Announcement

Collapse
No announcement yet.

FOSS and Game Development: the Problem

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by kraftman View Post
    You're mixing things up. Nobody said they'll have to work for free. They should work normally, but using Open Source development model. Nobody from big players ever did this, but it's probably because they have specialized teams that don't need some random volunteers help. My point is Open Source development models isn't worse than proprietary one in games development.
    No, I'm not "mixing things up." The burden is on you to show it is possible to raise $100,000,000(US) to cover a triple-A game budget.

    Comment


    • #32
      Originally posted by yogi_berra View Post
      No, I'm not "mixing things up." The burden is on you to show it is possible to raise $100,000,000(US) to cover a triple-A game budget.
      Last I heard, a "triple-A" 360/PS3 game cost about $10-20M to actually develop depending on how "big" a game you're going for and how much in-house stuff can be leveraged (e.g. Gears of War 2 is said to have cost about $10-12M to develop). A $100M quote probably included an absolutely massive marketing budget.

      Comment


      • #33
        Originally posted by yogi_berra View Post
        No, I'm not "mixing things up." The burden is on you to show it is possible to raise $100,000,000(US) to cover a triple-A game budget.
        I was just talking about development model not about possibilities. Just for example iD making games by using Open Source development model will release as good games as using proprietary one.

        Comment


        • #34
          Originally posted by kraftman View Post
          I was just talking about development model not about possibilities. Just for example iD making games by using Open Source development model will release as good games as using proprietary one.
          But your development model doesn't pay your animators, voice actors, musicians, environmental artists, writers, texture artists, low-level programmers, level scripters, project leads, office staff, caterers, testers, and marketing agents. You'll need to raise money to do that.

          e.g. Gears of War 2 is said to have cost about $10-12M
          According to imdb Skyrim's budget was $100,000,000. Before they lost the rights to it, Interplay was supposed to raise $30M to show they could produce "Fallout: Online" Ten to Twelve million is a low budget.

          Anyway, the community isn't going to raise One million (Novacut has yet to raise $30,000), let alone the budget to a Triple-A game like Skyrim.

          You might be able to develop a casual game via the open source model, like Pissed off Penguins is attempting to do, but any non-casual game is out of reach for the development model.

          Comment


          • #35
            Originally posted by yogi_berra View Post
            Anyway, the community isn't going to raise One million (Novacut has yet to raise $30,000), let alone the budget to a Triple-A game like Skyrim.

            You might be able to develop a casual game via the open source model, like Pissed off Penguins is attempting to do, but any non-casual game is out of reach for the development model.
            I'm not sure those figures are completely relevant for an open development model. Yes, you do need to pay coders and artists, but you also have to take into account you can leverage copyleft code and content. Independent developers can also contribute to the project for free, even if it's just fixing bugs (speaking of triple A games like Skyrim or other stuff by Bethesda ). And even releasing incomplete games like 0AD helps in the long run.

            Besides, I think we do have to consider just how financially-efficient proprietary development is, at least in practice. Maybe all this intellectual property protection makes it less important to minimize costs to get excess profits. I'm not saying proprietary development is overall bad or inefficient at satisfying these needs, companies aren't stupid, they're probably doing what makes most sense under the current conditions. But it's something to consider.

            I also don't think we'd mind open games lagging proprietary stuff a year or two, in terms of content complexity. Or perhaps it could take a bit longer to deliver, compared to a proprietary game. It'd certainly help if there was a lot of good content out there to draw upon and I think that might get easier at some point. IMO, things could get better, even if not totally on par and on time with triple A titles.

            (Also, speaking of TES games, it's interesting how Bethesda managed to attract modders. Perhaps that's something we can learn from.)

            Comment


            • #36
              Originally posted by Eduard Munteanu View Post
              Yes, you do need to pay coders and artists, but you also have to take into account you can leverage copyleft code and content. Independent developers can also contribute to the project for free, even if it's just fixing bugs (speaking of triple A games like Skyrim or other stuff by Bethesda ). And even releasing incomplete games like 0AD helps in the long run.
              If the strategy is to rely on the pity contribution, your game will end up looking like Open Arena or fail entirely. You have to attract talent and the promise of regular paychecks and benefits tends to do that.

              Comment


              • #37
                Originally posted by DMJC View Post
                There is no free artist shortage, just look at mod teams. The real question is, why aren't they targetting open source engines? 90% of mods suck/die etc. but that 10% is pro quality work, and that is what the foss game devs need to attract and retain.
                I believe this goes back to the nature of FOSS and artists.

                Mods are generally not Open/Free. The artist retains "rights" over their work that they do not have with FOSS, such as the right to tell people not to modify their creative work in ways they don't approve of. This kind of thing really matters to a lot of artists. There's a lot more emotional investment in art than there generally is in implementations of algorithms.

