Page 3 of 3 FirstFirst 123
Results 21 to 29 of 29

Thread: Unvanquished Still Looks Amazing For Open-Source

  1. #21
    Join Date
    Feb 2008
    Location
    Linuxland
    Posts
    5,196

    Default

    Quote Originally Posted by 89c51 View Post
    Then how on earth you don't have html5 video. Opera supports this afaik.
    What kind of question is that? I disabled it of course

  2. #22
    Join Date
    Jan 2009
    Posts
    1,711

    Default

    Quote Originally Posted by curaga View Post
    What kind of question is that? I disabled it of course

  3. #23
    Join Date
    Jun 2010
    Location
    ฿ 16LDJ6Hrd1oN3nCoFL7BypHSEYL84ca1JR
    Posts
    1,052

    Default

    Quote Originally Posted by elanthis View Post
    why make a game and give it away for free when I can make the same game and sell it and become a millionaire and never need to work for anyone but myself ever again?

    If you want big content RPGs and story-driven action/adventure games -- things like Dragon Age, Red Dead Redemption, etc. -- then some things need to change in FOSS land. A reasonable revenue stream for FOSS games would need to be found. This precludes packaging them with distros, who are known for stripping revenue streams from individual upstream projects in favor of using their own. A unified installer framework (for the love of God, Linux needs this) would allow game developers to offer easy installers on their sites right next to big Donate buttons, or ask a user to enter a payment amount of their choice to get the precompiled binary (Humble Bundle style). There'd also need to be some good physically centralized places to find FOSS teams, since working together over the Internet keeps teams working at a fraction of the efficiency that working together allows, and that efficiency is needed to get games out the door before they become stale. Lastly, people would need to learn to accept that these games would most often be released as FOSS but would not be developed as FOSS because of the importance of keeping projects under wraps, both for marketing reasons (you publicly announce when you have something impressive to show and not a second before) and to keep player interest from waning before the game is even released.
    I think I read your earlier post and it sounds mostly reasonable but I don't completely buy it.

    Obviously for a story driven game it isn't the best idea to release every bit of work as soon as you have completed it.

    But you still could do a reasonable mix of holding things closed and releasing other things.
    I think about something like wolfire is doing with overgrowth, but a little different:

    The engine is open source and every work gets immediately added.
    The assets (level files, speech, dialog, textures) that end up in the final game are not released / are only accessible with NDA for volunteers who help develop for a little share.
    Initially there are some demo assets released, just big enaugh so you can test some things the engine allows you to do.
    Occassionally there are videos released that show concept art, the progress in the code and maybe some of the assets in a way that doesn't reveal the story.
    The game will not be for sale at the end of development but it is continually funded by preorders and donations.
    In a strictly linear game you could manage to not release a 8+ hour game but maybe one/two/three hour long parts at a time. That would allow great feedback but the experience would maybe be too splitted up.

    Maybe this doesn't work. But at least I don't see a really strong reason why it absolutely couldn't.


    Quote Originally Posted by elanthis View Post
    A community of hobbyists cannot build those today, the exact same way that a community of hobbyist film makers cannot make Avatar today, and a community of hobbyist engineers cannot create the USS Nimitz. There's just a point at which both the resource costs require a huge company's cash (which in return expects huge profits) and at which you require full-time top-notch people (who in return want/need to be paid for their hard work).
    Why would only hobbyists be involved in this? Top people are of course not there because it has not been proven to work or proven to make money. But on the other side there is still this cliché about linux developers being all hobbyists who develop in their free time when in reality there are many major companies and to people involved. Of course the comparison is flawed because linux is "the other type" of software, but it's still something you can think about.

    In an open source game not everything would need to be opened in the end, I can see that voice actors etc. would be concerned about their work becoming creative commons but that stuff could be put under a proprietary free of charge license (they get paid for their work in the development phase).

    Quote Originally Posted by elanthis View Post
    Outside of a few rare and unexpected gems, FOSS simply isn't going to be a hot bed of game development. The best use of FOSS is to be the repository for old games. Which I still think should be mandatory; release a game under a FOSS license 3-5 years after initial release. This is critical to ensuring that the cultural influences of games are preserved for generations of gamers, historians, and artists to come. (For one example, I can't for the freaking life of me get one of my most favorite games, Fallout New Vegas, to run on any of my computers anymore, because of its numerous bugs; the profits from that game are drying up very quickly, and releasing the source would allow folks like me to go in and surgically fix the game, allowing it to be playable with a minimum of fuss by others on continue to be supported on future popular gaming OSes, and would even allow ports to Linux or OS X or iOS or so on, in addition to being a huge boon to the modding community that keeps games like that alive and interesting to regular gamers.)
    Freespace 2 is a great example for this. There's the freespace scp project where enthusiasts keep developing the engine and I think some months ago there was a big release of a mod based on the freespace scp but I can't quite remember what it was...

