For week 2, I was moved to a different group, due to my initial group being too big. I was quickly brought up to speed by Conor, and informed on the group's goal. Our group chose the blacksmith environment, and the concept was to have it be owned by a dwarf blacksmith. The twist being he's also deeply passionate about cooking, so combines both professions into one. He crafts weapons and cooking utensils. He also both sells weapons, and runs a little eatery.
The plan was to have two main rooms of the emvironment: the dining section, and the blacksmithing section. The main centrepiece of the environment would be the furnace itself - as the group áwanted to take a Ghibli-inspired direction with it by incorporating a sea-creature design into it. There were already some inspiration sources such as Breath of the Wild, studio Ghibli, and MediEvil's 2019 remake. As I'm super familiar with the latter, it allowed us to decide on that as our main inspiration for the art style of our assets.
I spent the class looking at concept art from the 2019 remake:
Key features I picked up from looking at the work, (and how it translated into the game models themselves) was how proportions were exaggerated. Not chaotic in its cartoonishness, but instead what I like to call "diet Tim Burton". The original game was inspired by the works of Tim Burton and Danny Elfman to begin with, and its evident in the final product. I characterise the MediEvil 2019 style as "blotchy" with its textures. Not quite painterly, as strokes cant really be seen in the final textures, but the colours aren't 100% flat throughout either. I also observed that, particularly on metallic materials, models have a lighter colour around their edges, acting as a highlight.
After studying the style more, I got to sketching out some rough weaponty concepts. The idea was naturally to have weapons the blacksmith made scattered around the smithing room - including a wall of "easter egg" weaponry; each of us in the group would model their own special weapon for this rack, each referencing a preexisting weapon from a piece of media that we like.
Following the sketches, I took one, (the scythe) and attempted to fully render it in a style similar to the MediEvil 2019 concept art. It was something out of my comfort zone, but despite being a little tricky at first, I ended up really loving the result.
WEEK 3
week 3-4 - concepts, scythe, knightly sword, rack, style guide
Over the next week, I volunteered to create a style guide for the group. As the one with most experience with the MediEvil style, and knowledge on the right terminology needed to outline what elements of a given style we're using to fuel our own, I was eager to take up the task. I thought the strongest way to convey the differences between a photo-realistic style and the stylised one we were aiming for, would be to take a realistic model, and compare it to a model of the same object that instead follows the guidelines I'll be outlining for our own style.
I decided to use the scythe concept for the test model, as it was the sole concept I'd done a rendered depiction of. The cylinder and attachment piece were both created using cylinders; the blade itself was made from a plane that I extruded and adjusted to my liking. The fun part came with the texturing itself. Figuring out how to achieve a blotched effect was really fun - it just came down to having a black mask with a dripping rust generator and blur scope filter applied. For edge highlights I used the metal edge generator with a blur scope filter. As the early few models didn't have great UVs, I would alternate between metal edge, mask builder, or curvature. Once I'd learnt how to UV efficiently and had went back over the previous models, I was able to refine the edge highlights. Using the UV border distance generator enabled this process to be much more effective.
I followed a tutorial to achieve a basic stylised wood texture, before editing it myself to the desired effect. It was mainly achieved by using grunge dot textures and applying a directional blur on them. I'd then toy about with the height levels in order to achieve a wood grain effect. Following this, I used a mask builder with the blur filter to create a thicker, more blurred edge highlight.
After I'd created this test asset, I began writing up the style guide. I looked at existing examples to help me see how to organise my own, and what kind of language was used. I started the style guide off by outlining the general style we were aiming for. I used the term "diluted Tim Burton" to describe ours as whilst MediEvil was heavily inspired by the works of Tim Burton, it can be observed that the edges and proportions in MediEvil 2019 aren't as harsh or sharp - not as overtly gothic. In the same sense, the MediEvil 2019 style isn't overtly cartoonish and exaggerated either. Its by no means realistic, but is neither comparable to the levels of cartoonishness that styles like rubberhouse would show. Following this, I outlined a step-by-step texturing guide using SUbstance Painter. I outlined the methods I'd used for the test asset, including the layer types, generators and filters used. I concluded the guide with a side-by-side of a photo-realistic scythe model with the one I'd made, to highlight the differences. I additionally included other examples of MediEvil 2019 concept art that really exemplified the style we were aiming for.
My next task to complete for the next week was conceptual art for the following items:
Weapons rack
Fuller tool
Knightly sword
Bastard sword
Axe
Morning star
Mace
Zweihander
Great sword
While I wanted to have detailed and coloured concepts, I knew aiming for fully rendered concepts, (like the scythe made earlier) would be too time-consuming. Instead, I kept my concept style to rough but detailed sketches, base colours, and some basic shading and highlights.
