Low-Poly Plant Room

Pandy, Sydney, Julie, Vivian, Tania

DIGF-2004-001

December 7, 2017

 

Concept/What it is:

After being inspired by Haru’s classes of (what some would call) biological design, we decided to create a low-poly computer game based off of a Unity game called Superhot by the Superhot Team. Using the word “game” very loosely, we built an interactive room with fantasy-like plants being the main focus. The player would be able to walk up to the different animated plants and watch them react using colliders. In a way we could call it a prototype for a larger product because expanding this game to be more interactive and scenic is something we all would want do in the future. We wanted to stick with the aesthetic of having everything white and making the interactive objects coloured. Keeping the room simple yet fun, the goal of the space is to create an almost relaxed vibe where players have the time to just explore and wander without the feeling of being rushed. Our overall vision was to make an interactive deliverable that would be extremely interesting to play around with using a VR, if we had the time/money to use a VR device.

Technical Documentation:

  • Equipement
    • Computer or Laptop
  • Software / Platforms
    • Unity Engine
    • Rhinoceros 5
    • Maya 2017
    • Post Processing Stack (Asset store in Unity)

 

Process:

Building the Room:

No one on our team had worked in 3D unity before, some of us only had limited experience with 2D which works in significantly different ways. The first step was to look into how to build models and prefabs and interact with them in the space. Starting of working with one sided planes in order to do a first attempt build of the room, we wanted to bring in all of our assets into the scene in order to test for lighting and interaction.

We started running into problems once we brought the objects from Maya; the objects were skilled scaled to the wrong size and were not grouped individually as one whole piece. This caused problems with placement and and with the effects in the room the animations as well were difficult to apply. We needed an animation controller but with the objects having so many groups it was difficult to sort through them. We initially wanted a trigger animation with a proximity sensor as well as the light trigger but because of the way the objects imported once a collider or a mesh renderer was added to the object the animation became static and no longer applied. The issue was solved by using the prefabs in the system and adding the box collider to those so that they could be triggered.

The windows were originally open, so that there was the view of outside scenery. We used the terrain generator in unit which allows you to do some custom modeling, and mass generates trees and other objects. This idea was scrapped when we downloaded the post processing script to give the lighting a softer, more ethereal feeling. This was when we discovered that my version of unity had not updated properly and was outdated, which was interfering with post processing. At first we were unsure of the source of the error, so we scrapped the outside world in order to focus more on the internal experience of the room.

Once we discovered this error, we focused on trying to uninstall and reinstall, in which we encountered a glitch that was discussed on unity forums which said that there was no current patch for, stopping the program from running. We had to uninstall and reinstall the newest beta version which worked. After this, post processing installed properly but would not run or show in the game build. We decided not to worry about it.

Then the second build involved taking floor prefabs and making a more finalized version of the room with lighting interactions. This time there were less issues that before but the box collider mesh issue was still a problem so the c# script that was attached to the trigger had to be on a premade unity asset.

We learned to make all models including the space in 3D modeling program like blender or Maya, since this would have solved these issues. And to look more heavily into the import of models and their textures and animations. A large portion of project time in unity was spent troubleshooting, re-uploading assets and correcting errors.

 

Lighting in Unity:

We definitely went through some ups and downs when it came to lighting in Unity. We had the vision of realistic light reflecting off of white surfaces and adding intense shadows and glows (in reference to SuperHot). We have succeeded in doing that by the end of the game, but we tried different versions of lighting before we came to that final result. The first thing we tried was normal directional lighting which looked okay but we could not figure out the controls to make it look a little more aesthetically pleasing. So, we switched over to a different technique. We made the planes to be about the size of the windows and added a light material to them. We enabled the static option and fixed the values and colour of the light according to what we desired. The lighting and shadows looked great, but we ran into the problem of not being able to make the planes transparent and therefore not giving them the actual “window” feel. So, we switched back to directional lighting and got the hang of the different controls. We ended up with our desired, glowing effect and it still looked good with the animation lighting. We also tried adjusting the lighting so that it would be able to change overtime as if it were real day/night. Turning on real-time instead of baked lighting apparently does that, but we could not figure out how to adjust the time intervals so that the user would not have to wait a day in real life to see the change in lighting. That is definitely something we would want to look at in the future of this game.

 

Animations (Plants, Furniture, Decoration, etc):

The idea for the room and the furniture was to create low poly items that were also white to fit the contrast between stationary and interactive items. We wanted the furniture to have a simplistic and modern appeal to it. We settled with (mostly) white, geometric furniture to fit this aesthetic. The original idea was to create geometric furniture in Rhino while the organic furniture was to be done in Maya. With the idea of modern, the table and windows were done with Rhino 5 as it has a built in glass texture. The rest of the furniture and decorations were done in Maya.

Unfortunately, problems arose when we tried to export the 3D models. It turns out that the glass material in Rhino 5 was not able to transfer smoothly into Unity. It was also not able to distinguish plastic materials as it turned our white cabinets to a dark black brown. Fortunately, the models that were done in Maya were able to be exported perfectly. Determining the colour for the decorations was a tricky situation. We wanted the plants to be the focal point of the game but we also did not want the decorations to blend with the furniture. We finally decided on a darker version of the colour palette we decided to base the game on. The decorations were done as a darker version because it allow the plants to be emphasised as it allowed the decorations to be dulled out.

 

All the plant assets were created with maya, and used a set colour theme for emphasis. This also gave a visual cue as to what the player could interact with. We stuck to a low poly count when creating our plants for aesthetic reasons, but also to keep modelling simple as we only had a couple weeks to work on it. The animations were done via Maya keyframes. Due to time constraints, the animations were kept to simple rotations and translations. With more time, we would have added more complex movements and interactions. One of the biggest issues with the plants was importing them to Unity; while some worked fine, some did not retain their colours, and some did not retain animations (which is why some animations were missing during our demo). Working with Maya was definitely a learning experience for us. We discovered the limitations of what we could do with our given timeframe, but also learned a lot about the exporting process and which shaders and shapes will not export to Unity. Additionally, we learned a lot about file and asset management and making the overall plants seamless and easy to unpack in another software. The designs of our plants were loosely based on what one might see in real life, with a surreal fantasy twist for interest.

 

Layout of the Interactive Space

screen-shot-2017-12-07-at-11-02-06-pm

 

Examples of Furniture

24899237_1510375385667125_1184426542_n 25105692_1510375389000458_1289445545_n 24899250_1510375382333792_1552450227_n 24898919_1510375402333790_2013627008_n 24992173_1510375392333791_1145243278_n 24891490_1510375395667124_901263108_n

 

Examples of Plants

 

screen-shot-2017-12-07-at-9-28-27-am   screen-shot-2017-12-07-at-9-29-11-am screen-shot-2017-12-07-at-9-29-54-am screen-shot-2017-12-07-at-9-30-56-am screen-shot-2017-12-07-at-9-31-22-am screen-shot-2017-12-07-at-9-31-56-am

 

Final Game Screenshots

process1 room room1 room2 room3 screencap1 screencap2 screencap3 workinprogress

 

Final Game Video (Must click link because it exceeds max. size for blog site):

https://drive.google.com/file/d/1EmM0zMIjfQHS8mnv1Fkov4UM2_g3rps3/view?usp=sharing

 

This entry was posted in Experiment 4. Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.

Post a Comment

You must be logged in to post a comment.

Use of this service is governed by the IT Acceptable Use and Web Technologies policies.
Privacy Notice: It is possible for your name, e-mail address, and/or student/staff/faculty UserID to be publicly revealed if you choose to use OCAD University Blogs.