Linux Gaming: An Upward Trend | Hacker News Reader

Linux Gaming: An Upward Trend

pandoralive.info

135 pointsekianjo65 comments

I found it interesting to note from the article that the two main AAA titles cited (Civ 5 and Xcom) were ported by the two biggest Mac porting houses (Aspyr and Feral respectively). They may be new to the Linux market but they have extensive experience porting from DirectX to OpenGL and an impressive library of AAA completed projects.

Valve's Steam Machines may prove to be an interesting new business opportunity for these Mac porting houses - and hopefully the possibility of bringing games to not one, but two platforms, will mean yet more Mac ports as well.

As was noted in the Xcom vignette in the article, they'll likely have some difficulty dealing with Linux users (as opposed to their experience with Mac users) from a customer service perspective. They aren't in Kansas anymore!

I am not sure, I have never seen a Steam machine anywhere or anyone being interested in them. With the quickly growing power of ARM CPUs and GPUs, I think in a short time, an Android or iOS-power mediacenter will be enough for 90% of the people.

Of course, since Android uses OpenGL as well, it's probably good for Mac.

Steam machines weren't released yet.

Also, while mobile GPUs are definitely stronger than they were 10 years ago, they still have a lot more to go to even match the performance of a console. Not to mention top it/reach that of a PC.

Parent stated "enough for 90% of people", not "enough for 90% of games".

Read the previous sentence or the grand parent. The context is games.

I do not see why AAA PC game companies would care about what 90% of the people in the world think. What they care about is what gamers think, since those make up most of their customer base.

They care because ideas propagate.

The latest devices, using Metal and Android extensions certainly seem close to matching up with the WiiU, X-Box360, and ps3. I don't think they will match the ps4/XBone but the Unreal Engine/CryTech demos using those beta builds has been very impressive.

Apart from SteamOS itself, I feel like OS X's rise in marketshare has helped as well; Once you've ported your engine to OpenGL, supporting Linux as well as OS X isn't much harder in my experience (as long as you've got a decent abstraction layer).

What else has driven it? iOS/Android? I doubt those, other than perhaps making AAA developers consider OpenGL at all for various projects. The latest consoles, while x86, aren't OpenGL at all, so that can't be it. It's a really interesting state of affairs; I'd never thought I'd see the day where AAA games that weren't made by id were released on Linux!

I'm not sure about this. Obviously there's a cost in the actual port of the game, but I suspect there's a lot of work besides that to support any new platform.

I made some amateur games in python and they're trivial to run in Windows and Mac (being Linux my dev platform), but there are lots of small details that can ruin the user experience just because I didn't do proper testing in all the platforms (like some Windows OpenGL implementations not supporting shared contexts, or openal glitches in Mac).

My point is that is not just having a port, you have to support it, and that costs time and money that may not be worth the effort given the number of potential users of that new platform.

EDIT: sent the comment too early

Obviously there is an additional cost; I think the point the parent was making is that the cost to add your third platform is lower than the cost to add your second platform, especially since one of the hardest bits (changing graphics stack) is mostly shared work between Mac and Linux.

The cost is still non-zero, of course, and for the most part Linux gamers have to live with a lower degree of polish than Windows or even OSX gamers because of that.

I think another factor is probably the rise of a few big engines, particularly Unity and Unreal Engine, which are taking most of the work out of supporting multiple platforms. There's still some porting involved, but most of that will be focused on the additional code written for a game, rather than attempting to translate the entire engine from DirectX to OpenGL.

I think we're getting more games for OS X and Linux because users have more ways to make themselves heard. We've got Kickstarter, reddit, Twitter, HN and a million different ways to tell them we want more than a Windows version.

Developers are figuring out that a Mac and Linux port earns sales from a dedicated minority of gamers.

This may explain the additional indie games support Linux gamers get, but certainly not AAA games'. There was no way to tell Firaxis that we wanted Civ 5 and how sizable the Civ 5 Linux community would be and what they could expect in terms of profits. There's something new at work here.

I recently installed Steam on my Ubuntu 14.04 Laptop and was positively surprised at how many titles are already available for Linux. Now I can finally play a quick game after work without having to reboot my machine to Windows, which is a true win ;)

I hope that the additional push created by this will help drive more "mainstream" users to Linux, which should help to make the whole ecosystem more mature and accepted.

