Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Graphics Programming Games Build Linux

Unity 4 Adds Linux Support 150

dartttt writes "After more than 14,000 votes by Linux users and efforts by Brian Fargo, Unity has added Linux support to their popular 3D game engine. Starting with Unity 4.0, Linux will be supported as a publishing platform allowing Unity games to be played natively on Linux. Only standalone desktop games will be supported initially. From the article: 'Unity Technologies, maker of a widely used video-game engine, today announced that its fourth-generation product will introduce new animation technology and extend its support for Adobe Systems' Flash Player, Linux, and Microsoft's DirectX 11.'"
This discussion has been archived. No new comments can be posted.

Unity 4 Adds Linux Support

Comments Filter:
  • No source? (Score:5, Funny)

    by Hatta ( 162192 ) on Monday June 18, 2012 @02:32PM (#40362659) Journal

    How about a kickstarter to liberate the source of Unity?

    • Re:No source? (Score:5, Informative)

      by Sir_Sri ( 199544 ) on Monday June 18, 2012 @02:55PM (#40362949)

      Um.... do you have any concept of what that would cost? You'd have to be offering a huge pile of money. Right now they can commercially licence their engine for all sorts of projects. Even if those projects don't make money Unity can.

      Have a look at their people page, they have probably 110 employees. That's probably 12-13 million a year in revenue alone. Are you going to try and get a kickstarter for 100 million dollars to effectively shut them down, or to guarantee them income to keep working indefinitely?

      Don't get me wrong, there need to be more open source game tools (no matter how many you point me to there can always be more). As someone on the teaching side of things in trying to train game developers it's a real problem to know what tools you want to use, because the emphasis shouldn't be on the tools, but fighting with tools puts the emphasis on them. But Unity is pretty good about giving away a free trial, and being a good example of the sort of experience you'll have in industry, with some stuff opened up to you. That's about all we can hope for. Asking for a commercial engine that costs millions of dollars to make and maintain to just give up that kind of money is a pipe dream at best.

      Now, trying to get them to pull an id software and release old versions of the engine as open source (say release 2.0 or 3.0 when 4.0 goes live) might be a more realistic goal and would still be awesome.

      And by the way, you can negotiate your way into source code for Unity3D. I've never worked with anyone that thought it important enough to try until today, though. I literally advised a company this morning that Unity is probably their best bet for an engine given what they want to do, and they were wondering about source licences, which is the only reason I know that at all. Given that, it wouldn't be a huge shock to see old versions end up open sourced, if nothing else because you can't keep something bottled up indefinitely.

      • It makes perfect sense for Valve to buy and opensource Unity to get more games for their upcoming Linux-based game console.
        • by Sir_Sri ( 199544 )

          How would that make them money? The steam for linux store is going to be about selling software, not giving it away for free. What do you need steam for at all if you're giving it away for free? They aren't in the business of running a pile of servers for charity.

          • I think the parent's point is that Unity is a good tool and if Valve bought at and released the *tool* the result would be more (presumably payware) games that run on Steam.
            • Or they could not buy it, and let people do what they're doing already, and still have the potential increase in payware games on Linux...

  • now devs and companies just have to click a few buttons and we got some games... hopefully
    • by Zrako ( 1306145 )

      now devs and companies just have to click a few buttons and we got some games... hopefully

      After you pay for the second copy :(

    • by Anonymous Coward

      I've done one extended game jam with unity now. It started as a weekend project which was polished up afterwards. So about 16 hours of coding and design and another 24 odd hours of polishing between two people. I'd show you the game but the flash version is being auctioned off right now. Sorry. :)

      Overall I would say that it was remarkably easy. Though I would not attribute that to just being able to push buttons. There was a non significant amount of coding involved. And by that I mean, of the total of 80 h

      • Yeah, it continues to bug me that MonoDevelop lags so far behind it's Windows origin, SharpDevelop, when one of the selling points of .NET is cross-platform code - surely you should just be able to build SharpDevelop for Linux...

  • Fuck yeah! (Score:2, Funny)

    by Anonymous Coward

    OMG! OMG! OMG! More proprietary software is coming to Linux!!! Fuck yeah!!!

    • Re:Fuck yeah! (Score:5, Insightful)

      by SJHillman ( 1966756 ) on Monday June 18, 2012 @02:39PM (#40362755)

      Given the choice between having a proprietary option and having no options, I'll take the option to have proprietary software available every time.

      • by Lunix Nutcase ( 1092239 ) on Monday June 18, 2012 @02:45PM (#40362811)

        You just made baby RMS cry.

