Steam for Linux entered Beta five years ago today

There was never really a technical hurdle to supporting Linux. We've had OpenGL for years, and id Software's ports and Epic's experiment with Unreal Tournament 2004 proved a long time ago that Linux was more than capable of running AAA games. The real problem was, and still is, market share, and I'm a little concerned that if Valve ever expressed any lack of confidence in the Linux platform that a lot of developers would simply walk away.

I would not underestimate the distribution part. Having a channel to users for paying and actually getting the software is more complex than it sounds.
Now I see market share, same as you, indeed as _the_ concern.


Last edited by jens on 6 Nov 2017 at 6:56 pm UTC

My question is, is there enough momentum in the Linux gaming market for developers to continue supporting Linux in the event that Valve gets bored and kills off their grand experiment? That's my primary concern at the moment, that we're almost entirely dependent on Valve and Steam for the majority of our games, and if Valve goes away then we're going to be right back where we were five-years ago with a niche market that only a handful of independent developers even bother to support.

I guess the critical moment was when all major engine creators added Linux support to their products that make it (comparatively) easy to publish cross-platform. There is barely any reason NOT to support Linux these daysz

Let's be honest, as a game developer/publisher, Valve has been a non-factor for a while now, so if they chose to stop porting games to Linux, really, who'd notice? And Steam Machines never picked up any steam to begin with. Linux gaming has grown without them just nicely.

No, their great contribution is and was porting the Steam client to Linux in the first place, opening a mainstream publishing channel for developers on Linux for the first time ever. That deed is done and time won't get rolled back on that. The biggest help we're getting from Valve these days is them getting involved in improving Linux drivers and components for gaming. Which should hopefully drive down Linux related support costs and performance issues, which is the LAST sound reason larger studios still have not to publish on Linux.

There was never really a technical hurdle to supporting Linux. We've had OpenGL for years, and id Software's ports and Epic's experiment with Unreal Tournament 2004 proved a long time ago that Linux was more than capable of running AAA games. The real problem was, and still is, market share, and I'm a little concerned that if Valve ever expressed any lack of confidence in the Linux platform that a lot of developers would simply walk away.

Knock as hard as you can on the biggest piece of wood you can find.... NOW !
:'(

What year was that event?

http://www.teamfortress.com/linux/

Found it: It was at 2013 :D


Last edited by wolfyrion on 6 Nov 2017 at 8:04 pm UTC

My question is, is there enough momentum in the Linux gaming market for developers to continue supporting Linux in the event that Valve gets bored and kills off their grand experiment? That's my primary concern at the moment, that we're almost entirely dependent on Valve and Steam for the majority of our games, and if Valve goes away then we're going to be right back where we were five-years ago with a niche market that only a handful of independent developers even bother to support.

I guess the critical moment was when all major engine creators added Linux support to their products that make it (comparatively) easy to publish cross-platform. There is barely any reason NOT to support Linux these daysz

Let's be honest, as a game developer/publisher, Valve has been a non-factor for a while now, so if they chose to stop porting games to Linux, really, who'd notice? And Steam Machines never picked up any steam to begin with. Linux gaming has grown without them just nicely.

No, their great contribution is and was porting the Steam client to Linux in the first place, opening a mainstream publishing channel for developers on Linux for the first time ever. That deed is done and time won't get rolled back on that. The biggest help we're getting from Valve these days is them getting involved in improving Linux drivers and components for gaming. Which should hopefully drive down Linux related support costs and performance issues, which is the LAST sound reason larger studios still have not to publish on Linux.

There was never really a technical hurdle to supporting Linux. We've had OpenGL for years, and id Software's ports and Epic's experiment with Unreal Tournament 2004 proved a long time ago that Linux was more than capable of running AAA games. The real problem was, and still is, market share, and I'm a little concerned that if Valve ever expressed any lack of confidence in the Linux platform that a lot of developers would simply walk away.

Depends what you mean by technical hurdle. I'd agree that since forever it's been probably as easy to develop a game for Linux (alone) as for Windows (alone). But it was always much harder to develop for more than one platform than for one platform, no matter which additional platform you're talking about. And since for games, "one platform" meant Windows . . .
I agree with Kimyrielle that the big change is support for Linux on the major game engines. For a really big chunk of the game market from most of the indies up through some fairly major outfits, that lowered the cost of publishing games cross-platform so that the break-even point happens at a much smaller market share. That one factor makes releasing for Linux even at its current market share financially viable, maybe even closing on uncontroversial, if you're using one of those engines. There have been other increments as well. Vulkan may prove to be another again, as well as advances in packaging--Snaps and whatnot--and the move to Wayland. But the engines were the big one.
So if Valve walked away it would obviously not be a good thing, but the situation would be IMO very different from how things were pre-Steam. Developer assumptions are different now, and there are probably more developers with some idea how to build a game with Linux in mind, but above all it's simply a lot cheaper and easier to make games with Linux as one of the release platforms.

