Week #05 – Silhouettes & Scene Setup
Over the course of the first week, I’ll be focusing my attention to the ideas generation phase, in which I’ll be experimenting with shape language, colour theory, and plenty of other aspects surrounding 2D design. This may include: silhouettes, rough sketches, colour scheme testing, etc.
I began with some basic sketched shapes and used Photoshop’s symmetry feature, and shift-key straight lines to produce some silhouettes. The process was condensed into a GIF as a visual demonstration and I annotated each set accordingly. This method was quick and easy, and required very little effort to complete. However, I wasn’t happy with the majority of these results. Afterward, I moved into creating some rough sketches that gave these designs a bit more depth, but I found this task really difficult to complete as I’m not a great 2D artist. At this stage, I thought it would be a good idea to create some simple blockout shapes to draw over, to make the 2D digital art aspect easier – but I also struggled to produce any great blockouts.
It’s important to be upfront and honest when it comes to time and project management, so I should admit that I did absolutely nothing throughout the last half of this week, and not a lot in the first half week 6 of development. By this point, I had completed all of the research required in task 3 and was struggling with the 2D aspect of task 4. I stepped away and took a break. As the research doesn’t have to be completed until mid-December, I took the time to relax. I’m a little worried about burning out before the final.
Because I’m struggling to produce a concept and design I’m happy with, regarding the jumpship, I plan to work around it for the time being. Week 3 will be dedicated to creating the assets behind the focal point, such as planets, stars, the skybox, and other bits and bobs. This week, I managed to create a setup that I can reuse for each destination I plan to create; I’m very proud of the versatility of these blueprints and materials. This process is broken down and explained in task 4, where I cover all of the exposed parameters that make this setup so adaptable. I’ve included customisation for the cloud formations, speed or rotation and density, as well as a hue shift, speed of rotation, and interchangeable textures for the planet’s surface.
Week #06 – Concept & Scene, Finished
During week 6, I will continue to work around the jumpship to complete the surrounding environment, so that I can solely concentrate on the modelling later on in development. This section of building up the environment will involve creating an interesting space-themed skybox that will surround the scene.
This week, I was able to create a skybox that suited the overall warm colour palette around Mars. In addition, I also created a star material and introduced it to the environment. At the moment, I’m not happy with the result whatsoever and I’ll be spending the next week making considerable revisions to my work. For example, Mars doesn’t have an atmosphere that supports the build up of moisture, therefore I should remove the clouds/atmosphere. Even if Mars did have a cover of clouds, there shouldn’t be such a large gap between them and the planet’s surface – it just looks unnatural. The star is also far too large and draws too much attention.
The last half of week 6 was dedicated to updating all of my written work and experimenting with a final jumpship concept; it’s important to remain up-to-date as and when possible as not to fall behind or risk leaving too much written work to complete toward the end of the assigned time period. As stated above, I spent the last half of week 6 ensuring all written work was up to date and all processes were documented where necessary. I went back to my concept stage and worked on the digital 2D side of the project, which was far from enjoyable. I ended up settling with the design inserted above and it became apparent that I need a lot of practice in this field, especially if I’m to apply to De Montfort University. Nonetheless, I’ll still be uploading this work and its process to my portfolio to demonstrate my progress over the year.
Week #07 – Restart!
As is tradition with every project I start, it’s about time I restart or adjust my proposal to suit my current interests. In this particular scenario, I’ll be spending week 7 using what I learnt while creating my initial scene, to start over and create something better.
I began the new instance by searching for a more realistic texture source, which I discuss in more detail over task 4. Then I looked for a tutorial on advanced materials in Unreal Engine and learnt a completely different way of setting up my planet materials so that they are stored in one graph, rather than multiple, and can be wildly customised in every instance. This includes exposing attributes such as atmosphere opacity, a hue shift/tint, the atmosphere mask, planet surface texture, and fresnel intensity/colour.
I felt that the initial skybox was far too intense, too low-res, and its colours blended in far too well – this had an undesirable effect. This time around I went with a more neutral and subtle appearance that met the planet with great contrast, attempting to create a clear divide between the foreground and background. Despite the texture being 8192 x 8192, it was far too stretched over the inverted sphere and still appeared to be of really poor quality. It would be absurd to increase the resolution any more, not that UE supports files above 8k, so I simply doubled the texture coordinates and rotated the sky sphere so that any distortions weren’t visible from my chosen render angles.
Week #08 – Outer Glow
Toward the end of week 7, I began to experiment with an outer glow effect to surround the planet as if it were lit by a bright star from behind. However, I couldn’t create a result I was happy with. Throughout week 8, I experimented with a plethora of methods that could potentially create the effect I was looking for. I toiled with extra meshes, subtracting and dividing fresnel effects in materials, and even looking into achieving this goal through post processing. Once I felt I had tried hard enough, I eventually turned to the Unreal Engine community on Reddit. Here, it was suggested that I indeed add a new sphere mesh with inverted normals and applied a regular fresnel. I was really close to this answer, and once it was presented to me, I felt I missed the obvious. Outside of wrestling with this ridiculously simple task, I didn’t do a lot of work. With all of this free time over the winter holidays, I should have been able to create a jumpship for the final product. However, I spent too much of that time playing games.
Week #09 – Particle Systems
Over the course of week 9, I will be learning about particle effects through online tutorials while also using prior knowledge of post processing volumes to create lens flares and dirt effects. As planned, I made sure to start Epic Game’s official YouTube tutorial series on understanding the fundamentals of particle systems. This course consists of a total of 2.5 hours of content, which I watched while I worked on other aspects of the project. During this time, I added a lens dirt grunge texture to the post processing volume and played around with some lens flare effects – the results of this can be viewed in plenty of development screenshots. Week 9’s particle system turned out to be a very tricky task that took a lot of time from this project; it was very hands-on and required a lot of effort and attention to finish, despite being such a simple concept. After the star particle system was finished, I moved straight onto the material. Which was the same level of difficulty and took just as much time, unfortunately. Luckily, I was able to sit down with Colin and we broke down an example together. This process gave me a better understanding of how animating and masking textures really works inside of material graphs. Eventually, I layered 3 instances of the same blueprint, with different hue shifts exposed, to create a more dynamic warp drive effect.
Ash taught me how to use animation sequencers in UE4 so that I could export a high quality video of the final products, and Michael showed me how to use Premier Pro to turn the exported frames into a working video. I’ll most likely upload these two clips to YouTube so that I can embed them here and on Artstation.
Level 3 Year 2 