        • Re:Fuck yeah! (Score:4, Insightful)

          by Microlith ( 54737 ) on Monday June 18, 2012 @02:53PM (#40362921)

          Well then he can weep, despite it moving one step closer to his goal. Actually, didn't he acknowledge that proprietary software on a Free platform was better than proprietary software on a proprietary platform?

          • Re: (Score:3, Funny)

            by ArsonSmith ( 13997 )

            A hand job and a kick in the balls is better than two kicks in the balls.

          • Quick Internet search nets me a wikiquote page for him: (http://en.wikiquote.org/wiki/Richard_Stallman)

            Some GNU/Linux operating system distributions add proprietary packages to the basic free system, and they invite users to consider this an advantage, rather than a step backwards from freedom.

            Whether this is before or after something else he may have said is unknown.

            • I believe Microlith was referring to RMS's statement linked from this story [slashdot.org], which might be taken to mean that non-free games on a free platform are the lesser of two evils: "At least you avoid the harm to your freedom that Windows would do." He goes on to say something to the effect "instead of actually playing games, support these vaporware projects".
          • by unixisc ( 2429386 ) on Tuesday June 19, 2012 @02:29AM (#40367153)

            Yeah, he very specifically did say that in one of his recent essays. He doesn't like this unliberated software, but he'd rather that it sleeps w/ his liberated OS, rather than an unliberated one.

            I actually tend to believe that the opposite choices are the way to go. Had the FOSS movements (not talking about just the FSF, but everybody involved in having source code automatically available w/ binaries) actually started w/ useful apps and making those liberated or open-sourced - things like Office Suites, Image & video editing software, Publishing software, financial software like and so on - that would have been better for FOSS as a whole. People would have gotten used to the likes of Open Office, GIMP, VLC and so on ages ago, just like they're used to Firefox and Chrome, and those would have been ubiquitous on computers. This would also have given these platforms the opportunity to get feature rich and customizable, letting people install either just the features they need, or all the bells & whistles.

            Once that was out there, it would have been relatively easier to migrate them to FOSS OSs, be it Linux, BSD, osFree, ReactOS, et al. The initial port may have been a bitch - all those API translations and so on - but once that was done & out of the way, making upgrades to say, Linux versions of FOSS titles would have accompanied the upgrades to Windows equivalents (incidentally, while on that subject, such software should not have to be re-written b/w different versions of glibc or GCC or GTK or Qt - once it's written in each library, it should automatically be supported by its successors). Only caveat I see - the business models behind these would have needed to be worked out, but aside from that, it would have ensured a much wider acceptance of FOSS. In other words, if these programs need to be sold, do it, so that the projects don't remain in the red.

            In short, what keeps FOSS from being widely embraced is its focus on lower layer s/w like kernels and userland utilities, rather than actual programs that end-users need. Stop making 20 text editors, 10 music players (KDE, I'm looking @ YOU), and so on, and actually produce the type of software that people need - be it things like Quickbooks, Photoshop & so on (close the gaps b/w GNUcash & Quickbooks, GIMP & Photoshop, Calligra vs MS Office and so on). Once those are successful, it will be easier to talk people into installing BSD or Linux or other FOSS OSs, since these titles can be ported there, given the availability of the source.

            • Had the FOSS movements (not talking about just the FSF, but everybody involved in having source code automatically available w/ binaries) actually started w/ useful apps and making those liberated or open-sourced - things like Office Suites,

              emacs, vi, vim, latex

              Image & video editing software

              Ok, GIMP lagged Photoshop to market by six years. However, it's worth noting Photoshop was the first in its class.

              Video editing? PCs really weren't up to snuff. I remember having to run an MPEG decoder in grayscale mode under Windows 3.1 just because decoding chroma made it run too slow.

              Publishing software

              LaTeX...which was (and is) a WYSIWYM editor similar in some respects to WordPerfect for DOS (remember that?), but was really a user-friendlier means of working with TeX, the standard bearer for publishin

              • emacs, vi weren't the type of apps I was talking about. I was talking about things like Libre-Office. In the KDE suite, I've noticed a lot of applets, but very often, a lot of them have duplicated fuctions, such as - in editors itself - Kate, Kedit, KJots, KWrite. Or, for multimedia, KPlayer, KMPlayer, Kaffeine, Dragon Player. Incidentally, for video editors, I was thinking something that's the equivalent of Microsoft Movie Maker, but the only ones I found were Cinerella and Avidemux, and they were too

                • I was talking about things like Libre-Office

                  So I did a little digging. It looks like the first WYSIWYG functionality in personal computer appeared with the Apple Lisa in 1983, and WYSIWYG quickly made its way into the GUI releases of Word and WordPerfect around 1985. Still, this is 1985 we're talking about. You're lamenting the lack of consumer-oriented open-source desktop applications back at a time when any commercial enterprise trying to build and sell such a product was taking a huge risk. The reason Microsoft's mission statement was "a computer

