This project was the development of a video game we have named 'Aftershock'. It is a first-person shooter, in which we have recreated the graphics of games played on the PlayStation One. We have also focused on making the gameplay fast-paced and intense, with a heavy focus on movement-based mechanics. In this evaluation, I will be reviewing every area of this project, from the initial proposal and planning to the finished product that we have today. I will be breaking down each stage of development and talking in detail about the areas I believe I was successful, as well as the areas I feel could've been improved upon or avoided.
The time planner was crucial for me and the completion of this project. I believe I stuck very closely to this plan and I think it helped me to get everything done in the time required. In hindsight with the experience I have after development, I would've split my time more to focus on environmental assets and level design, as this is something that took a lot more time than I was anticipating. I was given positive feedback from everyone looking at my proposal, so I never made any changes to the plan. I am still confident with the idea we proposed even after its development. I massively enjoyed creating a movement-based shooter. I think it was a good test of my level design ability, and it was interesting to learn a completely different workflow for texturing models. Having to figure out the best ways to lower poly count on my model while still maintaining good topology and quality was a learning curve that I found very valuable. I also think it was the best balance of my skill because it would have been impossible to create as many high-poly, realistic assets in the time provided.
Our initial planning featured three primary stages. We started the way we usually do by creating a mind map using a website called Coggle.it. This mind map allowed us to visualise all of our ideas in one central location. We spent a couple of hours brainstorming every aspect of our game, with subsections including; game features, user interface, promotion, and level design, just to name a few. Any ideas that we came up with would be put into the Coggle and were available any time we needed to reference them or remind ourselves what was left to do. We then created a to-do list using 'Trello' another free website that allowed us to put our plans onto a timetable. This was to help encourage us to get work finished for deadlines that we set ourselves. The last two things I needed to consider were my role within the project and the reasons behind our decisions for the game. I was already very aware of what my role would be, as the artist of the project, so I also considered the university course I will be studying in my next academic year, to see if I could learn some of these skills ahead of time and implement them into this game.
I believe our mind map was the most valuable planning document we made. It was extremely useful for visualising our plan and I constantly used it to double-check what I had left to work on. I used it as a checklist for my assets, level design, animations, and more. I don't think anything could have been improved about this tool, as it covered everything I needed to work on. Our Trello was also useful but wasn't a resource I used as actively as the mind map. It was great for separating our ideas into more digestible chunks and worked perfectly for setting ourselves deadlines, but I never really felt it was necessary to use. I think if we could change this I would have put more time into focusing on our Trello so that it felt like a better use of my time, rather than just something we were told to use. Writing about the reasoning behind my work was also very valuable to me. It gave me a sense of affirmation and allowed me to feel confident about what we were creating and my reasoning behind things. I constantly doubt myself and my ideas so writing it down and explaining my thoughts behind each decision was massively beneficial to me.
If we could plan again I would probably use Miro instead of Coggle. Miro is the industry standard for mind maps and flowcharts and is generally a much more advanced software. Unfortunately, we only became aware of this a couple of weeks ago so we never had the chance to use it. I think the added functionality of Miro would give us more freedom in our plans by including things like flowcharts and images. I would also find a good way to use Trello that felt effective and felt like a contribution to the work we were doing, rather than just writing down our plans in a different piece of software.
My primary research was conducted around the same time as our initial planning in order to gather audience opinions about our ideas. We also had a few ideas that we were unsure about that we wanted to hear a different opinion on before deciding. We created a survey wanting to use the answers to somewhat sway our final decision before development started. For example, if our target audience had a majority of people that didn't like retro video games, we would have needed to reconsider our art style for the project. Our research was very useful for the development of this game. Primarily because it gave us assurance on the features that we weren't fully set on, and also because it gave us some inspiration and ideas that we hadn't thought of ourselves. One of our questions asked the audience about mechanics they would want to see in a movement shooter, and the reply "mid-air dashing" was an interesting response that we chose to include in the plan.
I don't think the sample size we used for this survey was enough to get an accurate representation of people's opinions. A class of games students isn't your average group of people, so I think it would have been smarter to branch outside of people studying the subject. Unfortunately, there was no real way of reaching out to more people due to our limited resources and finances.
If we could create this survey again, with more adequate resources, I would attempt to share it outside of just my course. 24 responses aren't enough to give us any accurate data so I would aim to reach numbers in the hundreds at least. I think it's valuable to gather opinions from a much more diverse audience and we didn't achieve that with our class which is made up of mostly white, male, avid gamers, between the ages of 18 and 25.