                Originally posted by yogi_berra
                I'm not sure those figures are completely relevant for an open development model. Yes, you do need to pay coders and artists, but you also have to take into account you can leverage copyleft code and content.
                Pretend I wanted to start such a game today. The first question I'd need to ask you, then, is "what copyleft code and content" ?

                If there were significant bodies of such code, this entire discussion would never have happened in the first place, because the very existence of such assets would mean that FOSS games were already being developed and released in sizable quantity and with sufficient quality.

                Originally posted by Ex-Cyber
                Last I heard, a "triple-A" 360/PS3 game cost about $10-20M to actually develop depending on how "big" a game you're going for and how much in-house stuff can be leveraged (e.g. Gears of War 2 is said to have cost about $10-12M to develop). A $100M quote probably included an absolutely massive marketing budget.
                Gears of War is very atypical. The Witcher also came in around $12, and again it was atypical and a lot of people were really amazed by the quality of the game for how cheaply it was developed. $10M is _tiny_. Marketing budgets on games are roughly only 30% of the total budget. Some of the bigger AAA games went over $230M in costs.


                Originally posted by blackiwid
                1. Manpower: game projects in this dimensions are very big, so you have to have much manpower for a long time.
                That's kind of my thesis. Manpower should be the easiest problem to solve of all of them in the FOSS community, and yet it turns out that there's barely anyone with the necessary skills making games for Free.

                2. Moneyproblem:
                Again, if we assume that FOSS means free labor and the community means free advertising, money should not be the issue that it is.

                3. tools: I just made for fun to test the blender game engine a tictactoe game, I must say, this blendertool-integration of a engine with the gui and the python api there to make games also you can make games without even programming with python is a bit gain for linux game development
                You can't make decent games with those tools. Note that those tools _are_ used a lot in the industry, but only for prototyping.

                Pay attention to even a mediocre "real" game. Pay _real_ attention to it. All the little details. The menus, the controls, the feedback systems, the AI, the effects, etc. No tools allow these to be made without real programmers putting in a lot of time because for every game those details are pretty different.

                We're slowly seeing more and more tools made for these tasks, but I've yet to see any in a FOSS package. UnrealEd and the like can do some really impressive things (but still, having worked with plenty of design/art teams who've used Unreal as a basis... you can only go a very limited distance without lots of programmers), but Blender is nowhere even remotely close.

                4. getting together a team with same goals, for comercial games the goal is clear earning money,
                I don't agree that the goal is making money at all. It's a requirement a developer has when he's expected to work 40-60 hours/week because he needs to eat and pay bills, certainly. However, companies are generally very VERY keen on hiring _passionate_ talent that really believes in the project, meshes well with the rest of the team, and so on.

                Also note that game programmers get paid WAY less than programmers typically get paid in other IT/CS fields. Average game programmer salary is a measly $50k, with the "top talent" at most companies making only around $120k. Comparatively, Microsoft's *interns* make $90k, and run of the mill government and financial sector programmers make six figures. I took a pay rate hit of almost $30k/year when I switched to the games industry. And game artists don't get paid worth crap. We call it the "game tax" -- since we're willing to work for less because we love the job so much, there's no reason to pay us more.

                Originally posted by crazycheese
                Point nr 0. I don't quite understand the reason behind your discussion. Try going to MSDN and start talk "windows is dead end and you should go develop somewhere" - you will be banned. If you up to developing games, you should be up to searching methods to succeed in your goals. Not arguments to justify the laziness. You don't need discussion for that, just sofa. You want to start opensource game or just talking?
                You're making some huge logical jumps. Going to a Linux forum and saying "Linux is dead" is not at all the same thing as going to a Linux forum and saying "Linux's development model doesn't work for one niche of application." In fact, it's a lot like going to a Windows forum and saying, "Linux makes a better Web server." If you get banned for saying that, you're on a pretty worthless forum run by a bunch of insecure idiots.

                The rest of your post is filled with way too many (very weak) strawman arguments that don't even have any real connection to the actual discussion.

                Originally posted by Eduard Munteanu
                I'd conjecture artists are less willing <snip> simply because they're not comfortable with working for free (which needn't be the case).
                Sorry, but this just made me laugh out loud, literally. Artists are very, VERY used to getting paid jack shit, and often working for free. Even in games, artists make a tiny percentage of what programmers make. If I recall the figures correctly, the average artist salary in the games industry is around $23k/year, which I think is only just barely above the poverty level. Almost no art internships are paid, and artists are expected to have a massive and impressive portfolio (which can take months or years of near full-time work to accumulate) to even be considered being hired for a "real" job. Also, there is a surplus of artists with the necessary skills, so pay rates have been driven down further and there are a lot of them without work. Certainly the artists working full-time right now and getting paid decently aren't as likely to be up for working for free on a Free game, but there are plenty of artists with nothing else to do but bone up their portfolios.

                No, money is not the problem for artists.