  4. #24
    Join Date
    Jun 2010
    Location
    ฿ 16LDJ6Hrd1oN3nCoFL7BypHSEYL84ca1JR
    Posts
    1,052

    Default

    Quote Originally Posted by curaga View Post
    It's occasionally useful to remind oneself how low can humanity go, lest one forgets that

    @ChrisXY:
    I do have a downloader script at hand for the one decent video per year. But it takes a lot to convince me that _your_ video is worth wasting time over among the cesspool of lolcats, attention whores, and people imitating the Dudesons. Wait, I think I just repeated myself thrice.
    See, because of your self-induced problem you will never find out if it would have been worthwhile.

    Quote Originally Posted by curaga View Post
    As for the content, there's exceedingly little to be gained from a video about most of the topics you mentioned. Reading is much better and faster for technical matters (Rpi) than waiting someone go "umm.. eh... now...",
    Well, in a video you can actually see it in action. A working home made (rather inaccurate though) ink printer is just more impressive to observe than to read about.

    Quote Originally Posted by curaga View Post
    Intellectual Property likewise. If there's something there on a video that's not available in text elsewhere, well, their loss.
    It's a humourus presentation and I don't believe it translates well to text.

    Quote Originally Posted by curaga View Post
    For music, why would you subject yourself to the bad quality there? It's heavily compressed AAC if I remember right, with clearly audible distortions even over standard computer speakers, not to mention any better set.
    1080p videos have 128 kbit aac. That's not really impressive but good enough to listen to. I actually use spotify premium (let's see if I keep that or if they annoy me too much) but not every live performance is available there and there is no video.
    And again, presentation is a big part of the experience for http://www.youtube.com/watch?v=3YMXpFca0ko or http://www.youtube.com/watch?v=fmYAXUHgvEQ

  5. #25
    Join Date
    Feb 2008
    Location
    Linuxland
    Posts
    5,196

    Default

    Quote Originally Posted by ChrisXY View Post
    See, because of your self-induced problem you will never find out if it would have been worthwhile.
    You're correct there. But I'm hedging my bets; the 99.9% figure is bound to be close to accurate. That means I'd waste time watching 999 videos for every decent one.

    That's ~83 hours wasted at 5min per vid.


    Contrasting this with text, one cam skim a page in a few secs to determine if it's worth reading. To do so for a vid requires watching a minute or so.

    It's a humourus presentation and I don't believe it translates well to text.
    Cool if you like that, but sounds like it'd be a waste of time for me to watch then.

    1080p videos have 128 kbit aac. That's not really impressive but good enough to listen to. I actually use spotify premium (let's see if I keep that or if they annoy me too much) but not every live performance is available there and there is no video.
    And again, presentation is a big part of the experience for http://www.youtube.com/watch?v=3YMXpFca0ko or http://www.youtube.com/watch?v=fmYAXUHgvEQ
    The 1080p videos are also huge, are they not? 100-300mb of bandwidth wasted for video when you only wanted the audio

    Derail lvl 1:
    I've found a lot of great music at Jamendo. Music and only music, in both mp3 and ogg, with high-quality files in torrents available.

    Derail lvl 2:
    (Until the recent redesign though, which made the Jamendo site both a pain to navigate and very laggy. It seems the majority of the artists think so too, so unless they change things, I'm afraid that move will cost them a lot of their artists and users, if it hasn't already.)

  6. #26
    Join Date
    Nov 2007
    Posts
    1,024

    Default

    Quote Originally Posted by ChrisXY View Post
    I think I read your earlier post and it sounds mostly reasonable but I don't completely buy it.

    The engine is open source and every work gets immediately added.
    The assets (level files, speech, dialog, textures) that end up in the final game are not released / are only accessible with NDA for volunteers who help develop for a little share.
    Then it's not a FOSS development process for the actual game at all, is it?

