We do need a modern game engine for pascal - there are a number of people working on the next generation of game engines for our preferred language - Castle for example has matured into an excellent engine and it's getting better all the time. We don't yet have an Ogre3D or a Unity but we're closer now than we ever have been.
But that's not the main reason for a community engine - at least not for me. I wanted to see the community come together and pool its collective knowledge on a large project. Something we can all be proud of, spawn new collaborations. Something that people can learn from and even use in their own projects. It doesn't have to be Ogre3D, there's engines in development right now such as Castle that are tackling the high end - it's not for commercial success or making money - it's to have fun, together.
"Make games, not engines" is fine if you're making small games but isn't going to cut it for large games. I'd love for someone to be able to pick a modern pascal game engine off the shelf and write a modern game but we're not there yet.
I'm personally not interested in making games, I'm interested in making engines, developing new techniques and systems to abstract the game development process. I'm fully aware that engines are massive project and I'm fully aware that many engines fail - I'm not some kid dreaming of making the next quake! I'm a professional developer that's taken on a massive task in his spare time because it's fun to me
But at the same time, "make games, not engines" is the realm of the hobby coder making small indie titles that have access to things such as Unreal and Unity. People still need to render 3D objects, use shaders, utilise collision detection, path finding, steering etc etc
There's still an engine inside every game even if that engine is a one off specifically tailored for the job - there's a clear line between code related to gameplay and the code it calls as an abstraction to get the job done.
I understand the core mentality that the article is trying to get across - if somebody wants to make a game they should focus on making the game as it'll be more productive towards that end goal. It's trying to dissuade hobby coders from reinventing the wheel or writing code that they can't or won't use. It's written assuming people will be writing games in C++, javascript or any other far more popular language where all the tools already exist.
"make operating systems, not compilers" only makes sense if there's already a modern tool chain you can use. Making a modern 3D game in pascal right now, given its available engine/game related resources is like writing Linux in Cobol from scratch - yeah you can do it but the job would be way easier with the latest GNU C++ tool chain.
Almost everybody shouldn't be writing engines - it's bloody difficult and most want to make games so "make games, not engines" applies to them. But *somebody* has to write the engines, develop modern examples with Vulkan, GL4.x, write steering libraries, skinning libraries, asset trees so that people *can* make modern 3D games. Without a Unity, Unreal, Ogre3D etc to start from - that task is monumentally difficult.
Yeah you *could* of released Skyrim using the morrowwind engine and it would of been a good game but not as good as what they did release. We're moving forward, not backwards.
Can we please get back on topic and try an analyse what went wrong with the project so if there is another effort, we can make it work this time.
Reply With Quote
Bookmarks