For my secondary research, I primarily looked into things to enhance my level design. I researched map layouts, the positives and negatives of maps in different games, and how I would implement certain features of maps into my levels. I also looked into shape theory in level design, and how using different shapes can affect the players decision making. And finally, I also looked into a range of games that we wanted to reference for certain mechanics and features. I broke down different aspects of each game and explained which features I wanted to use, and which features I would avoid. All the research I conducted was relevant to our project and helped me with development. The main thing I researched that I was unable to implement was the shape theory. I found it difficult in the time we had to include this in my level design. The fact that our game is non-linear and multiplayer made a lot of what I looked into challenging to implement. I think I did enough research to create a good game, but I think I should have looked into colour theory more and tailored it specifically to this project. My research was very impactful on my level design. I attempted to mirror the design of Halo maps and I wouldn't have done this naturally without looking into their levels before the project.
The main thing I discovered in my research was the concept of 3-lane maps. This concept is very clearly implemented in a lot of games that I have played but I had never considered how it was used until my research. Thankfully my experience playing games that include 3-lane maps allowed me to understand how it was implemented, and also how it doesn't work. I also knew that this map layout would only work in a team-based situation, and would be best implemented in an attack/defence game mode.
If I could do my secondary research again, there wouldn't be anything that I would change. I would focus on researching more content, than changing anything I've researched for this project. I'd research more into different map types. I'm sure there are more layouts than just 3-lane maps and I can imagine a lot of these layouts would have been more beneficial and relevant to the type of level I wanted to create. I would also research things like colour theory, and put some more thought into shape theory to see how it could have been better implemented into our game.
Our production started with creating a Unity scene and getting GitHub set up. Since I was working in a team, GitHub was the best option for syncing and sharing our work between computers so we made sure to get this working very early in the project. GitHub allows one of us to make a change inside of our Unity project, and then send the files across to the other. This was a crucial tool in our development and a massively important thing to set up early on.
We then branched off in our work, and I turned my focus toward creating a good character model. I had no experience in character design or art, so I had to research a decent amount into the workflow and techniques used in character modelling. I ended up using two reference images of a male figure and using them to get my proportions correct. I also wanted to try texturing the character I made, but I never got around to this as I was struggling to understand the best way to UV unwrap the character, and I had no idea how to implement my Photoshop image texturing onto a model like my character. After I created my character I then rigged it so that we could use animations. These animations were sourced from Mixamo, a free tool from Adobe, that contains a library full of high quality, motion-captured animations. I tried to keep my rig fairly low quality, using as few bones as possible. I used two bones for both the arms and legs, with 1 bone for the foot, and two for the hand, to make the player holding the gun seem more natural. I initially had way too many triangles in my character, with it being over 1000 when my target was 500. After some rethinking, I managed to lower the poly count down to just under 600 which was good enough for the quality of the character. My research was pretty crucial for my character art. Since I had no experience creating characters, if I'd approached this without looking into other examples and tutorials I would have completely failed.
The next step in my development process was creating the weapon models (Of which we planned to create 14). I wanted to make every weapon before moving on to any environmental assets, so I spent a couple of weeks focusing solely on creating and texturing the guns. My workflow was as follows; Block out the rough shape of each section of the gun. Don't be afraid to split the gun apart into multiple sections. Using the Boolean modifier (A Blender modifier that allows you to cut the shape of another object into a face) create any holes that the weapon needed. Add extra topology in areas where the gun seems too boxy to be realistic. Clean up topology where necessary. Unwrap UVs separating out the sides of the gun to allow image texturing. Try and keep the seems away from the camera where possible. Using Photoshop, split an image of the weapon you're texturing into the main parts and lay them out to take up as much space on the image as possible (The texture map has to be 256x256 resolution). Import texture map into Blender and position my UVs to match up with the texture map. I did research into the best workflow to follow, and from my research, this seemed like the best. I think it was very effective for recreating a PS1 aesthetic and I would use it again.
I then moved on to grey boxing levels. My first level didn't work with the player physics. I don't think the design was necessarily bad, it just had a lot of areas to be improved upon and areas that didn't work with the player physics at all. I moved on to creating a more outdoor level to compensate for this and it worked much better with the player physics so I kept it. My first map was created using the 3-lane map layout and my second map was more of a round map with a clear central point, which I believe was more effective for our game. My research helped me pretty substantially in this area, as I wouldn't have developed as much of an understanding of level design without the research I put in. I wouldn't know about 3-lane maps and I probably wouldn't have known where to start without this research.
