#Texturepacker sdl code#

It is much harder to attain faster build times with C++ code base because of all the standards, templates, standard headers and stl library. Faster build times are much easier to achieve with C. In C, despite it's shortcomings which many C programmers are aware of, there is just much less stuff to think when you are coding, so it doesn't gets in your way and leaves you to think about the actual problem, algorithms and data instead of what idiom, syntax, feature and construct to use. As for C++, each subset of it can be defined as a new language, and every C++ programmer has their own stuff, so the code may eventually become a mess of different C++ standards and idioms, and makes it very hard to read and maintain. So other C programmers can get right into your project without adapting to your standards. It's just simpler, there are much less idioms, patterns and stuff that you have to learn about the language compared to C++. But in lower-level systems and performance critical code, C is a much better choice than C++: In my opinion, orthodox C++ with a very strict feature-set is still a valid choice for game code and even some high-performance code. Maybe I'm not a good C++ programmer and I don't think I'll ever will be, because after C++11, I never even bothered to learn the new stuff. I can say after many experiences with C++, I'm still not comfortable with the language and it still bothers the hell out of me, especially when the project scales. In previous project (()) (fisherboy's game engine), I wrote the engine and game itself in C++ with a rather () approach, but with some minor changes integrating some more 'Modern C++' features, especially in game code. Read this (./fisherboy-technical) for a more complete technical explanation of that game. In this post, I'm gonna discuss the flaws and the stuff that nearly brought me to tears in the previous project _(fisherboy)_, and the stuff that I would like to improve in the next project. **Fisherboy postmortem and future plans**