Ditto! Not to mention the fact that Xbox controllers (wired ones, at least) seem to be truly plug-and-play. They just work. I play BattleBlock Theater almost exclusively on Linux.

Of course there are still some glitches. I can't play games while using Xmonad since the games don't play nicely with non-reparenting window managers. But if I load gnome it's fine.

Things seem to work better for me if I float the game window; maybe try that.

How do you get the game window to float if it never appears? Like if I have the Steam window floating, for example, by dragging with mod+<left mouse>, if I just launch the game normally will it Just Work(tm)?

Good to see more games coming to Linux, even though I game only rarely. They forgot to mention also Carmageddon:Reincarnation, funded by Kickstarter, now already available on Steam in preview for Windows, and should be coming to Linux later as well! One of the few games that I'm really looking forward to :)

> CD Projekt Red has announced that The Witcher 3 will be available for SteamOS (and therefore Linux)

This is incorrect. CD Projekt Red never announced such thing. These false news are caused by one banner which appeared on Steam front page. It was quickly pulled though (so it most probably was some prank), and GOG / CDPR commented that no such thing was announced by them.

If you actually want such port to be made, vote for it: https://secure.gog.com/wishlist/games/to_cd_projekt_red_brin...

OP here, Thanks for pointing that out. I will correct it.

Please, spread the word about that wishist item on GOG. I'm sure more people would be interested in it, than 1600+.

Actually I could not find any official refutation from Projekt Red that Witcher 3 is NOT coming to Linux... do you have any source?

This question was directed to GOG (CD Projekt Red sister company) regarding their own preorder. I.e. whether preordering there will include the Linux version (since it was supposedly announced on Steam). GOG representatives answered like this (this was all by e-mail):

> Hello

> As for now there were announced only PC, Xbox One and PS4 versions of the game.

> Regards

I see it as an official refutation. There is some hope still, since they said "as for now", but I definitely won't preorder anything until it will be officially confirmed.

Biggest issue (for me at least) is 3D graphics performance, or rather lack thereof.

My laptop dual-boots - I use Windows for gaming and Linux (Mint currently) for almost everything else.

Any games I can (Dwarf Fortress, etc) I run in Linux, but there are many games that either just flat-out cannot run or run too sluggishly to be playable.

Have to agree here at the moment, though I blame GPU makers. Ati's Linux drivers are problematic (open source Radeon is great but lacks performance wise).

I don't know about ATI but nVidia has been great for me on Linux going back to 2007 when I switched to Linux (Ubuntu) FT.

Are you using proprietary drivers or open source drivers? There's a very large performance gap (with proprietary being better).

On even my old machine I tried running UT 2004 on Windows then natively on Linux then on Wine on Linux and Linux beat the performance on Windows hands down.

I'm not the only one, when Valve ported their engine to Linux the performance was higher, I believe I read recently it's still higher than on Windows.

edit: I use nVidia graphics cards and the restricted drivers in the Ubuntu repository, old machine and my new one.

I'd have to agree, I can run most of the source engine games at max quality 60fps on my laptop running Windows. When I switch to my linux install on the same box it's much slower and a much less polished experience.

X-Com is currently eating my time together with Pandora: First Contact (which was inspired by alpha centauri). Still can't warm to Civ5, but getting a Civ4 port would be great. I'm also quite intrigued by the civ: beyond earth game, though only time will tell how good that is. And probably not on release, Firaxis has a history of needing a lot of time to polish the games.

Anyway, good times. I'm throwing money at all the Linux games. Hope the trend continues.

Regarding Civ 5: you should pick up the two expansions (Gods & Kings and Brave New World). Together they add:

- Religion - Trade routes - Ideologies (enhanced late game Social policies) - Tourism (enhanced cultural victory) - Spying - World congress (binding resolutions! and makes diplomacy actually useful, buying votes from AI civs etc) - Unique civs, like Austria who can buy city states to expand their empire

Vanilla Civ 5 was quite disappointing and dull (get science, make army, total war). With GnK and BNW it is a far more nuanced game, and there are more valid win conditions than just domination.

I think both expansions are on sale as part of Steam's summer sale.

