Common employability requirements for my chosen focus?

 What features would be beneficial for a junior environment/shader/material artist?

Over the past few months I have had to put a lot of thought into what kind of job I would be interested in leading into the job search I will likely have to do next year, and originally I thought that this would be purely technical art. After a bit of introspection and a bit of a mental crisis over the last month or so I realised that I was more interested in the node based systems to make materials and shaders, and the presentation elements in engine that tend to make the project come alive. Things like particles and shaders are the easiest for me to get stuck into and enjoy the most, which I think is a desirable factor already - I have to be passionate and interested in the field.

After doing some quick hunting on sites like LinkedIn, I found a couple of job listings that I found interesting compiled them below:

Material Artist:

Responsibilities:

• Be able to make photoreal or stylised textures and materials
• Balance visual quality and technical consistency for real time applications
• keep consistency with the visual artistic style defined for the game
• Assist in R&D of new techniques and help tech artists create tools
• Communicate and receive critique from lead artists and directors

Qualifications/Skills:

• Digital portfolio of previous high quality work
• Sound knowledge of the texturing pipeline used in games or VFX
• Experience with substance designer, substance painter, 3DSMax/Maya or equivalent, Photoshop, Zbrush
• Good understating of PBR materials and real time engines (unreal, unity, etc)
• experience of asset creation (both tiled and baked assets)
• Good observation/research skills of what real world materials look like (geology, rock formations, terrain, urban planning etc)
• Strong understanding of art fundamentals (colour theory, composition, lighting for mood)

Extras:

• Interest in real time shader creation
• interest in lighting
• Interests in latest advancements in game graphics and visual trends
• Photogrammetry experience
• Organisation, time management, presentation and communication skills
• strong English skills (written and verbal)

 Texturing and showing the role of the animal in game/world

Animal Role:

Before texturing, it was important to establish what the role of this saddled animal would be. By establishing this it would allow me to hone in on an appropriate mood and add details that would suit the animal's purpose. 

My Shiba Inu dog is meant to act as a companion to whatever player it may come across; A lot of my inspiration on my initial design was oriented around people dressing up their own pets so I thought keeping along that theme would be both more relatable and realistic. More specifically I chose saddle elements that were reflective of a helpful camping companion, keeping a sleeping bag role on its back as well as useful storage pouches that would help the player store consumables perhaps. I thought that his scarf would be a cute edition to show he was ready for whatever conditions they were going into, and my plans for the bags would be that they would be adorned with small badges and patches to add personality. Most of these would be patches added on through texturing since the small badges might have been much more expensive overall, and I thought that having some rough stitching would be a cosy detail over the clean edges of all of them being badges.

Texturing:

I wanted most of the textures to be simple colour blocks since the dog itself is quite a small animal, and I thought having larger shapes be readable would be less chaotic. I kept the fur texture on the sheets minimal since I wanted to experiment with a fur shader to artificially add depth; Shiba's don't have large pronounced waves of fur, their coats are dense and fluffy so a shader to keep the silhouette felt most appropriate.

Base colours applied in max

Quick render in unreal

Badges Close Up

 Animal retop + differences between static vs deforming mesh

Sculpt edits:

After receiving some critique from lecturers on my sculpt, it was pointed out to me that the yes on my dog were too small. Especially for an animal like a dog where humans can gauge a lot of emotions from the eyes and eyebrows, making this area read properly was very important for the cuteness factor. This was the final sculpt vs before:

Before crit

After crit

Difference between static and deforming meshes:

When retopping my mesh for my animal, it was important to keep in mind the function of the mesh. The mesh had to be structure so that it would deform well when moving joints such as legs, the tail or the head of the animal, so I had to keep a couple of features in mind:

• Deforming meshes need a primarily quadded mesh to deform evenly, so most shapes in the mesh need to be 4 sided polys. 
• These Quads should form loops to make the mesh keep forms better. It also means that the mesh is easier to manage while retopping.
• The mesh needs very few triangles, limited primarily to areas of the mesh that wont deform much (like flat spaces or areas necessary to keep the loops in place)

These features differ from static or non-deforming meshes since they can implement triangles in their geometry easier. By not having these extra loops and geometry, these assets tend to be cheaper in vertice cost compared to deforming meshes.

Retop of my Shiba Inu:

I attempted the retop process initially and it was commented on that I was not including enough loops around deforming areas, so the mesh would not deform very well overall. I redid most of my initial attempt within the day, and this resulted in the final result below:

Facial geometry

Body Geometry

I made sure when I was retopping to make use of the conform tool and the relax tool to create a more even geometry density, since that also would help the model if I ever rigged it and would allow the silhouette to be more consistent to my sculpt overall. I made a very basic rig for my dog using the CAT rig method, and checked how my geometry handled the basic walk cycle provided in 3ds MAX 2020. I think overall the mesh deformed properly, especially considering at this stage I did not do any custom weight painting for my rig.

