly. Or at least the assumption part of it is silly. Nothing wrong with informed opinion of course.

What if Valve realized that exposing their success with L4D better performance under Linux just helped MS to realize they need to reinvest in directX and in effect did not help Valve? What if Gabe did not like it that much and thought they should be less open about the progress? They surely realize OpenGL limitations and so went into Mantle->Vulkan efforts inviting a lot of players into the project. I assume similar happened to engines producers. Valve realized whole infrastructure needs to be ready first to go 100% competition with M$. Since we clearly see improvements with drivers, engines, work with VR etc., I just think Valve is just quietly improving SteamOS into preparing to hit big. Look at how Nintendo, Sony or MS are working on new platforms. They are quiet until they are sure they can hit hard. This is valid and verified strategy.

I don't think they're pushing for a big hit with SteamOS / SteamMachines anymore. It's exactly what's said here, it's the hammer above Microsofts head.

Vulkan, while being a hammer over the head, is a serious competitor. So Microsoft needs to invest a lot there, since they want people to develop DX12 for Windows and XBox, and don't want to become a 2nd class citizen API whise being the "port" rather than the main platform. One who could have changed this would have been Sony, if they decided official support for Vulkan in PS4. But there is no official word about that yet (not one that I've found).

I do not consider Linux distributions or SteamOS a serious competitor to Windows. Not in the next decade, not if Microsoft does not fuck it up royally. My bet is that Google will go competitive one day on the Desktop market with an Android / Linux base or even their own Kernel.

Why? Windows is known, is sold with the PCs, has all titles, and runs the software people know as well as everything runs on Windows, which not necessarily runs on Linux. People are used to things, and they want the exact program, not some "other program which can do this". Android or a "ChromeOS" or what ever it will be called is a more serious competitor, since Google does have all the apps.

Call me pesimistic, but I think that's what we're facing in reality.

AAA is too ambiguous of a term. Some simply mean big budget games by it, and not all big budget games are made by backwards thinking legacy publishers. Some are made by quite innovative and forward thinking ones, who are interested in Linux too. And in my experience, big budget doesn't necessarily guarantee artistic quality, especially if mass market is the target. Personally I'm interested in more good games, and whether they are big or small budget is really secondary for me. I acknowledge though, that perception of art is to a degree subjective, and good can mean different things for different people.

The hipster is strong in this one. :)

Seriously, I love a good indie title, but if you play a game like Mad Max or Tomb Raider, you'll see where all those millions went. There's just tons and tons of highly polished content, from graphics and animation to audio design, made by large teams of professional artists.

Millions went somewhere, but it bears a strong resemblance to current day mass market film industry. Most of it is pulp fiction style writing, with flashy graphics and CGI, and barely any depth in the script and story. Or in the games context - reactivity with choices and consequences. Sure, there are good authors out there. But most of the mass market is after flashy things, not after depth. I value more than "polished content" in art.

Of course this is all subjective but assuming a game must be "less good" if it's "mainstream" is just silly. Or at least the assumption part of it is silly. Nothing wrong with informed opinion of course.

It doesn't have to be worse, but it easily can become worse. As soon as mass market kicks in, quality of art can easily be compromised for superficiality with flashy look. I'm not trying to downplay the value of good graphics - I like it same as the next guy. I'm just pointing out the pitfalls of mass market / commercialized art driven by legacy publishers.


Last edited by Shmerl on 5 Jan 2017 at 4:42 pm UTC

Seriously, I love a good indie title, but if you play a game like Mad Max or Tomb Raider, you'll see where all those millions went. There's just tons and tons of highly polished content, from graphics and animation to audio design, made by large teams of professional artists.

Millions went somewhere, but it bears a strong resemblance to current day mass market film industry. Most of it is pulp fiction style writing, with flashy graphics and CGI, and barely any depth in the script and story. Or in the games context - reactivity with choices and consequences. Sure, there are good authors out there. But most of the mass market is after flashy things, not after depth. I value more than "polished content" in art.

You are talking about your personal taste in games and movies here, and that's fine. We're all entitled to our opinions. Though surely you shouldn't judge the entertainment value of a game or a movie based on its budget? Of course this applies equally well to the AAA-only indie-bashers. No need to pick a side, you know.

Of course this is all subjective but assuming a game must be "less good" if it's "mainstream" is just silly. Or at least the assumption part of it is silly. Nothing wrong with informed opinion of course.

It doesn't have to be worse, but it easily can become worse. As soon as mass market kicks in, quality of art can easily be compromised for superficiality with flashy look. I'm not trying to downplay the value of good graphics - I like it same as the next guy. I'm just pointing out the pitfalls of mass market / commercialized art driven by legacy publishers.

Sure, a big budget can be used to polish a turd, but that doesn't mean everything that shines is hiding shit underneath.