                For all we know, we could have some excellent free MUDs, but that's not something a gamer usually considers a top-notch game.
                Actually, MUDs are notoriously anti-Free. It always confused (and really bothered) me back when I was in the MUDing scene. The logic came down to "we don't want other people to steal our gameplay and features."

                Which now that I think about it today, goees back to games being much more than just algorithms and code; even all the things implemented as algorithms and code equate to something very, very different than a compiler or a web server. They're an artistic expression, an experience, the foundations for worlds. Not saying that justifies them being proprietary, just that the mindset between the "why should they be Free" is entirely and absolutely different from software oriented towards function rather than experience.

                It makes me wonder if a viable compromise is procedural content generation. For example, recent TES games have been using this for stuff like faces and such, which seems to show you can get high quality generated content.
                Experiences so far have not worked out well. Aside from random dungeons in things like Diablo (and its Roguelike predecessors), procedural content has mostly been uninteresting to the market. TES's FaceGen (a proprietary middleware they license) is state of the art and yet the quality of the faces it produces is generally considered to be very below par compared to traditional character art, and it tends to have performance issues due to the graphics pipeline requirements.

                I absolutely believe that procedural content will be a huge part of the future, but it's not really a viable approach on a larger scale just yet. This is certainly an area that FOSS projects could pioneer in, though!

                Originally posted by kraftman
                You're mixing things up. Nobody said they'll have to work for free. They should work normally, but using Open Source development model. Nobody from big players ever did this, but it's probably because they have specialized teams that don't need some random volunteers help. My point is Open Source development models isn't worse than proprietary one in games development.
                If we were talking purely about coding, I'd agree. Just in terms of code, there's no reason not to use an Open development model.

                One of my main points, however, is that games are not just code. Even many of the bits that are "code" are things that transcend algorithms and architecture and the things programmers normally deal with. The Open model has potential costs on those aspects of the project. To a degree those are just hypothetical costs, but experiences with what works and what utterly fails in the marketing and community aspects of games from the last 30 years strongly implies that being too Open before release can hurt more than help.

                The FOSS community I believe has trouble grasping those costs because they're so very alien: absolutely no other kind of software besides games has any reason to think about "spoilers" or "game experience algorithms" or "marketing media." All other software is _functional_ by nature; their purpose is to consume, transform, and generate data. A game's sole purpose is to affect the player's mood.

                Comment


                • #38
                  You got your quotes wrong I didn't say:
                  I'm not sure those figures are completely relevant for an open development model. Yes, you do need to pay coders and artists, but you also have to take into account you can leverage copyleft code and content.
                  That was Eduard Munteanu and I believe he was talking about reusing existing code and content. Oh and $23k/year is poverty level but not the average which is actually around $60k.

                  If I remember correctly $23k/yr is the high end of what the 2d anime studios pay their tweeners in Tokyo, which is insane considering Tokyo's cost of living is so outrageously high.

                  Comment


                  • #39
                    But then, which kind of games?

                    There is a basic question in the first post which remains unanswered: which kind of games is best suitable to be developed as FOSS?

                    Storyboard based games, like old dear LucasArts/Sierra adventure games are out of the question: if you can access the solution of each and every single puzzle of the storyline before the game is done, there's no game at all for you - it is like being a magician who cheers and applaudes at his own tricks.
                    Shooters should do better: no need of hidden story, just blast. And after you blasted every nasty thing who wants to destrtoy you, you can make mods and receive from others, just to start blasting again. But this, sooner or later, will become a lot boring.
                    Beat-em ups? like the shooters, but they become even more boring. And there is a big problem: who takes care of balancement? all characters must be balanced in their up and downsides, or someone might create a super-undefeatable char with a set of highly-destructive special moves (who are not secret even in the first place, since the game is opensource).
                    What about simulation games, them? Tanks, fighter-bombers... the playabe ones are not that different from shooters (the more realistic the flight model is, the less playable the game is: it takes years of studying and practice at the military academy to become a fighter pilot, a gamer is not supposed to take a full course just to play a computer game), you can add new planes or enemies... but like beaters, who takes care of the balancement (= who says no if my home-made tank can resist a dozen H-bombs, run 600 mph speed and carry 2000 nuke bullets that can clear a city in a single shot?)
                    role-playing? Ultima Online-like multiplayer ones might do, but single-player ones like non-online Ultimas are out of the question (they're more or less adventure games with a RPG-ish touch)
                    RTS/TBS? single-player ones must be storyboarded (no-go!). Multiplayer should fit best.
                    Racing? yeah... but like shooters and beaters, racing is racing, plain and simple. And balancement is fundamental (Formula-1 cars aren't dragsters and they're a lot different from Indys or NASCARs).
                    Tycoon games? I think they might go fine. Even single-player ones don't need a full storyboard, just goals to reach.

                    Suggestions?

                    Comment

                    Working...
                    X