Thursday, 4 May 2017

RP4222 Design Practice 2 Submission Post

Welcome to my submission post for the RP4222 Design Practice 2 module. This blog post will act as my submission package for the module and will consist of the following:
  • Presentation screenshots of my scene.
  • Presentation video of my scene.
  • A write-up detailing my design process, the motivation behind my decisions and the steps I took to try and achieve atmosphere in my level.
  • Links to dev diary/reflective diary blog posts about this project. 
  • My learning agreement poster.
Presentation

















Presentation Movie




Learning Agreement



Write-up

Idea Generation / Concept:

After the Design Practice 1 submission in January, I spent a few weeks doing some side projects and dabbling in new software. Then in mid February, I began idea generation for my semester 2 practical work. I put a lot of time and effort into idea generation for this scene. As with the previous environments that I have created during my MA, atmosphere has been the overarching priority and goal in this environment. Here's the criteria that I worked with when generating the idea:

  • The scene needed to have a lot of potential for atmosphere. I had to be able to make good use of all the things that create atmosphere such as lighting, level layout, particle effects, weather etc.
  • The scene/project had to be of an achievable scale so that I could get it all done in time. With my Milecastle environment that I submitted for the first practice module, I ended up having to rush to finish it, so I realised that I needed to keep realistic goals and prioritise and manage my time better on this module.
  • I wanted to go with a historical theme again, as with my other scenes. My reasons behind using historical themes is that I have a personal interest in history which means I am more interested and I care more about what I create. Also, I think that having historical settings can help create atmosphere because it adds a level of relatability and mystique/intrigue. Portraying lesser-known periods of history is also a good way of making a scene interesting and unique.
  • I decided early on that I wanted to make gameplay for the level and to spread this project across semesters 2 and 3, with semester 2 including the creation of all the art assets and environments, and semester 3 being the creation of the gameplay and level design itself. Therefore, I had to choose a theme that I wouldn't mind spending a long time on. I also had to choose something that would give way to interesting gameplay.

With this criteria in mind, I began to generate ideas and soon arrived at the idea for a scene containing a stone circle and taking place in Iron Age/Celtic Britain. This setting and time period would allow some interesting visuals, themes and gameplay. Information about this time is somewhat scarce, which meant I could take some more artistic license with my scene this time around.

At this early stage, I drew up some level layouts and rough concept sketches, conducted extensive research with books and online resources, and modelled a level blockout. This gave me a decent starting point for moving forward with the scene. I made a blog post about this which you can see here.


The story of the scene is as follows: the scene is set on the Celtic festival of Beltaine, which is a precursor to the modern May Day and involves the lighting of fires to mark the beginning of Summer in the Celtic calendar. The player plays as a Druid and has to collect a branch from 3 sacred trees and burn them on the altar in the middle of the stone circle. Upon doing this, dawn breaks over the scene, the rain stops and the level is completed.

Research:

Research is always a very important and ongoing process for me when I create environments. I like to make sure that the contents of my environments make sense. I think that accuracy and believability are very important to create a congruent and consistent experience for the player. My research activities included extensive reading from a variety of sources from books to websites to journals, visiting relevant locations (such as my trip to Ilkley Moor) and watching documentaries. My research covered topics including prehistoric monuments, Iron Age life, druidism and Celtic culture and mythology. This research allowed me to pin down the story of the level and build the level and its features with this in mind.


I've also spent a lot of time researching the creation of atmosphere. Again, I've looked at many different sources for information and opinions to inform my own practice including books, websites and magazines. I've also looked closely at games and analysed the creation of atmosphere in them. For example, I wrote an essay about the creation of atmosphere in Battlefield 1 and in doing so, I gained a lot of new ideas and put them to use in my own level. My weekly Spontaneous Sunday blog posts were also ongoing up until the Easter break and helped to inform my decisions.
Creating the Scene:

Creating this scene has been a very in-depth process that I have documented through regular blog posts. Click here to see an archive of all the blog posts about this project. I have used many features of the Unreal Engine 4 that I had never used before and have learnt so many new skills. I've also started using Substance Painter. Between the end of my last project and now, I have pretty much completely overhauled my 3D asset creation workflow. My old workflow was outdated, inefficient and didn't always yield great results. My current workflow feels a lot more solid and yields better results.

The design process was definitely not a straight road. There was a lot of things I had to do and a lot of crossover, ongoing work and revisiting things, but to cut a long story short, the process of creating this level went more or less as follows:

Idea gen & concepts -> creating level blockout -> creating the landscape and textures -> creating basic trees and rocks, torches, stone circle placeholder model and cliff faces to act as level boundary -> adding water, fog and the Aurora effect -> creating the altar -> creating the 3 sacred trees and the stone circle itself -> creating more varied foliage types such as ferns and flowers -> a huge amount of VFX, post-processing, lighting and optimisation work -> Making the music and audio -> Making the presentation shots/video and submission package.

Here's a series of pictures showing how the stone circle itself developed over the course of the project:








Atmospheric Features:

Models and Textures

I aimed for a realistic art-style in the environment. This meant that I worked without somewhat generous poly-count limits so that the models could look as good as possible (but not impact performance too badly). The sacred trees are focal points of the level so have the highest polycounts, ranging between 3000 and 6000. The stone circle rocks range between 800 and 2000 poly. Meanwhile, the less important or smaller models, such as the foliage, have lower poly counts as they are less likely to be closely inspected by the player and have to be light on performance. I used Maya to model, Mudbox to create high poly's and Substance Painter to create my textures.

