Production Postmortem - Sunshine RV

Hello everyone, welcome to our final production blog for Sunshine RV! In this blog post, I will highlight what went well during the production process, what could be improved, and what I learned from the experience. For those of you who just joined us, welcome! Sunshine RV is a first-person simulator where players manage specific and time-sensitive orders from customers by buying wares and harvesting ingredients to brew various sunshine concoctions. With that, let’s get right into the postmortem.

Statistics

• Throughout the 14-week production period, 1251 points were created.
• 910 of those points were completed
• 341 points were left in the backlog at the end of production

What Went Well

As all or most projects have issues, I can proudly say our team did a great job of completing a well-designed game. I will be speaking about the issues later on, but let’s have a moment to discuss what went right. First and foremost, communication was exponentially perfect. Although communication might be a broad and simple aspect of what went right, it is the most important. Without communication, the team would go downhill, along with its cohesion. We communicated through Discord and had different channels so each discipline could talk amongst themselves about questions, expectations, or just anything related to their field. With that being said, we also had a Google Drive for the team to put their work in. For example, we had 3D models along with it’s textures placed in specific folders so our level designer could find them easily. As the producer, I made sure we had proper scheduling to ensure we were meeting deadlines and expectations. I would send out weekly announcements to the team to make sure they’re doing well and moving tickets over to “To Be Verified.” We went through two-week sprints, so that meant I had to set a deadline so the designer and I could make sprint reviews so our team could see where we’re at and how we’re moving along. Conducting sprint reviews helped the team boost morale because they got to see their work being presented and slowly got to see the end product come alive. Also, having a deadline for the sprint review, assisted me as the project manager in making sure certain tasks were being prioritized so we could have effective builds of the game.

Tagging along, different departments had very few hiccups during production. Specifically, the art and audio departments. There were no major issues, other than cross-departmental collaboration of animated models, but we will get to that shortly. The cause of these departments doing so well was the amount of freedom they had when it came to how work got done, as long as it went with Sunshine RV’s recurring rustic theme. This motivated the ICs to provide a great amount of high-quality work. When it came to sound design, it exceeded expectations because we had a sound designer on the team who solely wanted to do sound design. We were able to get the needed SFX in and done early which left plenty of time to get started on music. By having a sound designer, we didn’t have to use royalty-free music or SFX which was a huge advantage on our end. 

What Went Wrong

I decided that I wanted to take on the role of being the producer. This was my first time leading a group of more than 10 people. It was quite interesting if I’m being honest. There were highs and lows, but all together, it was an experience I will never forget, and the best part, it gets better from here. The first major issue that arose was the implementation of the core game loop. The problem was that we wanted 3D models to be integrated with UI elements, but it proved to be more difficult than expected. So, the designer and I spoke about possible solutions and we decided to adopt a UI-based gameplay, instead of using the 3D assets. We figured that we should go around the problem and not go through it. From there on, the process was fairly smooth for the most part.

There was one issue that lasted through a good amount of production, and that would be animation. Specifically, speaking about character models. We did not have a character modeler on our team and no one had experience with modeling a human. So, we decided to go for a basic human model with low detail. Once the model was done, we took it through the rigging process, but it was far too complicated. The joints were displaced along with the vertices and faces on the model. We went through different iterations of the human models and finally got it to work after two sprints. With that, we wanted to have clothes on the characters but ran into an issue when it came to weight painting. Weight painting is the process of making the clothes flow correctly onto the human model. With all the animation issues, we decided to reduce the scope and get what we had in. So, we decided to texture clothes onto the characters and have only two animations in the game. 

As for level design, if you have seen our game, the player does not wander in an open-world environment. It is a point-click-based game and players solely move around the RV. So, I initially thought, I would have the level designer drag and drop finished props and fill the scene. However, I later discovered that level designers can do much more than have players traverse through levels. They can figure out systems design, specifically speaking about games that have a lot to do with numbers. Our game has numbers that pertain to the cost of trees, glassware, time to harvest, etc. We had to understand the flow state of the game so players don’t feel bored waiting for an order to finish or waiting to harvest a tree. We want to keep players immersed so I specified to our level designer that it’s important to find the correct flow state of the game and that consisted of constant playtesting.

What I Will Change

As it was it was my first time being a producer, managing different disciplines and trying to align their work together was difficult at first, but through time, I was able to manage it and got used to the workflow. The problem, in the beginning, was that we were focusing on different features at the same time, but in reality, we should have focused on one feature before moving on to the next. Production was slow in the beginning, but once we got familiar with how things should operate, it was full steam ahead and production was moving at a steady pace. 

As I previously said, animation was shaky to begin with because we didn’t fully understand the process of it. However, I did sit down with the animators and discuss how to write user stories and essentially understand how the process works. I did not take into account the possible curve balls that might come. For example, the human models have to be done a certain way. So, next time, I will take into consideration the possible issues that may come up and we can take it on as a team and finish it promptly.

One major thing we would do differently would be to have one-on-ones. Since we had 14 weeks to complete a game, we didn’t have the time to do so because we didn’t want to take work time away. If we did have time, we would do so. They’re important because it’s the time to step away from work and get their feedback on how they're doing and what they think we should do differently or what we should keep doing. We want to keep the morale of the team high so we can produce efficient work.

What I Will Do Again

Before the start of production, we assigned roles to each person on the team. Each person was tasked with filling out a spreadsheet for the role they wanted to take on. From there, the designer and I assigned their roles and by doing this, everyone knew the value they add to the team. This also prevents power struggles or confusion at work during development. Along with that, we had team-building exercises so everyone could get to know each other and be comfortable with one another. This created a cohesive working environment and it also encourages cross-departmental collaboration. One of the important aspects of production is to have effective organizational practices so assets don’t get lost or anything valuable in general. This was key for our team and that is something we would do again.

What I Learned

Throughout the whole production process, I learned valuable lessons and gained new skills on top of that. Models need to be done a certain way for animations to be done correctly. Understanding the flow of games needs to be prioritized so players are having fun and immersed the entire time. Most importantly, each team member affects the team. I know I’ve said this many times, but it’s truly important. Keeping the morale of the team high will help boost confidence and will make projects come out better. 

Understanding scope is important as well when it comes to big ideas. We had 14 weeks to make a game, truly not enough time, but understanding the scope, it can be done. As long as we have communication with each other, especially myself, with the designer. Listen to what people have to say because they may have a brilliant idea, and if it’s feasible with the time constraints, it can be done. This was a learning experience and I will be forever grateful to the team I led and to the game we created.

From Bootleg Studios, we thank you for checking out Sunshine RV and we hope you enjoy it!