The only way to prevent it from imploding on itself is to make off-the-shelf components viable, and to improve the tools used to develop content. Honestly.... doing good-looking 200 polygon characters is hard, but not nearly as time consuming as doing 20,000 polygon characters.
The problem is very few of these available tools are either general enough to be used across a wide variety of games, or they are so general that they require a 6-month training course for people to learn to use, or they are slow, or they are buggy, or they are too expensive.
We'll get there. Once we accept the fact that games are too freaking huge to have the attitude that everything must be written / created in-house. That's the kind of innovation I'm seeing here... coming from guys with really small budgets.
#1 - More complete components: The Torque engine (and similar high-level components). Two thirds of your game code is already written for you. You can focus on the CREATIVE, fun stuff, rather than the not-so-fun stuff like network coding or the core rendering engine (though it can be fine to improve / update it).
#2 - Higher-level tools for rapid "rough" content generation. I'm going to cite Joshua Ritter's path of creating levels in a DOOM editor (fast & easy, but very limited), and them exporting them to something more powerful and detailed like Quark. Rapid building / prototyping tools is where it's at.
#3 - Off-the-shelf content, like you find at Game Beavers or here at Garage Games. I'm thinking like a set designer for a local play should think... you don't want to build EVERY SINGLE prop from scratch, do you? But that's what we're doing in game development all the time. Instead, you go to the thrift shop, you dumpster-dive, you do everything you can to get everything "off-the-shelf" that you can... and THEN you build the signature pieces or the pieces you couldn't find anywhere else.
#4 - More use of higher-level languages for scripting of non-bottleneck code. Torquescript, QuakeC, Java, Python, Neverwinter Night's scripting language, ... all stuff getting used to get more work done with less effort (and making things easier to extend...)
#5 - And, as the author of the article mentioned, better tools geared more for game developers.
Torque 3D Owner Jay Barnson
The only way to prevent it from imploding on itself is to make off-the-shelf components viable, and to improve the tools used to develop content. Honestly.... doing good-looking 200 polygon characters is hard, but not nearly as time consuming as doing 20,000 polygon characters.
The problem is very few of these available tools are either general enough to be used across a wide variety of games, or they are so general that they require a 6-month training course for people to learn to use, or they are slow, or they are buggy, or they are too expensive.
We'll get there. Once we accept the fact that games are too freaking huge to have the attitude that everything must be written / created in-house. That's the kind of innovation I'm seeing here... coming from guys with really small budgets.
#1 - More complete components: The Torque engine (and similar high-level components). Two thirds of your game code is already written for you. You can focus on the CREATIVE, fun stuff, rather than the not-so-fun stuff like network coding or the core rendering engine (though it can be fine to improve / update it).
#2 - Higher-level tools for rapid "rough" content generation. I'm going to cite Joshua Ritter's path of creating levels in a DOOM editor (fast & easy, but very limited), and them exporting them to something more powerful and detailed like Quark. Rapid building / prototyping tools is where it's at.
#3 - Off-the-shelf content, like you find at Game Beavers or here at Garage Games. I'm thinking like a set designer for a local play should think... you don't want to build EVERY SINGLE prop from scratch, do you? But that's what we're doing in game development all the time. Instead, you go to the thrift shop, you dumpster-dive, you do everything you can to get everything "off-the-shelf" that you can... and THEN you build the signature pieces or the pieces you couldn't find anywhere else.
#4 - More use of higher-level languages for scripting of non-bottleneck code. Torquescript, QuakeC, Java, Python, Neverwinter Night's scripting language, ... all stuff getting used to get more work done with less effort (and making things easier to extend...)
#5 - And, as the author of the article mentioned, better tools geared more for game developers.