I have GnK but not BNW. I've been playing some co-op games with my friend. So far the games have been very slow-paced and lacking in the decisions-per-turn department. Sid Meier's Alpha Centauri continues to be the gold standard for the genre, in my opinion. It has a much quicker pace of gameplay and the strategies are far more varied and nuanced.

A lot of the design decisions in Civ 5 really turn me off, though none more so than the way they balanced the game in favour of small numbers of very large cities. I really hate how counterproductive it feels to build settlers!

I found it got a little better when I turned off battle animations (I actually kind of like having fewer decisions per turn.)

That said it still doesn't match Alpha Centauri. I miss the narrative angle especially, which was very unique in a Civ-style strategy game.

If your are interested in Grand Strategy and Linux Gaming, you should take a look Europa Universalis 4. Currently runs perfectly on Ubuntu on my second screen ;)

I second this, EU4 and Crusader Kings 2 are amazing games which runs perfectly on Linux. Only thing you may want to do set noatime to fix the problem with slows start up.

One game I am looking forward to the Linux port is Transistor by Super Giant Games. They said when it first comes out, it will be available on PS4 and PC. And they will be working on a Linux and Mac port soon afterwards[0].

As for the trend, I have normally seen the Linux port as a stretch goal for Kickstarters. While it's a shame it isn't a priority from the start, I feel that this is another sign of the increasing trend. Developers are considering it an incentive, which isn't a bad thing.

Another thing I think the author left out is Steam In-home Streaming [1]. It allows you to play any game that can be played through Steam, on a non-windows computer that has Steam on the same LAN. (I think even those that aren't on Steam, if you add it to steam as a non-steam game. I haven't tested this though) While not the most optimal setup (a powerful windows machine dedicated to streaming one game at a time, with a dumb terminal or another computer without windows), I feel it is designed with the SteamOS in mind to increase the amount of games that can be played. Those that already have a powerful gaming computer can hook up a Steam Box to their living room TV and play any game they have access to (and through Steams ability to share libraries, some you haven't purchased by your family (or friends) have purchased).

I've tested the Steam In-Home streaming myself. I was using a very under-powered laptop at the time. If that were the case, I'm sure the latency over LAN would be negligible. Which I believe is the fatal flaw that OnLive had.

[0]: http://supergiantgames.com/index.php/2013/06/transistor-comi... [1]: http://store.steampowered.com/streaming/

> Another thing I think the author left out is Steam In-home Streaming

I did not talk about it since this is not purely relevant for Linux Gaming. You can do in-home streaming between a Windows desktop and a windows box next to your TV as well. There's nothing Linux specific and it requires that you have a Windows PC install somewhere in the house. Only a tiny fraction of people are going to do that at home, while this may increase a little when Steam Machines are available.

> I did not talk about it since this is not purely relevant for Linux Gaming.

True. It isn't the cheapest or most elegant option. It may not be Linux specific, but I feel it deserved a mention as it can increase the availability of games to the Linux market. But I'm not you.

Is there some way to say in Steam "I'd buy this game if it was ported to my platform"?

Yes, via the discussion forums in the Community Board :) Lots of people usually start a Linux thread there for popular games, requesting specific Linux port in order to purchase the games.

Steam on distros other than Ubuntu (or a distro with similar package versions) sucks ass. I know they're working on the issue, but until then, Steam on Linux is completely useless to me.

What distros are you using? I've had no problems on Gentoo and have heard similar from Arch users. Sure it doesn't play nicely with the wider ecosystem (understatement), but it's been a long while since I could agree that Steam on Linux sucks.

Gentoo on ~amd64 with open-source radeon drivers.

You will have severe problems running anything OGL based at decent perfromance on OSS radeon drivers anyway. Sadly.

The radeon drivers have actually been improving a lot. Their performance is fairly close to using the closed ones now, although, Catalyst does suck on Linux, but at least the open source ones are getting better.

While I like and appreciate this trend, trending up from virtually nothing shouldn't really be too much of a surprise once SteamOS was announced.

It may not be a surprise, but there were a lot of naysayers whenever Steam arrived for Linux. Yet it seems to gather more and more support, and not just from indies.

