>>786145I think if you are into game development you also notice that my view is more realistic than is commonly believed. Take Rockstar for example, they aren't actually making individual houses and shops and streets in the game. Instead they take a top-down view of city outline and they have a machine learning software that constructs logical city structures such as road marking and parks and shops with sane and correct proportions using that said blueprint. They also have material generation algorithms that simulate the creation of stainless steel and other material which they slap on their 3D designs. Even their animation are made by tools that scan faces and capture real-life motion instead of developer working painstakingly on every pose and keyframe. So in reality they are making far less content than is commonly believed. Their competitive edge is coming to the gaming scene early and amassing huge capital. Imagine I had the AI and animation tools and other hardware and software they have, I'd certainly make a big enormous game in little time. I especially envy them for having that 100 thousand dollar tool that satellite view of a city and generates very realistic graphics from them.
The other thing about making video games is that even though it's hard to code entire game concept, it's actually easy to create good-looking games. You have fundamental concept such lighting, pbr, decals, shadow, bloom for example. And doing them will gives you great fidelity to AAA game graphics.
The main issue is not graphics actually. It's instead math which game developers are notoriously horrible at doing. In fact, barely any indie game developer has a good grasp of math. Like almost of them think that camera object is moving through the world space. You could look at it that way. But the more precise way to think about this is to realize that the drawing function is using cartesian coordinates -> f(x, y, z) and so to move in the world you don't move in the world space because it's always the same and you can't change it's geometry to make it render in other ways. Instead you use shift function and that's where camera comes it. It just applies offsets to graph of the world making it from f(x, y, z) to f(x - camera.x, y - camera.y, z - camera.z). So, basically, you are not moving the camera. You are shifting the graph of the world rendering.
I think this what prevents most indie developer from making large AAA-like games. They can't handle the math. They prefer the comfort of flipbook animation instead of interpolation keyframes using math. That's all. I think the game I have in mind is very doable with many of the current AI tools. I'm glad they finally trickled down, but this is a mere drop in the ocean triple A studios have actually. But imagine, just imagine, the game in my mind is not so far away in the future but instead just around the corner. I wake up early to work on the game when I think of this.