The other question of course is, how likely is Valve to walk away? But the question underneath that is, how likely is Microsoft to convincingly not be covetous of Valve's market? How likely is Microsoft to abandon things like introducing and gradually integrating app stores in Windows and trying to get games to be sold there rather than on Steam? I'd say fairly unlikely; that's a lot of revenue and control MS would be leaving in an outside third party's hands when they have a shot at getting it themselves. And it's even more unlikely that they'd be able to not only abandon the idea of doing that, but also convince Valve that there's no threat of them ever doing that. Companies, much less Microsoft, don't permanently abandon potential large revenue streams.
So my expectation is that, for as long as MS don't make any serious moves, Valve will play Linux-related stuff like now--quietly simmering on the back burner, ready to be brought to a boil when necessary.

Crystal balls are notoriously inaccurate, though. So I don't discount the worry--who knows what could happen? But Linux is on those game engines now. Things are a bit different.


Last edited by Purple Library Guy on 6 Nov 2017 at 8:48 pm UTC

If 5 years since first reading GOL makes you feel old, how old do you think I feel when I still remember typing this....

10 print HELLO
20 goto 10
run

10 print HELLO
20 gosub 10
run

*eg*

My question is, is there enough momentum in the Linux gaming market for developers to continue supporting Linux in the event that Valve gets bored and kills off their grand experiment? That's my primary concern at the moment, that we're almost entirely dependent on Valve and Steam for the majority of our games, and if Valve goes away then we're going to be right back where we were five-years ago with a niche market that only a handful of independent developers even bother to support.

I guess the critical moment was when all major engine creators added Linux support to their products that make it (comparatively) easy to publish cross-platform. There is barely any reason NOT to support Linux these daysz

Let's be honest, as a game developer/publisher, Valve has been a non-factor for a while now, so if they chose to stop porting games to Linux, really, who'd notice? And Steam Machines never picked up any steam to begin with. Linux gaming has grown without them just nicely.

No, their great contribution is and was porting the Steam client to Linux in the first place, opening a mainstream publishing channel for developers on Linux for the first time ever. That deed is done and time won't get rolled back on that. The biggest help we're getting from Valve these days is them getting involved in improving Linux drivers and components for gaming. Which should hopefully drive down Linux related support costs and performance issues, which is the LAST sound reason larger studios still have not to publish on Linux.

There was never really a technical hurdle to supporting Linux. We've had OpenGL for years, and id Software's ports and Epic's experiment with Unreal Tournament 2004 proved a long time ago that Linux was more than capable of running AAA games. The real problem was, and still is, market share, and I'm a little concerned that if Valve ever expressed any lack of confidence in the Linux platform that a lot of developers would simply walk away.

Depends what you mean by technical hurdle. I'd agree that since forever it's been probably as easy to develop a game for Linux (alone) as for Windows (alone). But it was always much harder to develop for more than one platform than for one platform, no matter which additional platform you're talking about. And since for games, "one platform" meant Windows . . .
I agree with Kimyrielle that the big change is support for Linux on the major game engines. For a really big chunk of the game market from most of the indies up through some fairly major outfits, that lowered the cost of publishing games cross-platform so that the break-even point happens at a much smaller market share. That one factor makes releasing for Linux even at its current market share financially viable, maybe even closing on uncontroversial, if you're using one of those engines. There have been other increments as well. Vulkan may prove to be another again, as well as advances in packaging--Snaps and whatnot--and the move to Wayland. But the engines were the big one.
So if Valve walked away it would obviously not be a good thing, but the situation would be IMO very different from how things were pre-Steam. Developer assumptions are different now, and there are probably more developers with some idea how to build a game with Linux in mind, but above all it's simply a lot cheaper and easier to make games with Linux as one of the release platforms.

