I’m two weeks in and I’ve learned something important: Scale. Let me start this over a bit.
I wanted to create a small sample scene, sort of Fallout-esque. I didn’t want to hit on the dead world, radiation theme, but I did want to give the impression of the location being abandoned.
The challenge was to create this using only free things from the Unity Asset Store. There’s quite a bit of stuff there and the majority of it is quality enough for a small, low-light test.
I had the idea of following a butterfly around the scene as it moved from one point of interest to another (I originally started with a cat, but the asset wasn’t free and I wasn’t sure if I was up to animating it).
Out of My Weight Class
Understand, I’ve only had a few tutorials — all very basic stuff. I generally know how to get around in Unity, but not how to handle building of a AAA production video. Idea was to challenge myself to something I clearly couldn’t do and force me to reach beyond the tutorials and fight with the Unity software in ways I wouldn’t ordinarily do.
A Good Start…
I was going well. I built a simple terrain using multiple textures. I added in a water layer and that worked pretty well, if a little awkward and not all that realistic. I pulled in some assets from the store, rocks mostly, and scattered them around the waterline, retexturing the terrain to match the grade of the slope. It was impressive enough that my wife and daughter were impressed when they saw it in progress, astounded that I’d done so much so quickly.
…or so I thought
It was about this time I pulled in my first “real” asset: a building that was to be the 3rd focal point of the movie. It took all of 5 seconds to realize I’d made a huge mistake (huge being the operative word).
The terrain and environment details were far too large, orders-of-magnitude larger than the other assets. I pulled in a sample character model to use as a baseline and the gravel on the ground were the size of the model’s.
At this point you might think “…just scale everything up…” and I considered it. However, I noticed when I took my camera and lights down to more realistic (and usable) sizes, the textures were rough. What were photorealistic assets were muddy and cartoonish when scaled to 100x size. Just blowing things up to match wasn’t going to cut it.
I need to start all over.
I saved the scene here just because it was a weekends work to get it to this point and I do like how it worked out initially. But this week I’ll be rebuilding things, paying far more attention to the scale of my terrain and assets as I work. I plan to create a base terrain and then immediately bring in my assets to get them situated in the scene and start carving out the terrain, the small stream, etc.
I also plan to experiment with the water texture and use another terrain as the water (or, you know, research river models and see if there’s a better way to get water motion).
How I’m Doing
The first two months of this have been a mixed bag in terms of sticking with the plan. I put in many more hours on videos, but learned that I needed to complete an entire course before I went back and did the exercises. I did write, but fell about 100 words short of my goal (unless you count this, and then I’m over goal).
I also spent a lot more time experimenting than previously thought I might. Experimentation in Unity or Maya requires more time for me at this point as I’m not all that familiar with the software.
As for art, I drew nothing. Well, except for the plan for the video, which only counts if I’m being generous. I also didn’t meditate once. I don’t want to be too hard on myself, but I’d like to get more drawing and writing in.
We’ll see how it turns out next update (week or two from now).