For the first few concepts, I wanted to include nautical details where I could, since they were already being included in the axolotl/catfish furnace itself. I was later told that it wasn't going to be such a key theme, so the later concepts don't have any of these nautical elements incorporated. I paid attention to the MediEvil 2019 concept art, to see how proportions and colours were exaggerated. With the swords, it became a little challenging to create concepts that weren't too similar to each other, but I believe they ended up being distinctive enough. The axe and morning star concepts in particular were really enjoyable to play about with - my favourite out of all the concepts being the morning star concept with the handle that resembles an arm and hand.
I asked the other group members to vote on which concept for each weapon they preferred, but I ended up having to choose myself.
With the concepts decided upon, I began working down through the list of weapon assets I'd been assigned to.
WEEK 5
For the knives, the process was simpler than I'd imagined it would be. I simply took a plane, then used the poly build tool to form the shape I'd created in the concept art. I then repeated this process for each of the 6 knives. Next, I selected the faces that formed the handles of each knife, and separated them by selection. Using the solidify modifiers, I then gave the blades and handles the appropriate thicknesses. The wraps were created with the same process that I'd used for the weaponry thus far.
The jars were really simple to model - just two cylinders for the base and the lid, a cylinder for the handle, and a torus for the rim of the lid. I imported the nuts/screws I'd created for the weapons rack to use as attachments for the handle, too. I made use of the bevel modifiers to keep the edges on the body of the jar a little softer. The process for creating the second jar was just a matter of altering the model of the first jar to match the concept art for the second. For the texturing, I took the smart materials I'd used for the knives and just adjusted them to my liking. I darkened the colours to match the concepts, and added a rust effect. This was achieved by using a dripping rust generator on a dark burgundy fill layer. I played about with the settings - including the height levels - before adding the blur scope filter. I also added a position generator on a fill layer of the same colour as before, so to create a sort of burnt effect on the bottom of the jars - to mimic how pots and pans get burnt after prolonged periods of being over an open stove flame. I was super pleased with the overall effect, and saved it as a smart material for later use.
The pot involved a similar sort of process to the jars; cylinders for the base and lid - this time with the edges along the top extruded to create a knob for the lid itself. I used a plane with the poly build tool and a solidify modifier to create the handle. Initially, I tried booleaning a cylinder with the handle to create the hole of the handle, but this messed with the topology in a way that confused me. Instead, I just altered the faces of the handle itself until I got the effect I wanted. Additionally, I used two toruses to create the loops that I'd added as details in the concepts. The texturing was just a matter of using the smart material that i'd created from the jar textures. I simply altered the settings of each layer to suit the pot, and that was that.
The kettle was also quite similar to model. The main differences was creating the spout of the kettle, and the handle with its wraps. The spout was a simple matter of extruding edge loops in the desired area. To create the wrapes, I created edge loops on the handle, then selected the faces that would form the wraps. I then separated by selection and used a solidify modifier. To create the little knob at the top of the lid, I simply used an icosphere. I used the jar smart materials for texturing, as well as the smart material I'd created for the wraps I used with the axe's textures.
Most of the components that were needed to model the cauldron were mostly just cylinder meshes. The base cauldron, the rim, the hook, the pole and the top component were all made from cylinder meshes - the handles and top attachment, however, are both made of toruses. I initially tried to model the jagged metal attachment as seen in the concept, but it proved a little difficult to model neatly, and overall wasn't necessary to the entire model. Texturing, again, was just a matter of applying smart materials.
The spoons were both really simple to create. I started off with the heart-shaped spoon. The spoon itself was just a plane with loop cuts and altered vertices. Before solidifying it with a modifier, I created and duplicated a bunch of cylinders in a pattern to create the holes of the spoon. I used the boolean modifier to cut the hole shapes into the spoon, then used the solidify modifier to add the thickness needed for the spoon. To create the dip in the spoon itself, I selected a vertice before moving it with the "proportional editing objects" setting on. The handle was just a matter of a solidified plane after being edited with the poly build tool. A cylinder was then booleaned with the handle to create its hole. For the second, circular spoon, the spoon itself was just a modified cylinder: modified to create a slight rim. The holes itself were also booleaned in different positions. Texturing, again, was just a matter of applying the jar smart materials and editing to suit. The edge highlights were a bit scuffed around the hole areas of the handle, but I was able to work around this.
