Interestingly, I did not feel my project was mature enough to start a blog about it until I freed myself from tutorials.
I think tutorials are a trap; they are a convincing trap for game development because, let’s face it, even if you are an experienced software engineer, if you have never before worked with a complex game engine such as Unreal or Unity, you need some tutorials to get up to speed. They also have the potential to guard you against “blank canvas syndrome.” It is an excellent way to develop a habit of consistent work. It’s hard to break down a big task like a game into smaller steps, especially when you are not familiar with the tools, and a tutorial can help to give you a sense of consistent progress.
Starting a longish tutorial (30+ hours) as a base for your game makes so much sense—a foundation you can build upon. And so I did. I found a tutorial that matched my game type, but I had to break free after about 70% of it.
What went wrong?
First of all, once I got some experience with Unreal, it was easy for me to spot some weaknesses/inconsistencies in design. Blueprints are fine; I get it. However, there is a good proportion of blueprints to C++ code, and each project is different. For my project, putting in as much C++ as possible apart from the UI layout, animation sequences, etc, makes sense. It’s because I feel comfortable using C++, and I use it daily at my job, and it’s a solo project. I don’t have a game designer or another person who is not a coder to use more approachable blueprints. For someone experienced at coding, they add only performance hits and less clarity. Let’s not forget their apparent limitations.
The real nail in the coffin was how the Skills UI was designed in the tutorial. Instead of a more intuitive drag-and-drop operation, one must do three button clicks to assign a new skill. At this point, I was done with the tutorial, and then I noticed how much time I had wasted due to its tech debt. By ‘tutorial tech debt, ‘I mean the extra work and potential issues that arise from following a tutorial that doesn’t align with your project’s needs. I spent two weeks cleaning up the code, and the cleanup is unfinished. Overall, I spent a month with very little progress on the face of it, but the foundation of my game is more solid.
Am I never going to use a tutorial again? Not as much as before. At this point, I can figure out most things on my own using the official documentation. A tutorial can be helpful to see one of the possible approaches, but I will be much more critical. I plan to thoroughly evaluate the tutorial’s relevance to my project, its design and code quality, and its alignment with my preferred programming language and tools.