Production | Sprint One

 Starting the Production

    Following the sprint kickoff and card assignments I took to finding references for the assets I was tasked with making: a walk signal, concrete barrier, and newspaper dispenser.

    The best and most accurate way to find reference was by looking at the city the production is set in being Boston. A simple look at downtown Boston and I found quite a few references for a walk signal. Digitally traversing the city also allowed for gauging the scale relative to people.

The latter parts of reference searching were less unique. There was a lack of newspaper dispensers throughout downtown Boston so Google images sufficed. I wanted to generate a variant dispenser so in addition to the metal dispenser I opted to create a plastic dispenser as well.

    The last reference was a concrete barrier that again was pulled from Google. There was a slightly more complex reference found but my designer favored the more simplistic reference. Less complex and less polys means less issues.

    Each reference was provided to my designer and producer respectively to ascertain whether or not these were sufficient to begin modeling assets off of. Satisfying the standards of my superiors I set off to begin my modeling.

    

    Knowing my references I began the process of modeling the assets. In this case I opted to model from reference instead of using image planes in Maya. To ensure I was modeling with some semblance of scale I used a human model from the content browser to scale and model appropriately. 

    Going in reverse, I modeled the concrete barrier quickly and simply. Making a damaged version was as easy as duplicating and moving the vertexes. I prefer to make one complete mesh rather than multiple objects. The copper rebar in the barrier was an eight sided cylinder that was cut and merged with the main base. That way I was left with one solid mesh.

    Next came the newspaper dispensers. The plastic variant on the left was quite simple. A cube scaled down squat was then extruded up from the top face. I hollowed out the inside and made a separate door object. It was styled after the green dispenser featured in the reference. The metal variant was the harder of the two. What gave me the most trouble was double extrusions causing manifold issues. This gave me the opportunity to figure out how to swiftly deal with these type of issues. The door on this variant was a separate object as well. Any of the references sufficed for the inspiration behind this asset.

    Lastly came the walk signal. This came from scrolling back and forth on the street view of Google maps in downtown Boston to get all around views of the reference. Much the former I started with a cube and extruded up, adjusting the vertexes as needed. To achieve the cylindrical pole I chamfered the top vertex and extruded that face up. To ensure I had as few tris as possible I deleted every other edge on the top face. The signal boxes are separate objects that can be removed as need be. The damaged version was added to ensure that there was a large breadth of assets. It was made simply by duplicating the mesh and adjusting the vertexes through rotation to achieve a bent look. I thought to remove the bases but the bottom may be made visible.

    After sharing the models with my leads I was approved to move on to UV'ing them. I was instructed to take these UV's and put them all within the zero to one space and assigned the same material of "Props_Texture_Set_Two" as a lambert. The process was fairly straight forward and the only issue that arose was the non-manifold geometry error. It lead to me losing the ability to properly unfold the UV shells and there being odd cuts. These issues were quickly resolved by deleting extra faces and edges. The downside I found UV'ing like this was when it came to readjusting all the shells to fit in the zero to one space. I feared losing texel density in the larger props so I scaled down areas that wouldn't normally be seen and objects that would retain less details anyway.

    Moving on to texturing I was able to approach this with creative license. Starting with the walk signal. I really wanted to capture the black painted metal being eroded, rusted, and chipped away by the passage of time. Small PNGs were used to make the sign and button on the pole.  I struggled at first finding a healthy balance between painted metal, rusted metal, and exposed metal. A large portion of time was dedicated to moving sliders and parameters around to achieve the desired look. Many smart masks were employed to hit the mark.

    The signal box was the easiest by applying a dirty glass material to the screen portion and adjusting the slider to look appropriate.

    The newspaper dispensers were both fairly simple. I enjoy the contrast between red and blue so those were the base colors that were selected. I utilized the Boston Globe newspaper motif to make the two assets more Boston-ie. However, my favorite aspect of all happens to be the spray painted "FAKE" at the bottom of the metal dispenser. I thought it fun and indicative of the view of the media at present. The plastic variant gave me some issues in regards to the bottom surface being far too reflective for my liking.

    Continuing the cycle of first to last and last first I end on the concrete barriers. This was the only shell to be slightly scaled down to make space for the other, more detailed meshes. Seeing as how this going to be nothing more than stone with a smart mask on it I didn't fret the texel density for this mesh dropping lower than the others.

    These texture assets were presented to my leads who approved my work allowing me to move on to the final step, implementing assets into Unity.

    This step was a bit rocky at first as I haven't had such extensive experience with Unity. After a quick walk through on the desired workflow I exported my models from Maya to Unity. There I made the appropriate file structure and proceeded to apply the materials to the meshes and creating prefabs to export.

    I ensured that the objects were zeroed and that they scaled appropriately to our scale reference. I made sure that these were satisfactory prefabs and with the approval of my producer the prefab folder was exported and submitted to our production Box.

    Being my first time in this version of 495 I found the workload and stress to be healthy. It has affirmed my workflow and techniques as skilled, timely, and thoughtful. The feedback and information received from my colleagues have been helpful and advanced my growth. I never expected I would be able to model, UV, texture, implement, and export at such a fast pace. Low and behold I have, for now. I plan to keep my work ethic up and continue to have a good grasp of deadlines and workloads.