Boss Arena + Monster - High Concept + Sketching

 Boss Arena + Monster - High Concept + Sketching

To begin this next set of projects, I wanted to outline a starting point for my concept. Since they are meant to be complimentary to each other, I think it is important to create a strong baseline to inform some basic sketching and reference gathering.

My initial thoughts were "what games do I want to use as reference?" - to which Dark Souls and Hollow Knight were in the forefront of my mind. I also remembered that I had created a phot bashed character with these inspirations at the beginning of the year, and I considered expanding on that universe since I already had a foot in the door. The more I thought about it however I realized that the Gothic medieval theme was very done (it was the main inspiration for the rest of my class let alone a cautionary search on artstation), so I went back to the drawing board. 

I thought about an interesting gameplay mechanic of a character that summons attacks from markings and drawings on walls, and then thought of a cave painting idea. However when I started looking at interesting caves, I found this type of cave called Cenotes which are these water filled sink holes in Mexico. Even looking at some pictures of them on google some had the subtitle "boss arena" so I thought it was fate and used that as my primary inspiration source. 

I still wanted to keep the idea of a character that is an artist of sorts as it feels like an interesting idea, and reminded me of the character Grimstroke from DOTA 2 which also just happens to be my main (who basically was a "chosen one" who corrupted a ritual, stole the souls of his village on accident, and now has a powerful ink and paintbrush as his method of attack to paraphrase. He also launches the soul of his dead wife to silence enemies so he is a lovely character all round - His backstory if you are curious can be found on the DOTA 2 Wiki: https://dota2.fandom.com/wiki/Grimstroke/Lore).

Grimstroke - DOTA 2

Now it is fairly obvious that Grimstroke's design is heavily inspired by Japan, and so I was planning on giving the character's base concept a Mesoamerican twist. If I still wanted to use the Ink idea, I would have to figure out how prominent in the scene it would be and how it got there. I also at this point gave my project the working title of "Inkan" but this is very likely to change.

I started looking at the gods and goddesses that are documented from Aztec, Mayan and Incan culture, with a particular emphasis on the Aztec gods; The geographical location of the Cenotes is in Mexico rather that northern south America. I then quickly realized there were a LOT of named gods and goddesses, so I refined my search to specifically caves, waters and artist gods and goddesses. I found a pair of Goddesses with descriptors on their appearance that I thought would help inform my design for my boss, Chalchiuhtlicue the waters, rivers and lakes goddess described to have skirts that turn into streams and is described to be the source of the Aztec deluge (think Noah's ark, world floods) myth, and Xochiquetzal the fertility and arts goddess who is described to be associated with the moon and to dress in elaborate clothing. This pair of goddesses gave me a decent foundation and idea along with Grimstroke to start designing a character, and the Cenote temple they would reside in.

I wanted the Cenote to look like a forgotten temple because I thought that the game could be an exploration game, similar to Uncharted or Tomb Raider, and you happen across these slumbering Gods and Goddesses who are angered that they haven't been worshipped in centuries and thus have forsaken their duties - in this case these Goddesses have weaponized art and corrupted water, and so I created the below sketch page:

Initial Sketch Sheet

I am going to move to blockout the arena design soon, and I will be iterating the design for my boss too. I know that I want her lower half to look like flowing ink, so the majority of the iterations will likely be the outfit and that silhouette using reference of ceremonial garbs.

In this sketch page I also started thinking about how you beat this boss other than "just hit her", and I thought about creating these seed pods growing from the corrupted vines that could be used to negate areas of the arena from ink, like erasing. Another idea came from a chat with Craig, where he asked what the worst thing you can do to an artist is - blinding. I think I may create a sheet exploring a stage system for the boss, the beginning being hostile but calm and considered, and the later stages when you shatter her mask becoming more aggressive and reckless. I will need to make sure that there is enough visual direction to the player that her mask is a point of weakness to attack that causes critical damage.

Saddled Animal Project - New Beginnings

 Saddled Animal Project - New Beginnings

Now that I am back from Christmas with the slice and pumpkin project under my belt, its time to begin the next brief: An anatomic correct animal with either a saddle or saddle bags. I initially read the brief with the intention of making the animal anatomically correct in proportion, but with some different mystical/fantastical style texturing. I had this idea of creating a Kelpie, a water spirit horse from Celtic mythos, and created a mood board of ideas to take to my first session. I knew that this idea was more out there so I did not pin all my hopes on being able to create this animal.

