Using Lumberyard, Amazon’s new game engine, team Workbench is designing and developing a highly-polished vertical slice of a game, which will be using Twitch to allow viewers to directly affect the game. The team aims to build a game in which the Streamer and the Audience members can build a relationship by simultaneously experiencing and interacting with the game, and each other. Another pillar for this project is the formation and creation of formal technical documentation of the development process and lifecycle of creating a game with Lumberyard, with a focus on a Unity developer’s perspective on the experience.

As the lead artist of this team, I am responsible for concept design and 3D modeling tasks, managing pipeline, inter-disciplinary communication, and assuring a coherent art style as well.

Project Type: Academic, Team of 7
Platform: PC, Twitch
Client: Amazon
Developing Software: Amazon Lumberyard
Developing Length: 1 Semester

Week 01 - 02
Our Project Started off by creating branding materials and learning Amazon Lumberyard as a new tool. Although I am not an UI artist, I also helped the UI artist on our team in providing logo options and rapid poster prototyping. I explored a wide range of possibilities relating to Lumberyard, Twitch, and Workbench. After several rounds of iteration, advisers are in favor of the logo I designed. The Workbench logo plays heavily on the concept of tools and connectivity. The gears themselves also bring forward the idea of cooperation and moving forward, both which are themes that are an integral part of our team. I then passed the basic shapes to UI artist to polish and finalize color tone of both the logo and poster.

Week 03
In the Mean time, I organized all artists working closely with programmers to create a “Gold Spike.” It is a small test scene in Lumberyard that contains minimal assets of everything: terrain, lighting, UI, model and animation. And then coded by programmers, the “Gold Spike” helped our team greatly in getting used to working in Lumberyard and setting up a pipeline.

Week 04
As our pitch idea passed by client, we started making sketches and concepts that can demonstrate the game we want to build. I created three concept sketches for the streamer character, Kenhai. Each design has its own unique lore, body shape, and combat style. While our client liked all of them, we agreed on the need to playtest before finalizing the design.
I have also created an environment concept by combining 3D rendering and illustration, which helped us efficiently on generating a visual style of our game. During this process, I imported some low-poly models into Substance Painter and applied some textures.  Then I passed the 3D rendered image into Photoshop, and added more details by illustrating and photobashing. 

Week 05
This week we received Quarters feedback from all faculties, as well as feedback from character playtesting. We recognized that the most popular character is the middle one, which matches how we felt within team. The faculties liked the middle character as well, but they consider her too mechanical, which might be too similar to the environment, a futuristic city in a hostile faction. In this case, we will move forward with the middle character and iterate based on the feedback.

Week 06
This week in art, we created concepts for two levels, and then discussed them with our client. The concepts focused on emphasizing the conflict and tension between the streamer character and the defending forces of the city of Krossis. We are also trying to show a transition from the wilderness to an urban environment, as the player progresses through the levels. Our client liked our concepts very much and provided us with valuable feedback. He especially likes how mundane architecture can be switched into military facilities as an extra defence.

Week 07
For this week, I created 3 iterations on the Kenhai monster we picked (based on playtesting) from last week, and finalized the design with team voting. Based on that and our client's feedback, I kept the mechanical structures on the Kenhai, but used black and red to distinguish Kenhai from other mechs. By adding sharper edges, I strengthened Kenhai’s monster-like representation. Based on this idea, the first iteration removed mechanical elements, but added exo-skeleton armor. The most angular shape within all three iterations made her look even more monster-like. The second iteration kept some of the mechanical structure, emphasized the shield uses neon lights exaggerate contrast. The third iteration comepletely chaged her shape and weapon, in order to make use of non-human silhouette. Based on some more playtesting and client’s suggestion, we choose to move forward with the second option.
I also created 2 concept sketches for Mech 1, the Gattler, and Mech 2, the Ice Box. Both of these choices are based on the previous playtesting results combined with our team’s preferences. They will be made into models in the following week. Each concept has their unique silhouette, following the descriptions on the mech document that used for playtests. The Gattler should be a long-ranged mech that has an extremely high fire-rate. To achieve this, I conceptualized it as moving in 4 legs to surpress backfire, and with two gattling guns on both sides to maximize offensiveness. The Ice Box, on the other hand, is a close-ranged mech spraying freezing gas to the front. In this case, I made it bulky and heavy-armored, so that he can sustain in close combat. 
Additionally, I settled the specific level design, together with our producer and created a 3D prototype of it. We will be playtesting it out once this first environment is playable. However, the actual environment will be further iterated based on the results of playtests.
We created a few small environment asset such as emergency gates and billboards and we are in the process of making more. They are not textured yet, but will in the following weeks. Also we looked into Lumberyard's default assets such as plantations, terrain texture and particle shaders, this will make our asset creating process a lot smoother in the following weeks.

Week 08
As we started to get into rapid development, I have created the model for the Gattler Mech. To visualize the final effects and asking approval from the client and adviser, I applied a layer of basic texture with Substance Painter. I then imported it into Lumberyard, and with Lumberyard’s outstanding rendering capabilities, the model looks great in the engine.Once I completed the model, I passed it to the rigger/animator on our team. By the end of week 8, he was able to finish the rigging process of the Gattler Mech. In the meantime, I also helped setting up a meeting between producer, UI artist and me, the lead artist. In this meeting, we designed a UI wireframe that includes every stage of our vertical slice from start to end, along with the audience game flow. We will follow this wireframe to implement visuals for UI. I also shared some of my thoughts in UI visual style, so that we can try to keep a coherent visual.

Week 09 + Spring Break
Moving into Week 9, I have finished the Ice Box Mech modeling and layered with a basic texture.An environment remapping is also going on in the meantime, where I started to replace cubes and planes from the environment blockout we had before with models in detail. This will be a long but beneficial process.

Week 10

Week 11

Week 12

Week 13

Week 14