Game Scene Final Project Post:

 Week 3

Slideset available here.

Here I realized I would have to reconstruct my house if I wanted to actually complete this project, which would be a time-suck. I tried to follow along with my original house as best as I could, but the entire reason why I had to do that was because my original house wasn't aligned with my UVs and had some custom pieces I wasn't able to pull out in time. It was ultimately easier to remake the house with the UVed pieces I already had. During this time, I realized I also needed to make a whole other house, but thanks to WoW's design allowing for palette swapping, I realized I had most of the pieces I needed. With some scaling of premade pieces, I was able to create a different house that would look coherent with the game world while also being not too difficult to make.

 

Here's the UV map for the farm; you can tell what parts of it were obtained from the House.

My props were last minute; the only ones I could really feel proud of were the Fence pieces, which I thought I was very smart for creating two different variations that I could construct in Unity. I should've taken the time to make a well; that was a piece I promised to make, and looked interesting.  

Week 2

Finally, my first relatively neat UV map! I had some distortions that I couldn't solve (but probably could've been fixed by deleting the history), and some elements could've been scaled up relative of their size in the scene; this probably would've resulted in much cleaner modeling later on. Once I had this done, I gave everything an ID, but I didn't move onto substance painter; I wanted to get the Waygate and props done at least.

 

 

The waygate I made consciously as a UV set; I only built the pieces that I absolutely needed, and then copied the rest when I was done UVing. This was aided by the Waygate's simple, and obviously mirrored construction. I did, however, make the bad decision of not leaving it as a UV set when it was time to texture; I wasn't too sure on UV stacking and I was a little nervous as to how it would go, so I fully constructed the model before IDing it and sending it to Substance painter. This would prove to be a problem later; since that meant a lot of pieces with different IDs were placed together, I would get numerous baking errors; which could've been solved by just creating a clear UV set with plenty of space for each mesh.

This was also the first model I sent through the entire pipeline; I wanted to know how much help I would need, so I just decided to go and test the entire process out on the Waygate so I wouldn't be surprised later on, since it was the simplest model, the one least likely to crash, and it had the least-terrible geometry.

Aside from baking errors, and issues with getting colliders attached due to not exporting the model, but rather the group the model was in (resulting in some weirdness), it  was a very smooth process. I began work on the props and final building.

Week 1

Initial idea is to do buildings from Warcraft; I'm a fan of the series, and more importantly, the art style. There's a lot that can be said about the 3D modeling for the game, but I'll settle with 'simple and effective'. A few distortions, some twists and lattice-work to add curves and cartoonish exaggerations in shape create so much personality that it's head and shoulders above most MMOs before and after it.

I elected to immediately get started with modeling, and I consciously decided to do the House from Warcraft III first. I decided that this model would probably take the most time; if I completed it, it would be my first modular project, the first project I successfully UVed neatly, and that's a big leap for me. I have not finished any project past the initial geometry stage; so if I could get that done with the house, it'd be something of a small miracle. Everything else can follow after that. My hope is that the other building could be done with assets pulled from the house, and that the Waygate and props are simple enough to be easily created, or alternatively pulled from other sources.

 

Starting out, the house was rough; I tried to be conscious of what pieces I was making, and attempted to solve problems in geometry by using what I already had. If I couldn't, I would make a new piece. During this time, I thought resized meshes still counted as modules and wouldn't need to be UVed separately. In the end, I decided to tear my old house apart to make a UV set of every definable piece, and made a UV map out of that in the following week.