The other question of course is, how likely is Valve to walk away? But the question underneath that is, how likely is Microsoft to convincingly not be covetous of Valve's market? How likely is Microsoft to abandon things like introducing and gradually integrating app stores in Windows and trying to get games to be sold there rather than on Steam? I'd say fairly unlikely; that's a lot of revenue and control MS would be leaving in an outside third party's hands when they have a shot at getting it themselves. And it's even more unlikely that they'd be able to not only abandon the idea of doing that, but also convince Valve that there's no threat of them ever doing that. Companies, much less Microsoft, don't permanently abandon potential large revenue streams.
So my expectation is that, for as long as MS don't make any serious moves, Valve will play Linux-related stuff like now--quietly simmering on the back burner, ready to be brought to a boil when necessary.

Crystal balls are notoriously inaccurate, though. So I don't discount the worry--who knows what could happen? But Linux is on those game engines now. Things are a bit different.

All good points. I don't mind conceding that I'm probably being a bit paranoid.

Five years ago I discovered GoL because I was searching news regarding Steam for Linux :D F*ck, I'm getting old.

If 5 years since first reading GOL makes you feel old, how old do you think I feel when I still remember typing this....

10 print HELLO
20 goto 10
run

I know I must have typed that and a lot more on my C-64, but I guess I'm too old to actually remember it.

Five years ago I discovered GoL because I was searching news regarding Steam for Linux :D F*ck, I'm getting old.

If 5 years since first reading GOL makes you feel old, how old do you think I feel when I still remember typing this....

10 print HELLO
20 goto 10
run

Now that takes me back to my TRS-80. Well, technically my dad's TRS-80.
2K of ROM, 4K of RAM.


Last edited by Purple Library Guy on 6 Nov 2017 at 9:54 pm UTC

If 5 years since first reading GOL makes you feel old, how old do you think I feel when I still remember typing this....

10 print HELLO
20 goto 10
run

10 print HELLO
20 gosub 10
run

*eg*

I don't touch basic for really long time, but code with gosub you will do a "stack overflow" because it's recursive, better do a goto. :P

Edit: maybe it was only with ON GOSUB... i don't remember well, to much time :P


Last edited by bubexel on 6 Nov 2017 at 10:08 pm UTC

Does anyone remember TF2 Tux hat?

I think I only ran TF2 long enough to get that in my inventory...

It is no coincidence that I joined Steam 5 years ago.

Where do you find that? I couldn't find it in my profile...

I think I only ran TF2 long enough to get that in my inventory...

I missed this opportunity. :-(

Where do you find that? I couldn't find it in my profile...

You get a badge every year.


Last edited by Eike on 7 Nov 2017 at 7:44 am UTC

You get a badge every year.

Ah, thanks! Shows how much attention I pay to badges, even though the top one claims I am a "Gaming God".

Ah, thanks! Shows how much attention I pay to badges, even though the top one claims I am a "Gaming God".

*readingitup*
2000 games or more?!? :O

Oh, I remember that time. I was very exited for this, even got an invitation to the closed beta. :)

At that time there was a crazy amount of 2 games in my library available for linux. Well, actually just one:


But it worked quite well:


The 4:3 dimension of the images is somewhat disturbing to look at, now...

Five years ago I discovered GoL because I was searching news regarding Steam for Linux :D F*ck, I'm getting old.

If 5 years since first reading GOL makes you feel old, how old do you think I feel when I still remember typing this....

10 print HELLO
20 goto 10
run

I know I must have typed that and a lot more on my C-64, but I guess I'm too old to actually remember it.

I knew that machine inside out and wrote an assembler and BASIC toolkit! Can't remember any of it now though (I do still have the 5¼" floppies I used in my 1541/1571 drives though!)

5 years, Wow! doesn't time fly when you're having fun :)

claims I am a "Gaming God"

2000 games or more?!? :O

In the early days of Steam for Linux, I bought cheap bundles even if they only had a couple Linux games. It's "only" 1094 Linux games (more than half, at least!), and certainly a lot of them I will likely never even download... :|

5 years... well all considered I can say Valve had a successful and fertile ground with Linux. I still don't know the meaning of the S machines (in my mind they feel like they are still to come out) but Steam runs perfectly and I am certain it made an important step for all gamers. Having that linux release simply didn't exist before. Now it is expected for some developers like paradox or indie games (generally talking).

I still think the availability of games wont make Linux more popular simply because it doesn't run as default when you buy a new computer and most people cant be bothered.

All in all, well done and let's never feel too comfortable with what we have, lets fight for more!