Sprint 3 Blog
This third sprint was a hard one for me as a level designer. In discussions with my team we realized that due to the nature of our game and how it is not a genre that has a plethora of games to use as research for the basis of level design we came to the conclusion that before we could realistically design further levels and have a high quality product, we needed playtests to see if design patterns that were already in use were going to work appropriately for us. This meant that while I did some level design such as setting up the level with proper spawn locations, updating prefabs, and fixing visual bugs, I had to find a new set of work to keep me occupied.
This resulted in completing 9 points sprint where I had a much heavier focus on modeling and texturing - a subject that is far outside my wheelhouse in terms of comfort.
Starting the model for our security camera itself I had to come up with a design that felt like it provided a familiar shape for the player with exaggerated proportions due to the camera view of our game. We realized quickly that going for realistic sizing on all aspects of an object would not result in a quality looking product and instead just looked like visual clutter. As a result the security camera received a longer frame with an extremely large focal lens and on/off light.
On the back of the camera you can see the base of the camera where it will be connecting with the wall. This base is set up with a ball and arm joints allowing for what the player sees a seamless pivot point for the camera to rotate on. The light also is set to switch between a red and green emissive material to tell the player whether it is active or not.
Moving onto the laser sensor it was important to make sure that it was very visually distinct from the camera. While they both provide extremely different purposes with regards to tracking thieves throughout the building, we did not want to rely on the visuals of laser and conal vision of the camera alone to sell the difference. A shorter length box, extremely different color composition, and very jagged, pointed corners helped to sell the laser sensor. In order to keep in tune with the red laser it projects, I placed a red glass texture over the area the beam shoots out of.
For the audio sensor I wanted to create a sonar dish that felt similar to what was seem in the deserts of Roswell, New Mexico. The dish shape on the audio sensor helped to make this particular tool look once again incredibly different from its cousins the security camera and the laser sensor.
I chose to use a copper based material to create the line work for the inside of the dish to give it the appearance of using more faces than it actually is. It has the added benefit of helping to break up the dish from the base in a top down view as the black coloring on the audio sensor does not come out well once actually in game (despite the lighting shown above). This is a piece that will be tackled again further down the line for re-texturing once we enter a slower sprint for me again as a level designer where I can go back and touch up on some materials. The audio sensor is also equipped with a ball pivot to ensure that the player does not see and strange clipping as the audio sensor rotates to pick up thief locations.
Finally we have the desk which was actually the hardest model for me. While it started off simple, the extrusions made to the bottom to create the edges of the desk as well as the storage on the sides of the desk proved to have some very difficult UVs to fix. In my attempts to create a minimal poly version of the table by removing every single face the player could not see I wound up somehow creating a table that could not be properly UV'd due to strange vertex connections on the pieces. Instead what I opted into was deleting the sections of the desk that were floating due to removal of all non-seen faces and creating new cubes that would act as the substitute. Getting these cubes into position using the vertex align tool and then once again removing hidden faces from the cubes got me into a much better position for this to be taken into substance painter.
Once within substance painter I had to make sure that I matched the top of the desk with the door provided by our primary modeler, Alex. Once I had a close color I went to work with the wood grain to create a clean looking desk. The slots where the owner would stick their hand to pull forward the storage sections actually have no geometry at all, as that entire section is done via a square-alpha brush and the straight line tool.
For the little bit of level design I was able to do, I expanded upon our initial room prefabs. I had realized that the rooms facing the outside would have windows due to the necessity for thieves, but the rooms inside the building had to be their own prefabs so that rooms inside the building would not have outdoor facing windows.
From here it was as simple as replacing any and all pre-existing rooms that needed to be accounted for to create the appropriate look for the building.
This sprint was a nice change of pace due to the ability to step into another role's shoes for a short period, but I am glad to be able to get back to what I do best. Next sprint I will be analyzing video footage we recorded from our playtests and going over player patterns to improve upon the design of this level where it feels it may be needed. Early viewings show that this level is actually very well balanced for its purpose as a tutorial. This just means I will have more time to work on the upcoming levels, test for balance with regards to cost of security measures, and further advocate for the player in regards to our scoring system.