I hope you don't take my posts here as personal attacks. I value your opinion, and your contributions on GOL. I'm just trying to point out that bashing games you admit you haven't even played does seem a bit counterproductive. I also think most of the stuff people watch and play is utter crud, but we have to recognize what is subjective opinion and what is objective fact.

I'm enjoying Mad Max at the moment by the way. AAA, and definitely shallow, but tons of fun. That doesn't mean I enjoy a good old-school adventure or RPG any less.

The result is the same, Linux is still the second class citizen, the big fluffy cushion down the cliff existing just in case Microsoft goes overboard.

It's really saddening.

In the long run, game development on Linux will be dependent on the same thing it's always been dependent on for every other platform, the size of the audience.

I remember when Commodore Amiga and Atari ST games were much better and more plentiful than IBM PC games. The fact was, even though those platforms were technically better at games than IBM PCs, DOS games started to become popular because more people bought DOS based systems to be compatible with their work computers. DOS became a bigger market and started to get the most attention by game developers.

Then DOS games were better than Windows 3.1 games, but as Windows became a bigger market, and Windows 95 hit, Windows games started to become the most prominent computer games.

The point is, when it comes to general purpose computing devices, both the graphical technology and the games that get developed are driven by existing platform popularity. Of course, if sufficient interest exists to push graphics technology and game development, then this creates a synergy and things progress faster.

The main advantage Linux has is that it can't be knocked out of development by a brilliant business move. If open source software had a theme song, it would be "Time Is on My Side."

While my brothers have Windows computers, they all have Linux computers in their households at this point as well. Linux computers have become extremely practical as second or third computers for households with multiple children, since they need less powerful hardware and require less maintenance than Windows computers, and since children adjust easily to a different platform. This type of growth of the Linux audience may seem slow, but it is beginning to have an effect. It may be part of the reason that older ports and retro games are more popular on Linux, because even older hardware runs them well, but eventually any growth of the market promotes more sophisticated technology.

I have to say, I'm really surprised by some of the pessimistic comments being made. Did some people think SteamOS was going to make Linux jump mountains this quickly? It was never going to.

The fact is that it has helped Linux improve in a great many ways in a relatively short amount of time.

It's going to be another good year for us.

So basically Linux support is a scarecrow now. They've put it on a life support and they are throwing bits and pieces at us to keep the crowd satisfied but they won't bring out the big guns because they don't want to go to a war with Microsoft. So unless Microsoft decides to push the store or seal the PC environment around Windows whatever that means there won't be a big Linux push. Some of the people might migrate but it won't ever be more that few % without drastic Microsoft moves. Also SteamOS would be more of a thing if it happened few years ago when people were dissatisfied with Vista, then 8 and when you actually had to pay for Windows. With 10 you just download ISO from the Microsoft website and off you go. With or without a license, it works anyway. Good enough for most of the people I guess. So this is it guys, if you think 2016 was good then hold on to it because all the subsequent years will be the same. Similar marketshare, similar amount of games released, similar treatment by developers unless they come to a conclusion that Linux port only pays the bills and it's not viable. Maybe Valve would be more successful with ReactOS lol.

Well, competition is always good and you shouldn't forget that Valve is a lot of ex Microsoft engineering people who are still loyal to their colleagues from the development departments there. It is a bit sad, that they didn't really push further with gaming on Linux and made the press call it a big failure of 2016.


Last edited by Milanium on 7 Jan 2017 at 7:49 pm UTC

I do find it funny all of the 'SteamOS has clearly failed' and 'it hasn't been updated that much' or 'it's on the back burner.'

Do people realize that it's based on Debian Jessie? And that Debian's release schedule is.. glacial at best? Do people want a 6 month release schedule like Ubuntu? How much extra work by Valve would that take? There is a reason why they chose Debian. Debian Stretch is going into a Soft Freeze soon, then hard Freeze, so there should be a HUGE update to SteamOS once that happens.

I'm wondering how much of Valve's work is just being put back into Debian proper. Anyone who has been following the development of it, should be quite excited for the potential SteamOS.

Debian Stretch is going into a Soft Freeze soon, then hard Freeze, so there should be a HUGE update to SteamOS once that happens.

Maybe. Jessie will be on long term support ["to end of April/May 2020](https://wiki.debian.org/LTS)" so there's no actual need to hurry. I'd imagine they'd want to do it sooner rather than later though, unless they want to deviate even more from upstream and take a heavier maintenance burden themselves.

I'm wondering how much of Valve's work is just being put back into Debian proper. Anyone who has been following the development of it, should be quite excited for the potential SteamOS.

I don't think they actively upstream anything much. Any changes are in the open SteamOS repo, but Debian maintainers pretty much need to actively trawl for patches. Valve doesn't seem to have an actual policy of keeping a minimal patchset (like Ubuntu claims to) unless I've missed something.