    Even if you decide that games themselves don't need to be FOSS or developed as FOSS, what is the motivation to releasing the engine from day one of development? If I have a medium-sized core team of 30 devs who are working on an insanely tight schedule delivering very specific bullet point features who can all work in the same office with the ability to directly talk to each other, work out problems on white boards, meet with the effects artists and designers and sound engineers face-to-face and hash out lists of feature needs, etc., what possible gain in all the world is there to adding the _overhead_ of publishing source code to a public repo, answering questions about the code, reviewing patches to code that are not part of the tight feature schedule that internal devs are already allocated to, and so on when that community can't even know what the end goal is or what needs our content and game design have on the engine? The company is supposed to pay a second "community lead" just to coordinate with people outside the company who have to be completely left out in the dark about what this thing is even for? And we're supposed to have the legal and managerial overhead to support a few NDA'd developers who are inherently forced to work at a fraction of their optimum efficiency because they're not full-time and not able to work in our office instead of just hiring them if they're actually good enough to bother working with? And I have to risk an otherwise toothless NDA with some random individuals on the 'Net, rather than with an employee or paid contractor who at least has a career to lose if he gets caught breaking an NDA in our industry?

    Again, games have tight schedules, and the slow meandering Internet community pace that works just fine for a kernel or a web server or the 180,000th ioQuake derivative simply does not work for a brand new high-quality content-driven game; it has so far had trouble even just cloning old low-quality tech-driven games. Having people come and go and scratch itches just doesn't work here. Especially if those itches don't even occur because there's no complete game to play and hence no reason to even find out what is missing or bothersome or interesting to change. If I jump on a pre-alpha FOSS scripting language, I can immediately start implementing features based on what I know the end product will be (a scripting language). If I jump on a pre-alpha game engine for which I have no idea what the final game is, I can maybe start adding random features I think might be interesting but which will just be rejected because they add complexity and maintenance overhead that the game doesn't need and that's about it.

    Let's not also forget that the idea that an engine and the game itself are totally separate for most games is a complete fallacy. Sure, the core low-level engine can maybe be separate (maybe), but not any of the interesting bits. The idea that we're going to spend even 1 second building some kind of large library of public-friendly test assets when all of the actually meaningful development, testing, verification, and design is going to happen with our actual private assets is also just absurd; it would be nothing more than a huge waste of time and resources. Core game mechanics are tightly integrated into the engine itself, because they have to be in order to actually work. This is one of several reasons why commercial game engines come with source for any serious developer; you're going to be building your game into the engine, not onto the engine.

    Engines can be thought of as application framework libraries rather than application platforms, in that sense. Those custom game features are tied to data. The engine I'm on right now, for instance, is impossible to separate from our core game, because it's an engine being built for a specific genre with some very specific innovative game features for a very specific franchise, and aside from the unmodified bits of the engine we are building on top of, there's really nothing we've written that is particularly generalizable to other genres; a lot of it is code for specific features that aren't even generalizable to other games in the same genre. There are few ways to release code and not also release the "NDA assets" (which includes the core idea of the game itself!).

    Point being, you either develop the entire game, or you develop nothing. Engines do not exist in isolation. Not one of the major engines used today (besides maybe Unity) was developed as an engine separate from a game: Source was made for Half-Life, CryTek for Crysis, UE for Unreal, etc. That in turn is why they are not generic or easily reusable for other genres. Engines that are developed without a specific game either never see the light of day, waste away at around 4% completion on SourceForge, or turn out to suck pretty hard and take additional years of effort to fix up and put into good shape (by building real games on top of it and fixing the problems as they come up) like what happened with Unity.

    Why would only hobbyists be involved in this? Top people are of course not there because it has not been proven to work or proven to make money. But on the other side there is still this cliché about linux developers being all hobbyists who develop in their free time when in reality there are many major companies and to people involved.
    Like I explicitly said, there are no companies involved in FOSS games. Paid FOSS developers in the kernel land and such are paid because there is a company there making fat stacks of cash off of the whole shebang in some way and paying those developers to keep that money rolling in. The support stories for those companies is entirely _enterprise customers_, not regular consumers. The income and support models for a consumer product and an enterprise platform are not the same at all, and what works for one does not work for the other.