I used PBR textures for every texture in the level, which hopefully has helped to achieve a level of realism and quality to the art assets. I also used height maps on the most important assets for extra quality. Using Substance Painter has not only helped me to produce better art, but has also provided me with skills that will hopefully make me more employable as a game artist in the future.



Setting and Theme

I chose the setting and theme because I thought it would make for an interesting and unique level and  good gameplay - I can't think of any games set in Iron Age Britain or in which the player plays as a Druid (in the historical sense rather than fantastical as seen in RPGs). The chance to see a setting that hasn't been portrayed much would add appeal if the level was part of a shipped game. Information about Iron Age/Celtic culture and religion is scarce, so this theme allowed me to take some artistic license and I think it also adds intrigue and mystery to the level, while still having it loosely rooted in reality. The setting is a forest/natural outdoor area with water, trees, rocks and so on. This gave me the opportunity to hone my skills in creating organic and natural shapes.

Lighting

Lighting is often cited as one of the most important features of atmosphere in games, so I knew I had to make the effort to get it right and use it to good effect. The scene takes place at night, so is generally dimly lit, but also features the Aurora Borealis/Northern Lights which act as the main visible light source, so the whole scene is tinted slightly green. Also, if the player gets lost, they can look into the sky for the lights and use them to help re-orientate themselves. I used UE4's curve feature to make the aurora flicker. I used light shaft bloom aka "god rays" to add more life to the scene and quality to the lighting.

The same shot of the environment, lit vs. unlit.
The burning torches add more light to areas that require it, help to highlight important areas and guide the player, and also add more varied colour and contrast in the scene's lighting. When the level is playable, the player character will carry a torch. This will let the player directly affect the lighting themselves - this feature of interactivity with the surroundings well hopefully elevate the atmosphere.

I also made sure to include darker areas without torches and with overhanging trees. This is intended to add variety and mystery. The player is intended to feel cautious and intrigued in these areas which will provide different moods throughout the level as it plays out. Conversely, the stone circle area itself is well-lit so that it feels cordial and is easy to navigate back to.

Particle Effects & Weather

With this scene, I used particle effects more ambitiously than ever before. The particle effects in the environment include: fire particles for the torches, glowing dust particles, mist/fog particles and rain. I used particle effects to add another layer of atmosphere and life to the scene that couldn't be achieved with the use of geometry and lighting alone.

The most prominent particle effect is probably the rain because I think the addition of weather to a game environment affects the atmosphere dramatically. The rain effect consists of falling raindrops and a subtle distortion effect that is meant to mimic the visual distortion caused by rain to both the human eye and camera lenses. I also added puddles to the level by using a terrain texture with low roughness and a flat normal map. When the level has gameplay, the sun will rise when the level is completed. At this point, the rain will stop. This will provide a more meaningful contrast between the environment's two states.

Place, Scale & Focal Points

A sense of place and scale and the presence of focal points/areas of interest can add a lot of atmosphere to an environment by grounding the player within the world and providing a sense of direction. The level's shape is loosely inspired by a Celtic cross, with the player starting at the end of the longest side, the circle in the middle and the 3 trees on the other 3 sides. This makes for a pretty simple and easy-to-navigate level layout.

A screenshot showing the final level layout (unlit and without foliage or fog)
I wanted to make sure that the level has a good sense of scale and that there is allusions to a greater world outside of the level's boundaries. To achieve this, I created a cliff upon which the Oak tree sits. This provides a long sightline over a larger forest area and a river running through it. The river and forest can also be seen from the Hawthorn grove, which sits on the edge of the river. A smaller stream also runs through the level on the approach to the stone circle. This stream can be seen to stretch off around the corner. This suggests to the player that there is more to the game world.

Foliage

A forest/natural environment such of this required a lot of foliage. I used UE4 instanced foliage tool to add grass, plants, trees and flowers to the scene and make a convincing and interesting natural environment. I also used simple wind nodes in the materials of each of the foliage actors to add a more natural and lively effect to the scene. One of my ever-present goals was to create a lively scene, so adding movement was very important. I wanted to have a scene in which something can always be seen to be moving or changing in some way, and the foliage really helped to achieve this.


Post-Processing Effects

I used post-processing effects to influence the atmosphere that is brought about by the level's visuals. I used effects such as colour grading, lens flares and depth of field. I think that post-processing effects can have a big influence on the overall look and feel of a scene, and this project has given me a better handle of how to use them effectively. I tried to create a scene in which the colours really stand out and work together to create interesting visuals.

Audio

I made an original score and ambient SFX track for the scene that I believe captures and aids the atmopshere. I used classical instruments such as harps and horns aswell as synth strings and a variety of percussion in the music. I tried to channel elements of Celtic music. I was inspired by soundtrack music in general and also drew inspiration from the music of games such as The Elder Scrolls and Age of Empires.

Conclusion:

I hope that I have managed to communicate the motivation behind my design decisions in my environment with this write-up. This project has been a very valuable learning experience for me. Not only have I learnt how to use new industry-standard software, but I've also majorly developed my skill in software that I was already using. This project saw me using software tools that I had never used or barely used, such as vertex painting in Maya and Blueprints, curves, particle effects, post-processing and sequencer in UE4. Aside from software skills, I've also learned how to plan and execute an environment art project more efficiently.

The ultimate question is whether or not I have succeeded in creating an atmospheric scene. In my opinion, I have created an atmospheric environment and I'm really happy with how it has turned out. I pulled out all the stops and tried to make use of every tool at my disposal to achieve my goal, while also consistently trying to broaden my mind to new ideas about atmosphere by doing research.

No comments:

Post a Comment