    • Re:Fuck yeah! (Score:5, Insightful)

      by oakgrove ( 845019 ) on Monday June 18, 2012 @02:42PM (#40362785)
      If you want Linux to ever be a remotely viable third choice on the desktop then proprietary software will be a given. If Android eschewed the presence of closed source apps it would be a non starter. Personally I don't have a problem with proprietary applications as long as the underlying operating system is Free. If the open source community isn't delivering for a particular niche then let the closed source people step in and first class games is a perfect example of this dynamic. If the community then steps up like has happened in so many other areas of software then even better but until that happens, I'll happily use unity on my Linux box right alongside my nvidia driver.
      • I'll happily use unity on my Linux box right alongside my nvidia driver.

        Not me; Unity sucks and Shuttleworth can go shove it... oh, you mean this other Unity.

      • There's nothing really keeping anyone from developing and offering proprietary software for Linux. However, I do think one of the bigger impediments to it is the lack of standardization of Linux systems: should you offer .deb packages, .rpm packages, Slackware-style .tgz packages, or what? Instead, most proprietary software makers seem to end up making some horrible, nasty custom installer (usually some giant file named something.run) which installs stuff wherever it wants and doesn't integrate with the s

        • I think if publishers would release their software as an RPM package that conforms to the LSB standard and have all assets/libraries/binaries go into /opt with the icon in /usr/share/icons and the .desktop file in /usr/share/applications and make sure to bundle up all libraries necessary for the application to run then the problem would be solved. That's just my take.
          • No, that wouldn't solve anything, because not all distros use RPM or follow the LSB "standard"; really, only Red Hat does. You can't exactly make something that's "Linux compatible" and it doesn't work out-of-the-box on Ubuntu, Debian, or its other derivatives.

            Bundling all the libraries needed as you say, or even statically linking them, however, would solve a lot of compatibility problems, even though it'd increase the size. Relying on other libraries to be installed is usually a recipe for disaster; eit

            • As long as something consistent is released, it will be repackaged with a simple wrapper (if existing conversion process implimented in "alien" utility won't do it already).

              The issue with packaging is completely bogus.

              • So you only want Linux to be usable by experts?

                Asking regular users to use alien is ridiculous.

                • That's what distribution maintainers do, not users.

                  Are you really that stupid, or is it your job?

                  • You seem to be a very stupid person. In case you didn't notice, this discussion is about proprietary software. By definition, proprietary software is NOT included in a Linux distribution, which consists of only Free (freely-distributable) software. The issue brought up 6 levels up by "oakgrove" is that proprietary software has to be usable on a Linux system to make it viable as a desktop system, like it or not. Since distros have zero control over third-party proprietary software, a Linux system would n

                    • You seem to be a very stupid person. In case you didn't notice, this discussion is about proprietary software. By definition, proprietary software is NOT included in a Linux distribution, which consists of only Free (freely-distributable) software.

                      There is plenty of proprietary software with packages maintained by distributions -- the package is a wrapper over whatever the software vendor distributed. If possible, package downloads the file automatically, or (thanks, Oracle, for your idiotic policy) it asks the user to get and supply the file. Most proprietary software requires separate license configuration after binaries are installed, or software has to "call home" to download associated data -- those operations have nothing to do with packaging,

                    • What package ? In what package manager?

                      Don't pretend to be stupid.

                      No sane commercial software developer should or will every give up control of distribution of their software to a bunch of F/OSS zealots.

                      Microsoft Core Fonts and all proprietary graphics drivers, among other things. Every piece of Microsoft software that is installed by winetricks.

                      Software vendors now produce installers for Windows -- those installers are given to the users, and run on whatever insane installation of Windows those users have. Following your logic, that could not possibly happen because vendors would not want users to run their installers.

                      In reality, unless vendor specifically disallows any external

  • by oakgrove ( 845019 ) on Monday June 18, 2012 @02:34PM (#40362693)
    Now that the engine is ported, how much additional effort is required by the developer to make their game run on Linux? A lot? A little? I'm readily curious.
    • by gman003 ( 1693318 ) on Monday June 18, 2012 @02:46PM (#40362819)

      Depends. I've never used Unity, but I have used UnrealEngine, Source and idTech, and I've done some light reading on it before.

      The most common scenario will probably be "needs some shaders re-written to work with Linux's outdated drivers", assuming, of course, that they'd already written GL shaders (and not just D3D). Best-case, all they need to do is check the "Export for Linux" box right next to the "Export for Android" and "Export for XBLA" boxes.

