Tuesday, 29 October 2013

Project A: Final Paper

            The development of a video game model/ character was always interesting to me. Finding the work order needed to create something from nothing and bringing it into a game with all of its details. Having to go through 2 to 3 programs can be difficult because they don’t all interrupt the modeling work the same way. 

            I took this task as my first research project. Looking into the different stages required to get an ingame modeled object. Starting with a simple concept art and building upon it in maya. Then further development of the model in mudbox by adding textures and color to it. I couldn’t just go through the pipeline from the start I had to do different tests to see if it was even possible to bring something through to the end.

            So my main idea was to create a living sword that had different aspects of it that would move and bring it life. I did some research on how I wanted to design the sword. I decided to go with a futuristic cyber punk look to it. The reason I did, is because it allowed me to bring my project further if I need too. I would be able to add liquids and light elements that would help push the design further. And I was fascinated by the cyber punk look of things. This research got me a set of images, I wanted to reproduce in my textures. The neon’s with metal bolts look. Once I got my look research done, I did some research on how to model in maya again, because it’s been a while since I used the program. Also did some research on mudbox and how to keep my model from losing its sharp edges, keep the sharp look of a sword. Then I had to understand the way unit work. I did some basic research on how to setup a simple first person controller on a terrain with some lighting.

My first test I decided to start with a simple shape, and see if I was able to pass it through all my programs and make it to the end of the pipeline. I decide to use a simple cube. I started by creating it in maya and then importing it into mudbox. Where I textured it and painted on it. From there I got it into unity using .fbx, and turned out to be the easiest way to carry over files from different programs. That test showed me it was possible to get to the end of the pipeline without many problems.

            I knew I was able to get to the other side of the pipeline, so I decided to work on my concept sword. There was no problems during the modeling portion. Once I got to transfer the entire model into mudbox, I found that many of my edges were stacking on top of each other. It was because I modeled each piece of the sword separately, I knew I would have problems because it wasn’t one big model. I didn’t have much time to texture the entire thing,e a simple cube. I started bythbringing models over to other programs, i n ven useful because i  inclusion of animated textures so I just used the top of the sword. I had a problem with my UVs at this point, I thought I was able to use the mudbox UVs for pretty much anything. But the mudbox UVs are only good for quick modeling jobs and exporting out as .fbx with a high poly count. I wanted to make normal maps for the top of my sword so it wouldn’t have a high poly count. But I wasn’t able to do so with the mudbox UVs. I tried using ptex, but that didn’t help change my UVs. I remade my UVs in maya and tried to import them on my model, but it messed up my paint layers because of the old UVs. I didn’t have time to repaint the model so I exported out as an .fbx and used maya’s reduce poly tool to lower the poly count a good bit. Once I got past that problem the importing the model into unity as .fbx was no problem and lastly I created a simple game scene.

            From this I found out that making your UVs before bringing it into mudbox is the best way to keep your model from keeping a high poly count. It also allows you to extract your maps from your mudbox model a lot easier. I didn’t have much time to see if it was possible to bring a model into unity with normal maps. Also the inclusion of animated textures and liquids were not possible because of the lack of time. Overall I was happy I was able to build something from an idea and bring it into the game scene. I feel this research has proven useful because I plan to enter the video game industry. I already know what to avoid when bringing models over from other programs, and it will save me a lot of time.


Download Links:




Bibliography:

-          Modeling Wire

-          Creating Basic Unity Game:

-          New Features of Mudbox:

-          Normal Maps








No comments:

Post a Comment