After grey boxing, it was time for making environmental assets. These assets were created for my specific locations, for example, the props I wanted to include in the construction site, were things like cinder blocks and concrete pipes. With more time I would have created a lot more assets, and I believe I underestimated the amount of time this asset development would take. I would put much more focus into developing the environment if we started this project again. I didn't research into actually creating the props. I already had a workflow to follow from the guns I made, and it didn't require any further knowledge when it came to making the assets, I just wish I'd known how much time level design takes.
Our feedback was universally pretty positive. We never had anyone telling us what to do and what not to do, so we simply carried on as we were. I didn't make any massive mistakes that feedback could have caught earlier, and the only thing I regret was not starting on my level design earlier, which I don't believe getting feedback would have helped with. It was nice getting reassurance from tutors that my models were looking nice so I knew I didn't need to change up anything in terms of workflow.
I believe I've learned a lot throughout this phase. As I've mentioned I had no prior experience creating characters, so that was an interesting skill to learn. It isn't a mastery of that area of development as I only made one character model and I never even got around to texturing, but it gave me a basic understanding of the workflow to follow. I also didn't know how to texture in a PS1 style. If not for this project I wouldn't have the experience of texturing 3D assets using photoshop, which could be a useful skill for the future. This texturing also gave me a much better understanding of Photoshop as I had very little experience using it before this.
I believe my time management was generally good. I output a lot of work in a fairly short amount of time. Despite not finishing the game, I think we put a lot more work in than our previous final project and I think we achieved a lot more creating this game. Our use of Trello and Coggle was probably the key area that saved us time and kept us on track so I will be using similar applications again in the future. As I've mentioned the only way I would have divided my time better, would be by spending more time on level design and environmental assets. This is a useful thing to have learned and I will keep it in mind for the next game I make.
I believe our final outcome was very positive. We achieved the vision we had at the start of this project and have achieved a lot of our goals. The only targets we failed to achieve were the creation of a trailer for our showcase and creating a single-player game mode. It isn't a problem of time management, I just don't think the game is in the best place visually for a good looking trailer. Instead of this trailer, we have created a gameplay showcase that shows off what we have created in much closer detail, and I think this is a much better way of showcasing our work so I'm not disappointed. For the single-player I also don't think it was a time management issue, we simply set our scope for the game too wide, and decided that it would be a better use of our time to create a high-quality multiplayer mode instead of trying to create everything that we planned in much lower quality due to rushing. Our project does reflect the initial proposal. We wanted to create a fast-paced movement shooter with a PS1 aesthetic which is exactly what we have. I also wanted this project to be a showcase of my skill, and flexibility when it comes to 3D work which it is. I'm happy with what we've made and I don't think I could have improved anything in the time we had.
I learnt a lot about different workflows which I will implement in future projects I'm sure. The biggest lesson I learned is about the challenge of creating high-quality levels. I want to create levels and environmental assets in the future so I believe it was very valuable that I learnt this lesson in a much safer environment.
The feedback we received on our final product was very positive. People universally agreed that the game felt nice to play and that the multiplayer was a good way to heighten the players experience. Our play testers enjoyed the wall-running mechanic, and the ability to emote so these are things we may include in future games. The points to improve on were the lack of polish our game had. This was understandable as we didn't have time to fix any major bugs or errors. People also wanted more functional weapons as most of the weapon models were not there for the player to use in game. Overall we had positive feedback and our players enjoyed their experience massively.
In conclusion, this project has been an amazing success for me. I have learnt a huge amount about different workflows and what the developers and artists of PS1 games would have needed to overcome in order to create their games. I've learnt about level design and some of the theories behind creating levels, the basics of character design and animation, texturing using image textures and photoshop, creating a PS1 aesthetic inside my game, and even new techniques in my modelling software. I'm happy with what I've learnt about low-poly modelling and the skills I've picked up will be useful in future. Now that this project is complete I do want to move away from this style of modelling for the time being. Low-poly has been great for me but I want to look into realistic assets and higher-poly modelling after this project. I have a lot of ideas for creating models that this project has inspired, and I also have a lot of new techniques that I can implement into my art in the future. This project also means I have a second published game, which is something a lot of junior positions at games studios require. I'm very happy with the outcome of this project, and I'm looking forward to taking my newfound knowledge and moving on to bigger things in the future.
Comments