This might be a very general question, but I can’t seem to find a definitive answer anywhere. From what I understand, you make the assets in Blender or other software similar to Blender and then import assets into Unreal Engine 4 and essentially that is where the functionality/programming is done. However, what determines the level of detail in terms of graphics. How is it that some guys just have boring graphics with no aesthetic appeal and others have amazing detail and they are both made within the same game engine. I am relatively new to game development and I really only know C++ decently well. I am just wondering does the appearance of graphics come from the engine itself or artistic ability within asset creation? Also, how does the level of graphics within a game relate to overall performance? I know this may seem like a stupid question, but any incites would be greatly appreciated as I am very passionate about learning how of all these dynamics work together.
It’s probably about 70% good art direction, assets, and aesthetics, 15% tech art magic and a good engine, 15% high graphics settings.
For example, the visual difference between Medium and Epic on Fornite night isn’t huge, but there’s a big difference in performance.
A little bit of optimization and planning can go a long way for providing good performance.
As ZacD mentioned above, the most of the graphics is done outside of engine (let’s say Blender).
Better artist, or better performed artistic style = better “aesthetic appeal”
There is no magic (almost) in the engine, that would make ugly art superb looking by programming.
More detailed graphics are more performance demanding:
3D models with higher poly count = more detail
Textures (or sprites) have larger resolution (or less/non compressed).
Programming wise - shaders on materials could be performance heavy.
Basically lowering graphic settings (or quality settings) could mean that you are:
- not using certain shaders or other gfx effects that are performance heavy (physics, etc.)
- using lower res textures (or you are compressing them)
- using LOD models on shorter range (Level of Detail = models with less polys, that are used for distance -> you don’t need many detail for objects far away from camera)
etc. etc.