Getting indie developers on board with Linux shouldn't be hard in the coming future. Linux is rapidly growing and is started to be seen by the public eye as an OS some people would actually personally use for their own personal computer. Also, Unity and Unreal Engine 4, the two biggest engines indie studios use to develop 3D games, both easily support developing for Linux. Lastly, and most importantly, these guys don't have a choice. If they want to compete with the triple-A market they have to expose themselves to the biggest market they can.

To get the triple-A companies on board is a tougher problem. The most-thorough and simplest solution is to get more Linux users. Triple-A companies use engines built on top of DirectX because DirectX works much better than OpenGL on Windows and everyone uses Windows (like, everyone. Windows still has 90%+ market share). If you pulled the rug under that logic by having everyone use Linux, you would force triple-A companies to support multiplatform engines and graphics APIs. Just because this solution is simple doesn't mean it's easy. Linux market share is currently pathetic. There are more Vista users than there are Linux users. While on forums and discussions and other virtual agoras, evangelists annually preach "the year of the Linux desktop" and with each patch to every WiFi driver pundits will make blog posts about how better Linux is getting and how no one likes Windows 8 and just-you-wait, the fact of the matter is, before Linux topples Windows in the market share graphs, it has to start toppling "Other" in the market share graphs. And it's pretty obvious why Linux isn't getting the market share it deserves. 1. No one wants to install a new operating system on their computer, and the fact that people have to know what an operating system is at all is a tragic case of a leaky abstraction, because people don't want operating systems, they want pictures and videos and music and Facebook and e-mail, 2. No one wants to deal with WiFi and graphics driver issues, and 3. No one wants to think for a single second past "Best Buy" when they want to buy a new craptop for their lovely little girl going to college (which I think is why Win8.x is beating both OS X and Linux combined despite being an objectively worse OS), and you sure as hell won't be finding any Linux at Best Buy. Marketing exists for a reason. Older computer geeks with long beards remember buying the Macintosh 128K for being the first computer with an effective implementation of a GUI and a mouse; many other people will remember buying the Macintosh because the 1984 ad caused Apple to explode in the public eye and turned Steve Jobs from one of those computer people into a heavily publicized rock star. Apple would continue to master and practically define hype, and while OS X isn't faring as well in the market share war, what Apple really is focusing on -- iOS -- is killing it. Until the Linux folks discover how important marketing stuff is, they won't capture the people who don't care enough about their machines to think past "what's the most popular thing being used right now?"

These are huge issues, and until they get solved, you won't be seeing Linux with a high marketshare, which means no triple-A company will bat it a single eyelash.

The easiest solution, although less effective than the first one, is to increase marketshare in platforms that are not Windows. This is because most platforms that are not Windows will require use of APIs that are multiplatform, making it frictionless to port to Linux. Somehow get people to game through the browser by making it possible to run triple-A games in Chrome? Good. Linux has Chrome. Get people to game mainly on OS X? Good. OS X games usually use OpenGL, and that's easy to port to Linux. Because this isn't as direct, you will always get some people who give so little shits about a set of OSes that have less users than "Other" does that they won't go through the effort of clicking the checkbox next to Linux in the engine project setup page, but it's somewhat of a solution.

Just check this out.

http://www.techonsolution.com/which-is-the-best-operating-sy...

The cross-platform tide lifted the Linux boat. That's it.

That may be it, but it's also the best possible scenario.

I am exclusively a Linux user. I work, browse the internet and game on Linux computers. However, I am strongly against so-called platform exclusives. Everything multi-platform is the best possible scenario for users. I wouldn't want Linux-only games; I want games to be available on Linux, Windows, OS X, etc. This is not always possible due to time and money constraints, but it's desirable. If that's what "lifted the Linux boat", I'm incredibly happy.

A blade of grass in my yard has an upward trend too.

This is the year Linux makes it big on the desktop.

Stop with this horrible trope please :(

Wait, this year? Are you sure? I thought the year of the linux desktop was like 20 years ago? And then 19 years ago, then 18, then 17...

I personally still think that Valve made a huge mistake by basing SteamOS on top of Debian. I think they should have adopted Android as a desktop environment per se (like Keepod [1] for example).

This gives them access to the same developer ecosystem (tools and documentation) that Android games has already established, the same OpenGL-ES framework that Android uses.

Additionally, Google has demonstrated Android-L applications running on the Chromium runtime and have access to the underlying hardware.

Using something like Docker, I could have been running Android-SteamOS on my laptop and still been able to get RoR+Postgresql working.