      However, it should be *possible* to make a Unity game that requires a ton of work to port. Either because you actively tried, or because you didn't use the engine to it's full potential and instead re-implemented half the functionality in system-specific ways. Think of Android - you *can* write native apps that don't run on non-ARM (or even only specific ARM) processors, but that's not exactly common.

      Of course, engine support historically hasn't translated into game support. UnrealEngine 2 supported Linux (think 3 does as well), as did several idTechs (even before being open-sourced), and yet we only rarely see games using those released for Linux. Although it may be a matter of how *good* the Linux support is - many of those may have required far more work than more modern engines.

      • by evil_Tak ( 964978 ) on Monday June 18, 2012 @03:04PM (#40363047)

        Actually, Unity's surface shaders are more or less 3d-system-agnostic. Some features will of course degrade when the underlying system doesn't support them, and some, although supported, will be too intense for the hardware (e.g. fog on mobiles).

        It is of course possible to create a platform-dependent game: in fact, it's as easy as File.ReadAllText("C:/Windows/blah").
        However, the majority of real content that has been tried has run out of the box with no major issues.

    • by Jeng ( 926980 )

      They mention that this is coming straight out of their labs so I would guess a lot of effort will be needed.

      No ship date at this time so there is still time for them to drop that feature.

      • Considering Wasteland 2 was promised to Linux users that backed it... and that they were working with the Unity team to permit Linux support... I'd say it's probably a fair conclusion to say that it will ship. If they decided not to ship it, there would be quite a few upset people on the whole Kickstarter process, inExile, and everyone involved.

    • by tigeba ( 208671 )

      When you are talking about switching between "like" platforms, for example Windows Standalone -vs- OSX Standalone -vs- Soon-to-be-Linux standalone the changes can be very minimal or almost nothing. My experience with the Windows/OSX standalone builds is that you can sometimes deploy with zero changes. The most common issues that seem to crop up are related to custom shaders.

      I maintain a bunch of games and demos that we use as examples for our networking middleware, and they basically never need platform cus

      • I'd second the minimal or almost nothing. I tend to target Android (and am extremely excited to target Linux, as we may move some embedded stuff to it) and have 0 problems targeting Win/Mac desktop or even web for testing. With a different screen aspect ratio I may have to tweak things a little (as we have stuff set up for the android tablet aspect) but other than that stuff just works.
    • by arth1 ( 260657 )

      Now that the engine is ported, how much additional effort is required by the developer to make their game run on Linux? A lot? A little? I'm readily curious.

      That depends on whether it really is native or just another Wine port. The latter isn't much effort, but also isn't Linux.

      • by slart42 ( 694765 )

        Unity on Linux will be a native Linux binary. For games designed to run on desktop systems, the porting "effort" should be no more then clicking a button -- unless you use custom C++ libraries. All the game logic in Unity is implemented in mono bytecode, and the graphics already use OpenGL on OSX (you generally don't need to write platform specific shaders in Unity, as shaders are cross compiled to the target platform).

        • by arth1 ( 260657 )

          Porting is a lot more than graphics. If you can change to Linux audio/keyboard/mouse support or screen/viewport configurations by the click of a button, I want that button.

          • by slart42 ( 694765 )

            Porting is a lot more than graphics. If you can change to Linux audio/keyboard/mouse support or screen/viewport configurations by the click of a button, I want that button.

            Get Unity 4 when it comes out, and you will have that button. Unity abstracts all of what you listed, so you won't have to deal with any of it when developing games in Unity.

  • cool more game for Linux! it seems that some companies in the game industry take notice of Linux and it's market
    • cool more game for Linux! it seems that some companies in the game industry take notice of Linux and it's market

      Was that an accidental typo or very clever satire?

    • by robmv ( 855035 ) on Monday June 18, 2012 @03:36PM (#40363349)

      I think this trend is being forced by proprietary OS vendors, Valve should be threatened by the Windows 8 Market that is locked (for Metro applications only, for now) and the prospect of a locked down OS X. If that future of entirely lock down stores arrive, Steam will be dead soon. That is the only reason they are looking for an exit on the Linux market.

      • Valve doesn't have any power and can't force game companies to port their games to Linux
      • It would be especially funny if Linux got a boost from software producers being worried about platform lock-in rather than users, but I think these businesses can see the future a bit more clearly than the average user who seems to think walled gardens are wonderful things. Of course, the users don't see where that extra 30% of their money is going.

  • by Anonymous Coward on Monday June 18, 2012 @02:37PM (#40362717)

    Pics or GTFO!

    • Pics or GTFO!

      It said nothing of the GUI working on Linux, only that the games would.

      • by Anonymous Coward

        To clarify: The Unity Editor (the application you develop games with) will not be supported on Linux with this release.
        The engine's GUI system (the feature you use to make in-game GUIs) will, of course, be supported on Linux.

  • this could encourage others to support linux with their engines
  • Since games like shadow gun run on Android and make use if the unity engine, want unity essentially 'ported' to Linux anyway? I mean I'm sure it is native code on Android and not just Java.
    • It's not the kernel that matters. It's the userspace part. So, no, it wasn't ported to X11 desktop Linux anyway.

    • The problem are the APIs used on Android generally don't line up to anything you normally find on non-Android Linux platforms. The entire JNI setup, for instance, isn't necessary on non-Andorid platforms. Instead you need to work through Xorg (X11) and deal with the standard *nix user space and not whatever Android supplies instead.

    • Re:Question (Score:5, Informative)

      by gman003 ( 1693318 ) on Monday June 18, 2012 @03:01PM (#40363017)

      It's not the kernel, it's the libraries. That's really all a game engine does - it takes all the libraries and presents a simple interface to them, while integrating the asset tools (ie. model file formats, etc.)

      On Windows, almost everything you need is in DirectX. Same for the XBox - it's pretty much the same library. Graphics, audio, networking, input, it's all there except a basic AI library and physics simulation.

      On OS X, there's a bunch of less integrated APIs. OpenGL for the graphics, some proprietary library for input, and so on. iOS uses mostly the same libraries.

      Android also uses OpenGL, but has it's own, different libraries for pretty much everything else. The same is true for the non-Microsoft consoles - either OpenGL or the OpenGL ES, and custom proprietary crap for everything else.

      Linux, again, uses OpenGL. But that's about it as far as "common code". Want to tell if Mouse3 has been pressed? Need new code. Want to play a sound? New code.

      Now, it's not quite as bad as it seems - most of the engine is, in fact, the "turn basic libraries into something that does all the work for you", and the renderer *is* the biggest library bit, but it's still quite a bit of work to go from Android to Linux.

      • Just to be clear there is a LOT that goes on under the hood beyond DX/OpenGL calls and AI.

        A substantial part of an engine involves efficiently managing all of that data so only subsets of the data are getting streamed/updated at any given time.

        • Yeah, I guess I did sort of misrepresent that. I was a bit rushed at the end there, boss was looking over my shoulder.

          For an analogy /. would understand, think of a game engine as Webkit, except in full 3D, rendering millions of elements every frame (and if you EVER dip below 30FPS there will be hell to pay). Oh, and half the elements have to think for themselves, and you have to run on everything more powerful than a wristwatch, and you have to have an integrated IDE that handles large-scale level geometry

  • by Anonymous Coward on Monday June 18, 2012 @02:51PM (#40362877)

    I thought this was another "gnome 3 is horrible" post. It's so horrible that after a few releases it NOW supports Linux.

  • I knew it!! (Score:3, Funny)

    by rikasa ( 1760376 ) on Monday June 18, 2012 @04:48PM (#40364053)
    I knew Linux could not be responsible for such a Desktop travesty! I wonder how much Shuttleworth was receiving to take the rap??
  • We run Unity for an embedded gaming solution (electronic gaming, think slots and similar machines). When I approached Unity at G2E they said they had it working on Linux (I suspect IGT or another big player in class 3 asked for it) but it wasn't released to the general masses. Linux is so much easier to manage than the alternatives and is much more cost effective.

    This also means you can run Unity games on the RaspberryPi or similar. Can't wait to make myself a Unity arcade cabinet based off the Pi or Riko
    • by Hatta ( 162192 )

      This also means you can run Unity games on the RaspberryPi or similar.

      When was the last time a closed source proprietary applicaiton was released for ARM Linux? You will get the architectures they deem profitable to support, no more, no less.

  • Judging from the number of the "This had better run on Linux!" comments on Neal Stephenson's sword game Kickstarter campaign [kickstarter.com] (they are likely to use Unity for CLANG), this should make some people happy.
  • I contributed to the Wasteland 2 kickstarter, and it was said all along that if/when it reached X dollars there would also be a Mac and Linux version. It easily passed that amount. In one of the follow-up newsletters, they stated that they had reviewed a number of game engines and chosen Unity, one of the reasons being that they had been working on porting it to Linux anyway and that there had been a substantial amount of work done already to that end, before being chosen for Wasteland 2. That and the devel

"...a most excellent barbarian ... Genghis Kahn!" -- _Bill And Ted's Excellent Adventure_

Working...