Dev Log 1 – Procedural Generation vs Human Ingenuity

Introduction

When it comes to games design, many people have to look at the landscape and decide how they will create it. A very integral part of the base game is the landscape. From its texture to its differences and its blandness, it can determine the outcome of someone’s feelings towards the game itself. In order to decide between the path of procedural generation and the path of manual generation, we have started to look over both types to see their pros and cons to it.

Blender

I have been focusing on the side of manually creating a simple landscape, ensuring I don’t spend all of my time on the landscape as to keep the time spent on this and the other, procedurally generated landscape. The first point we can make is the speed at which the landscape as a whole is generated. The landscape is much faster generated, and with better quality, as the procedurally generated coding is implemented. It took a fraction of the time to create the landscape, and the code allows for the size of the world to change while still being procedurally made. On the other hand, the manually-created world looks far worse while taking the same amount of time. To expand the world, one would have to add another grid on and keep randomly making it through their own skill at an even greater amount of time.

While this does make the procedural generation appear to take the high ground, there are more factors to look into. While procedural generation allows for the world to be created faster, the world also remains pretty bland. While I created my world, I was able to create mountains and peaks, flat plains and plateaus. I could even create land bridges and caves if I so wished, along with waterfalls to spice up the landscape. While it does take longer to do, a manually created world can also be filled with so much more life, and a manually generated world allows for thought to be put into each space instead of making it a bland landscape.

Conclusion

From the research we did this week, we can reach a conclusion over how to model things with the usage of procedural generation versus manual creation. While one could reach a conclusion that procedural generation is faster and therefore is the best method for creation, I believe there are spaces for each type of generation. Procedural Generation is good for simpler games that need a simple backdrop and running space while not being a flat plane. Manual creation can be extremely helpful in some more advanced games with a story and given landscape like Skyrim, where different locations scatter the area. A large space is much better filled by a third option, however. There’s room for a hybrid zone of Procedural Generation being used to create the primary landscape while small POIs can be created with manual generation. There is still room for all types of generation in the modern gaming creation industry.