I think fundamentally, the huge fragmentation in the desktop linux world (cue the Systemd Wars) is hard to work around. But Android already has a single driver (Google) and Steam could have built a viable desktop Android based SteamOS.

[1] http://keepod.org/

Android is hugely fragmented as well, and anyway Valve uses Steam runtime to go around the issue of distro fragmentation. I am running Steam on several different distros at home without any issue at all.

Using Android as a desktop environment is a terrible idea, because you'll only get mobile GPU chips to work with that kind of software stack, and they are massively underpowered vs the graphics cards you have on PC. That's only a good idea if you want to play stuff like Angry birds on your TV, but any serious gamer would not want that.

> the huge fragmentation in the desktop linux world (cue the Systemd Wars)

With Ubuntu adopting systemd (http://www.markshuttleworth.com/archives/1316), I guess the init system wars are basically over.

Completely disagree, and I'm really not sure why you're characterizing their decision to base on Debian/Linux as a "mistake" (it's going very well so far).

>I think they should have adopted Android as a desktop environment per se (like Keepod [1] for example).

Android is not a desktop environment. Just as there are numerous desktop environments for Debian/Linux, so too are there numerous desktop environments for Android. There is no one clear standard on either platform. Valve wouldn't escape fragmentation by using Android; they'd merely embrace another kind.

>Using something like Docker, I could have been running Android-SteamOS on my laptop and still been able to get RoR+Postgresql working.

You can do precisely the same thing with Debian/Linux. In fact, that's precisely what it was designed for.

>the huge fragmentation in the desktop linux world (cue the Systemd Wars) is hard to work around

The systemd wars are over. Systemd definitively won, full-stop. Besides, what on earth does that have to do with Steam? Why should it need to have any insight into that level of abstraction?

> But Android already has a single driver (Google)

This isn't an advantage; it means that they have less flexibility. Why do you think Google itself chose to base Chrome OS on Gentoo Linux, instead of working with someone like RedHat? Flexibility and freedom.

> Steam could have built a viable desktop Android based SteamOS.

They already did — on Debian. Exactly what makes you think they regret their decision? Seems to be going very well, so far.

First, Arch has not adopted systemd.

Second, fragmentation in Linux is different than fragmentation in android - the latter is about obsolete versions,while the former is about heterogenous operating system stacks.

Third, I don't understand why people believe that the Android os will force you to use mobile chips. If valve would have adopted android,it could very much have brought on nvidia and ati to port to android (I don't know if it would have been easier).

If the systemd wars have taught us anything ,then it is that flexibility is not always a good thing - which is why the CTTE voted to keep only one init system,rather than offer a choice. I do respect the fact that it is important for you.

Fundamentally ,I think Linux gaming would be much better married to Android because of the sane and coherent graphics APIs as compared to desktop Linux. Read Jonathan Blow's frustrations stemming from his inability to port Braid to Linux http://www.develop-online.net/news/jonathan-blow-criticises-...

I think a lot of people are resistant to the idea of having a touch centric interface on the desktop whenever I talk about desktop Android. But that is so untrue - you can build beautiful launchers for Android that replicate and improve upon desktop interfaces - Android L is bringing some of them bybrunning android apps in the desktop.

Lastly,why am I downvoted ? You may not agree with me,but it is not as if I'm trolling.

Arch was one of the earliest distros outside of Red Hat to adopt systemd. It's been using it for nearly two years.

Ah,my mistake - I was thinking of OpenRC. It is Gentoo rather than Arch,but I hope the example still holds.

I completely disagree. They've already (mostly) solved the fragmentation issue by bundling a set of libraries with Steam that roughly resemble Ubuntu 12.04, and I've run Steam on just about every distro out there and had good experiences.

The desktop gaming industry doesn't want a pre-existing developer ecosystem like Android has. It wants something like it already has on Windows, and more and more game engines are supporting OSX/Linux as build targets.

I just can't imagine playing Dota2 on Android and have it perform better than on Windows (which is currently the case for Linux), and I can't imagine porting it to Android would be easier than porting it to Linux.

Why would fragmentation in the broader Linux ecosystem affect the narrower SteamOS platform? Valve can safely ignore Linux fragmentation.

Reply on news.ycombinator.com