    When is the last time you bought a support contract for Red Hat Enterprise Linux or SuSE or Ubuntu for your home PC? There is no currently known way to release an entire game at AAA production values and get said fat stacks unless you charge for the game, so nobody is going to risk spending millions of dollars on engineers and content authors with no plan to actually make a profit (and a big enough profit to be worth the effort); they're risking enough just in hoping that the game will actually be a hit and not a dud. Even the Humble Bundle guys have indicated how very little cash the actual game developers get when all is said and done; a $2m indie bundle, split 8 ways when counting charities and humble tip, minus CC processing fees and taxes... really just not that much money for the amount of work involved in each game.

    If you made some dinky little game out of love and got to see something back from it, that's awesome, but if you're investing millions upfront to make the game, you expect a crapload more in return than the donation scheme has generated to date. Of course it may be that if Assassin's Creed III were in the Humble Bundle the bundle would make $1b, but no executive in charge of the sales of a game like that is going to be dumb enough to randomly chance it until it's been proven (and how that gets proven is one of the problems to solve, of course).

    In an open source game not everything would need to be opened in the end, I can see that voice actors etc. would be concerned about their work becoming creative commons but that stuff could be put under a proprietary free of charge license (they get paid for their work in the development phase).
    And then we're back to the game not really being FOSS at all.

    Are we arguing about the feasibility of FOSS game development or about FOSS middleware? Because I'm 100% on board with there being more high quality FOSS middleware. Bink, Scaleform, Havok, SpeedTree, FaceGen, FMOD, etc. could all be made obsolete by FOSS, if the right people put in the effort. Granted, the best FOSS has done for middleware in games so far is Bullet (not used in commercial games because it had some basic rigid body stuff; compare the feature set of Havok to Bullet some time and tell me whether it's worth spending cash on Havok or not) or OpenAL (comparing that to FMOD is like comparing 'ed' to LibreOffice Writer).

    I believe FOSS _can_ succeed in the middleware arena (it's already done well with the really simple stuff like Lua, zlib, libvorbis, etc.), but it certainly hasn't succeeded on a big level _yet_, and that's going to be the first bridge to cross before even considering completely FOSS games or high-quality FOSS engines on par with CryTek or whatnot. Seriously, if nobody in FOSS land can recreate SpeedTree of all things, what hope is there that FOSS is going to create an original title like Deus Ex or Grand Theft Auto?

    Freespace 2 is a great example for this. There's the freespace scp project where enthusiasts keep developing the engine and I think some months ago there was a big release of a mod based on the freespace scp but I can't quite remember what it was...
    Sure, idTech games as well. I again pointed out the importance of releasing older games as FOSS, because not doing so is damaging in a lot of ways. And hobbyists can certainly do some amazing stuff in small islands (the mod scene for even very new games shows this). They can do a lot more when they're working on old, well-understood technology. That is not the same thing as developing an entire game or engine from scratch. Not even close.

  7. #27
    Join Date
    Oct 2008
    Posts
    898

    Default

    Quote Originally Posted by elanthis View Post
    tl;dr
    Good games aren't being made with OSS development because no one wants to work 80+ hours a week for free.

  8. #28
    Join Date
    Nov 2008
    Location
    Madison, WI, USA
    Posts
    877

    Default

    Quote Originally Posted by curaga View Post
    Flash is shit in so many ways, I trust there's no need to write an essay here on that topic, it would be preaching to the choir.


    On HTML5 video:

    99.9% of Youtube is crap. Even the people that go there are the lowest of the low as witnessed by the Youtube comments, that place makes 4chan look like a civilized upperclass gentleman club.

    No, yet another video of your cat skateboarding is not funny.

    With Youtube being not useful at all, I haven't found any use for HTML5 video. inb4 porn.
    Exception to the rule: Step-By-Step repair guides for automobiles. I've got an electronic copy of the repair manual for my car (2001 Jetta), but sometimes a step-by-step video showing exact bolt locations and the orientation of the parts with respect to each other can be invaluable.

  9. #29
    Join Date
    Oct 2010
    Posts
    325

    Default

    Quote Originally Posted by curaga View Post
    Derail lvl 1:
    I've found a lot of great music at Jamendo. Music and only music, in both mp3 and ogg, with high-quality files in torrents available.

    Derail lvl 2:
    (Until the recent redesign though, which made the Jamendo site both a pain to navigate and very laggy. It seems the majority of the artists think so too, so unless they change things, I'm afraid that move will cost them a lot of their artists and users, if it hasn't already.)
    Well, since the thread is already derailed let's take it further

    Browsing the Jamendo library works the same when done from withing Amarok, no matter how they change their website. Though there's nothing Amarok can do about the lag...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •