The Opinion Slab: OpenConsole?

Welcome one and all for this weeks block…no chunk…nay, slab of opinion. As ever, everyone is welcome and encouraged to scribe their opinions into the comments on this post. Yep, even you oddbals with your wild staring eyes and can of Special Brew in tow are welcome to join in. Lets go…

In recent years the gaming world has become bigger, more impressive and more diverse, but all within a pretty restrictive set of metrics – if you want to develop for PS3, Wii or XBox you need licenses, special hardware and there is little opportunity to write homebrew games on these systems. Despite utterances from some of the vendors about encouraging some degree of homebrew, the situation is pretty drab. Then mix in the complexity of writing games for PCs and specifically Linux with its array of potential dependencies. Could we instead make our own open games console standard? As in, could we define a machine spec (a chosen set of components), a set of software tools, a development stack and consider this the OpenConsole standard? This could provide a specific target platform, a target set of tools, and potentially allow OEM vendors to release hardware based around said spec. There were similar attempts with the multimedia specification in the late nineties and it died on its arse. Could it work here? Could this be our opportunity to define an open games console standard and get our grubby feet into the gaming ring?

My take: I think it could work. Sure, there are challenges, and there would be some tough decisions (which graphics cards, which processors, which software etc), but the aim of this exercise would be to make tough decisions, define a platform and then build a strong community around that platform of players and importantly, developers. The platform could potentially offer a predictable gaming experience (which is always a benefit of consoles) and a specific set of development processes to encourage games production. There is no reason why the development platform could not encourage choice (support both C/C++/Java and Python as an example) but have the benefit of consistency and a predicable and known set of resources and abilities.

So what do you think? Share your views in the comments.

  • http://www.qdh.org.uk Karl Lattimer

    There’s already a pretty kick arse graphics engine http://www.ogre3d.org/ Quake3 is now open source under the GPL so the reality isn’t the lack of SDKs and APIs

    It is in fact the lack of good 3D artists, plot authors etc… that prevent open games from being made in the first place. World of Padman is a great example of what can be done by a few graphic artists and the quake3 engine.

    If an open platform for game development was introduced would it restrict the developers to open operating systems? I see that as a really bad thing, most gamers are windows users because of the fact that most game houses create games for windows, these gamers are regularly impressed with the loading times and FPS differences on Linux, the best example I’ve seen of this is UT2004.

    Shifting these users from windows to linux isn’t difficult once they see the benefits, but they’re always going to be dual booters.

    As for the idea of an OpenConsole standard, I think in this market its a pretty ludicrous idea. It is massively costly to develop games console hardware, especially competitive hardware. Sure, an open standard for a fairly low level abstraction of any hardware to a standard SDK is the only part of this idea I think /could/ work, but you’d need to get nintendo, sony and mickysoft on board in order to make it work in the real world. Then developing and porting games becomes much easier, possibly even as easy as a rebuild and that would REALLY open up the gaming market, if an abstraction layer was introduced to make all the available platforms essentially the same, and the big players got on board with it then after a year or two an OpenConsole could be designed built and marketed to developers and gamers alike.

    Worst thing about the whole situation at the minute is, the 3 players have already released their next-gen consoles, you have 5-10 years to wait until they start shipping their next-next-gen, so this is an ideal time to get started on something like this (as it probably is 5-10 years work to get it ready for a developer market).

  • DeadGuy

    :???: I could flip flop on this one. I could and would love to see a console that could have the diversity of home brew games but then there is the problem that all the competition would be multi billion dollar corporations. I would support it if say a large organization that i supported was the main backer.

  • jono

    Karl – I think you are maybe mis-understanding my point. I am not talking about building hardware or drawing up a standard with vendors – I am talking about the Open Source community deciding on a hardware and software configuration and calling it OpenConsole. As such, anyone who purchases a machine that meets that standard has an OpenConsole.

  • McLurker

    Don’t console manufacturers sell the box at below cost, and get the money back from selling licenses? This is almost the direct opposite of Open Source business models, where you give away software in order to sell other stuff. So, how could you make a console that isn’t far more expensive than a PS3?

    One possibility would be to get money from subscriptions to online games, I guess. But would there be enough money coming in to allow the Open Console Project to subsidise consoles? Or do we all love freedom enough to pay £100 or more extra for our consoles?

  • Luke Benstead

    One thing that frustrates me about the games industry is its reliance on DirectX. Using OpenGL and SDL (along with other APIs) you can develop cross-platform just as easily and reach a wider audience. I have even started a project on google code to try and provide some of the functionality of the D3DX library to OpenGL programmers (search for SheepLib) just to get people to switch.

    Although I see where you are going with the Open Console idea, I think it would be much better to collect a standard toolkit (equivalent to the whole of directX) of APIs and cross-platform tools for modeling, texturing and world creation rather than to specify certain hardware.

    For example the Open Console toolkit could include: – SDL – OpenAL – HawkNL or SDL_net or the SDL Net2 high level api – ODE or Newton Game Dyamics – CEGUI or similar – Some kind of world editor like Deled or Getic 3D (but open source) – GLSL editor (like lumina on SF)

    Then document all of it, write tutorials using it, then create somewhere for people to put their games written in it.

    Games are one of the main reasons people run Windows, making Linux the most appealing development environment for games developers will draw a load of people over. That’s my belief anyway.

  • http://mighmos.org Adam Petaccia

    As for comments regarding restricting this to open platforms, why not make the game a liveDVD based on Linux/BSD/whatever? It would be installable, but for the Windows, Mac, and whatever users, it would still be playable like consoles today? I think that is feasible.

  • http://loz.loz-n-ali.com/ loz

    I think this is an awesome idea. hardware is not a problem, if it went with x86 and some well supported linux hardware. Open 3d drivers are obviously an issue, maybe if AMD do the right thing here, there’s an opportunity. But basically the technology is there.

    I see this as being something that people could easily get into and behind. It’s a bit like many of the open source media centres. A console platform which would work in a similar way, live cds etc.. would be great.

    There’d be no shortage of testers for this, many linux users would get behind something like this straight away. If you build it, they will come. The industry doesn’t need to get behind it necessarily at the start. The console companies (Sony, MS, Nintendo) don’t have to get behind it either. Just the developers. Part of the company I work for is a games company, and the costs of developing on platforms is daft, the hardware costs are part of this. Smaller companies could have a crack at games without massive hardware investment costs. Look at the likes of innovotech, games like darwinia and uplink could come out, and be really affordable.

    People are willing to pay for games if there’s quality, this platform could make affordable varied playable games like the spectrum and amiga days a reality again..

  • Jerome

    Too bad the Indrema died…it could have been a good starting point for Open Console.

  • Rob J. Caskey

    Hardware wise, what differentiates a console from a desktop:

    Price: Consoles are usually cheaper, even if it is only because they are subsidized.

    Form-Factor, although desktops are coming closer and closer to consoles and consoles are coming closer and closer to desktops. Laptops are, in most cases, superior in form-factor.

    Testing – Console hardware and software is very very well tested. All hardware plugs and plays perfectly. It is known to be fast enough to run the software people want.

    Items 1-2 are not anything FOSS can do anything about. But #3, #3 is golden. Who knows whether hardware works reliably with feature X? We don’t need certs, we need the Hardware DB extended, user-friendlyified, with individual results released under an appropriate license.

  • http://gepatino.blogspot.com Gabriel Patiño

    Great idea, mostly because I had the same idea this weekend ;)

    One benefit, if you work in an open source fashion, would be that people could make their own consoles, as long as they buy the recommended components. This is a great issue for people living in countries were technology takes some months (or even years) to land.

    What I was thinking of this weekend wasn’t too complicated: – The games are burned on CDs/DVDs with all needed libraries inside the CD (forget about dependencies) – When you insert the CD/DVD, the console starts playing it, as any other console, but you could also copy an image to a local disk. This way you could run the games from the hard disk later, without using the media. This could be usefull for roaming gamers (?) who doesn’t want to take a bag with disks.

    The main problem here is how to attract big game publishers to this arena. It could even reduce publishing costs since you could download images from their sites (of course, paying for the downloads) so they don’t need to do any extra packaging, etc, etc.

    If you advance in this topic, please, keep us informed.

    Cheers form Argentina.

  • Jerome Haltom (wasabi)

    I don’t really see anything concrete in your proposal. The idea of establishing a “hardware platform”, by it’s very nature precludes systems that do not fit into that hardware platform. If we were to make LinuxGamePlatform1.0, somebody would disapprove and make YALinuxGamePlatform1.0. That’s just the way it works without centralization.

    If you add centralization to the mix, you end up with a single hardware vendor.

    Now, the first approach really is fine. If somebody wants to write a game that will run on any platform easily, it is not that hard. OpenGL, SDL, whatever. A higher level engine? Ogre, Axiom, CrystalSpace. Quake3. We have all of these.

    The real issue is nobody wants to make games for them. Because making games takes a shit load of time. And people want to be paid for thei time. And then they’ve already admitted their goal is money, then maximizing the intake of money becomes the reason for the endeavor. And then you end up using Direct3D because quite simply, it is easier, and lets you get more money faster while paying less for less skilled people. That’s why it’s so successful. Not that it’s technically better, but that you can find more people who know it. More people equates to lower salaries. Just look at XNA.

    And so the cycle continues.

    What we really need quite simply is industry lobbying. We have to get our software: OpenGL/SDL/Whatever into the hands of the people who want to make the games. It has huge potential.

    Will you see free/open source games on release anytime soon? Doubt it. Whether they program for Linux or for Windows, they need to pay artists/programmers.

  • http://www.qdh.org.uk Karl Lattimer

    Jono: Oh, oops, I thought you meant something like OIN where the corps come together to a solid aim… In that case, its a pretty similar kind of set of requirements as MythTV or Freevo.

    Game developers, content designers are all that’s really needed to get ‘something’ off the ground. They’ll end up defining the hardware requirements during game development, if there is an open standard for how games are launched, and some way of defining joystick configurations etc… The hardware stuff should be pretty easy, some kind of base UI / firmware interface possibly built ontop of Myth or Freevo could be an excellent starting block. A metapackage which has all of the OpenConsole standard libs and tools would carry the can for the certainty that the systems meet the required spec.

    Sounds much more interesting than what I initially thought :)

  • http://www.pobox.com/~meta/ mathew murphy

    If you shipped the OpenConsole with a bunch of emulators (just not the ROMs, obviously) and managed to give it a sub-Wii price point, you could probably sell enough to give you a reasonable install base.

  • Imran Chaudhry

    Interesting and promising idea Jono !

    Some advantages of an OpenConsole platform:

    • would be great for trying out concepts that are too risky to be tried in a commercial sense.

    • re-ignite the homebrew games scene we last saw with the Amiga/Atari ST?

    • having a set platform with limits encourages designers to really squeeze the most out of the hardware

    The concept reminds me of the 3DO of the mid-90s led by Trip Hawkins. Long-time Edge magazine readers may recall it. That was a console standard licensed out to Panasonic, JVC etc. Sadly, it flopped due to lack of titles but I do remember a decent Super Street Fighter II port for the thing! I saw one at a carboot sale the other day.

  • mika

    No, it would not work. You see, it’s not a technical question at all.

    The video games are like movie productions nowadays, in almost every sense. Huge money and loathsome projects. What they require are standard platforms. When you for instance create a game for XBOX360 the same runs practically instantly on Windows as well. Or vice versa. Although this would be “technically possible” using open source toolkits the Q&A problems are risks are too vast.

    Support is an other issue. DirectX family for instance is extremely well documented by the vendor, and it also comes with ready RAD tools and you can find skillful workers for it easily. Also you can find a lot of courses, 3rd party modules and such. It’s a whole ecosystem of support and tested references of abilities, and is in overall critical of importance.

    Also, there are people that know the current environments. Introducing new ones is always risky and there will be less interest and knowledge about it freely available. The new DirectX/(whatever the PS uses?)/(Whatever the Wii uses?) versions will be a lot like the previous ones.. They are pretty nice to move into.

  • mythic

    I had thought of something like this a long time ago. With all the emulators out there for consoles i got to wondering why someone doesn’t make a new platform specifically for emulation. The platform itself would be nothing more than a list of api’s, and an open sourced application that works like an emulator or virtual machine. That application could then be ported to work on any operating system or console. Obviously this system couldn’t be used for the most high-end hardware intense games, but i’m sure it would be great for the market nintendo has been going after for the past console generation.

  • peterd

    I think the hardware platform could remain quite elastic: x86 processor of a certain speed or higher X ammount of ram, or more a hard drive usb ports a 3d compatible video card with linux support tv out a network connection possibly tv out maybe even bluetooth… You would just need a minimum required spec.

    A platform that would support linux bios would be preferable, but not necessary.

    A “dashboard” interface would be nice, but seemingly simple enough to do.

    What would seem more important for proliferation is a series of rich frameworks that complement mods, (my favorite [currently a little outdated] example being vegastrike) and free (quality) models textures sounds and other related content.

  • Per Thulin

    Sounds a bit like the http://www.gamix.org project.

  • http://www.realnitro.be/ Jens Geiregat

    How about this portable gaming device? http://www.hackaday.com/2007/07/12/vc/

  • http://mirrorbingo.blogspot.com/ dailymirror bingo

    As for comments regarding restricting this to open platforms, why not make the game a liveDVD based on Linux/BSD/whatever?

  • http://www.personalisednumberplates.org.uk personalised number plates

    i agree totally a live dvd would be really cool