Week Summary - BA2a Week 11 (3-week development project week 1)

Development Project

For our three-week development project I decided to choose the product design brief, as a way to develop something outside of my usual focus, which is human characters. I want to focus on realistic rendering for this project, using it to further my knowledge of materials and refine my design process.

My brief is to design five items that might appear in a medieval fantasy shop, in a Dungeons and Dragons-type setting. I think one design a week wouldn't push me hard enough, while two might be too much, so I went with five as a compromise.

Ring

The first item I chose is a magical ring. I wasn't sure at first what enchantment to choose, so I started scribbling silhouettes in search of a starting point.

As I went on I found I liked the idea of a ring which used the face of an animal as if it was biting the user's finger. The only problem with this is the fact that the teeth would pierce the skin, which I decided to incorporate into the item, making it a blood magic ring powered by the life of the user. Going on with the silhouettes I found them becoming more and more like a dragon skull, as I liked the sleek and sharp shapes of snake and lizard skulls but also enjoyed the more antelope-like horns.

The next stage for this was to render it fully to show the materials involved, but I struggled with visualising the ring as a 3D shape. Instead, I opted to sculpt the shape in Zbrush to allow me to create an actual 3D shape which I could then work from later for a paintover.

Taking a rough shape from maya, I found it much easier using this method to create a pleasing shape for the ring. Having a 3D model let me see in real time how the changes I was making affected not just the profile silhouette, but how it looked from all angles, allowing me to make the shape interesting from all angles, not just the side. Below are some process shots.

And then the final 3D model.

After this was done I took the above screenshot and used it as a base for a paintover, but first I considered the type of material I would be painting. I felt iron would be a fitting material for a blood-based item, but a worn and tarnished rather than a polished iron, to give it an aged look and give it history. Below are some material studies where I tried to figure out how iron reflects light, what kind of colours it contains, and which combination of brushes gave the most realistic look. In the end, I settled on a pipeline of establishing colours with a basic hard round brush, blending them to a smooth gradient using a soft airbrush, and then adding texture by blending over the top using noisier, more textured brushes.

I'm very pleased with how the final painting turned out, I think there's a believable feel to it even despite the magic effects and the nature of the design. In places there might be a little too much texture, but as a concept piece rather than a final illustration, and in conjunction with the zbrush sculpts I think the design holds up really well. I am however concerned about the amount of time it took me to produce one item, and I think I need to streamline the process if I'm to finish five in the time I have.

Week Summary - Game Jam (Week 10)

(Day 0)

We received the theme and brief and formed our team. We had a meeting to discuss ideas, and decided that a 2D arena fighter-style game would be the best to aim for given the short time we had. Other members of our team all had experience with using 2D pixel art in UE4, so this seemed the best choice for things to run smoothly, and in the spirit of the genre we decided to use the NES colour palette.

The theme for the game came from a joke about a panda cooking game which we decided to run with, diversifying it a little by making it about various different kinds of bear chefs. We finished the day by assigning roles for the project; as I had the least experience in pixels but some in animation, we decided I would do vfx for the project as well as fill in other gaps that might arise as we went on.

Day 1

We decided that as we were making a fighting game we could have each of the five of us design a bear to include. I chose to make my design in pixels as practise for the rest of the work I would be doing.

I then started on the fx, starting with the most essential part which is the water at the bottom of the level, and then moving on to the attack effects

Day 2

Continued with the attack fx, trying to make them look fairly fluid using pixels was a bit of a challenge but I found ways of making it work, and I think the final effects look really great.

After I was done with the fx I moved on to character portraits for the menu, which we split between two of us. I kept them fairly basic in the interests of time and the nature of the pixel look.

I also made a button for the menu and a transition screen, the latter we unfortunately couldn't work into the final game but I think turned out well.

Day 3

With all of the fx done I used this time to add polish to the game, making a splash screen, logo, and some small UI elements, as well as a control screen.

I didn't realise that I'd accidentally mixed up the PS4 and Xbox controllers in the control screen until the submission day when people expressed some confusion when playing. Fortunately the controls are basic enough that people quickly figured out which joystick was the correct one.

Submission day

Although our team didn't win any of the categories, I think we made a very successful game in the time allowed. People seemed to really enjoy the competitive element and the spirit of humour we'd approached the game with, and though the controls could use a little tightening up to feel a little better to play I think we had a pretty strong core gameplay loop which could be expanded on.

Week Summary - BA2a Week 8 (Formative submission week)

Submission feedback

My feedback from the submission this week was mostly positive, which is a great feeling. I did receive some criticism, all of which I agreed with, including reframing my sell sheet to improve the layout and gear it more towards a target audience, and catching up with unfinished work. With regards to the game it was suggested to change the colour of the cleanable objects to make it clearer to the player which ones they were, and to edit some of the ending questions to make them less vague. I'll be making these changes throughout the next week.

Life drawing session

I'm trying to focus more on line weight and more nuanced anatomy in these sessions, as well as thinking more about the weight of the pose and catching myself out on bad habits I may have picked up.

Asset Creation

Continuing my work in substance painter, I've now finished texturing the rest of my model, which now just needs reassembled and taken into UE4.

Personal work

The end of this unit seems like a good time to include some personal projects I've been working on since september, including some commissions and monthly character design challenge entries.

Week Summary - BA2a Week 7

Game Project

As the bulk of the game was all finished last week, this week was more dedicated to small tweaks and polishing. I adjusted some timers according to feedback from playtesters and moved some cameras around to make navigation and general gameplay easier, as well as adding a logo to the start screen and a simple emissive map to the player character to make its eyes glow. I also adjusted some of the values in the post-processing volume (like the universal light level) to make the lighting function better and give the scene a better look.

Moved the starting camera to allow players to see the exit door and added a visual for the AI here

At this point I realised some of my playtesters were having trouble finding the last few objects in the room which they needed to clean to open the door. As well as adding some blocking volumes to stop objects being knocked out of the camera view, I added a function to them using event dispatchers which meant that if the door was clicked without first removing everything in the room, it would cause a point light in the object blueprint to turn on briefly and the amount of objects left to pick up to print to the screen. The blueprint for this as well as the conditions used to block the door opening are below.

I did attempt to add a pause screen to the game, but unfortunately the pause function didn't extend to things like timer widgets, which kept counting down even when the game was paused. I decided that getting this working wasn't a priority, but below is the blueprint section I was using.

Finally, I added the last main gameplay section, which is an interactive menu which asks questions that players have infer the answers to from the environment they just passed through. Due to the basic nature of the objects in this demo, in this case the screens were more of a trial-and-error guessing game, but I would want the final version of the game to let players figure out the answers without having to guess. In this version there are three questions with sets of answers which I arranged on canvas panels in the widget for the sake of keeping things neat. After getting all the correct answers (the game doesn't move on to the next question until the player chooses the correct option) the game displays a final string of dialogue with an end screen, and the demo is over!

I timed several people playing the game to see whether I was hitting the five-minute goal, and found that most people took around 4.5 to 6 minutes, so I think I would say it has a good length of play time. As for feedback, a lot of people seemed to really enjoy playing the game, as well as finding ways to break it, so I would consider this project a success by my own standards in that I managed to make a game that I and others enjoyed.

Asset creation
I started working on my model with substance painter, baking the different parts and painting the textures. Below is an exploded version of the roof section. I had some problems with the bake on this model due to a lack of detail in the low poly which I had to go back and fix, which delayed me a little.

I used two different fill layers for the wood grain here, one vertical and one horizontal, to make sure the grain was always going in the right direction. I may also include some emissive maps later.