Kelpie Mood board

After speaking to Kat and Pat, this idea was quickly thrown out as it was too fantastical for the brief. This left me in a awkward spot and I had a lot of different ideas that were not sticking with me - I thought about making a normal horse and then making the texture in my own time afterwards but I did not think I would want to approach this project again at the end of the term. I also considered making a fox mount, but because of the restrictions on budget fur cards were going to be limited and I did not think I could do a fox justice. I also considered a bamboo-delivering red panda, but that stayed at the back of my mind too. In the end I settled half heartedly on making a otter with some scuba gear, and set about modelling him using a zphere method, then applying an adaptive skin and doing sculpting passes on top afterwards.

Otter Sculpting Process

Sculpting the otter was a long process and I was struggling a lot with how to characterise or stylise the fur to appear like an otter. In all my references the otters had this pointy fur when wet but when dry they had a soft, short fur all over that made it very hard to model. I found an alpha of hair spikes online that I used on the model, and then used the weave brush with a split line alpha to add more detail to the spikes. The tufts around the face were made using the snake hook brush, smoothing it slightly and then adding texture on top. In hindsight I really should have used the pinch tool to make these tufts go to a point, but this result looked very cuddly and fluffy. I kept running into this recurring problem; If I did not add a texture, the model looked incomplete, and if I added a texture the result looked messy and chaotic. In the end I settled for the final sculpt result, but I still was ready to change it later if I came up with a better solution. 

At this point I started to seriously consider how I was going to apply saddle bags or a saddle to this otter, and I hit a brick wall. I could not figure out a way of making this creature carry either the planned air tanks or giving him a fantasy profession etc - and I had a panic. I had sculpted the otter slowly over the course of two weeks since I had been ill which caused me to deal with dizzy spells, light sensitivity and fatigue, and sitting at my tablet at home to sculpt was quite taxing. I took a step back, and I had some medication now at this point to assist and I was feeling a lot better because of it. I sort of went back to the drawing board, analysing my mood board and collected reference on other animals that might help. I noticed that there was instances of water rescue dogs with these saddle bags and equipment to save people in danger, so I started looking at these references closer. I soon realised that I was quite disenchanted with the idea of this otter; a combo of the frustration of being able to work only in short stints, with the fur texture driving me around the bend, and now I was completely lost on how or what to dress the otter in, I took the decision to look at other animals I wanted to sculpt. 

Inspiration struck when I remembered I had gotten a birthday present that could work for a starting point; I had gotten a Nintendo Switch shell, with a little chibi Shiba Inu with a sushi wrap around it that looked suspiciously like a outfit. It was stupid - but hey I guess inspiration comes from some weird sources sometimes, and I clung to it for life.

Shiba + Otter Ref Board

Switch Case Design

Dogs in general get dressed up all the time, and Shiba's are no exception. I was able to find a plethora of references to use to create a basic idea including a reference of a Shiba with a sleeping roll and little bags on its back, and so I began modelling again - this time a woodland adventurer Shiba Inu. I was even able to quickly sketch an idea that I would use to make the saddle bags, so I knew ahead of time how this would look at a final stage.

Shiba Saddle Bag Breakdown

Shiba Inu Sculpting Process

Final Sculpt

This model came together really fast - within 2 days of work to get all of high poly work done. I will likely make some changes, but I want to attempt retopping what I do have to make sure that the budgets can work for this many props. I am really excited to be able to texture this good boi, and I plan to make a parallax occlusion fur material to get the fluffy texture on the model. I am really happy I switched animal to this, however if I want to I could perhaps return to the otter at a later date to polish and upload as just a sculpt on its own.

Backyard Slice - Final Presentation and Decals

 Backyard Slice - Final Presentation and Decals

I am now in the final stretch of the project, and I am excited to show my results after what has been a massive learning and practice experience in this project.

Decal Generation:

The final asset I plan to create is a couple of decals for some posters, as I wanted to add some clutter to the walls in and around the shrine to break up the plaster and the tiling shutter textures. I wanted these posters to be very painterly and without discernible detail - they are not meant to be eye catching or focal points in the scene, and so I went to the internet to find some basic Japanese food posters and vintage travel posters.

After collecting a few, I applied a harsh cut-out and poster filter to blur the images into just the outlines and colours of the posters. I also colour shifted the posters to fit the scene more, adding more pink and copper oxide green to the posters. The final decal albedos can be seen below:

Poster collage decal albedo

Single poster decal albedo

On top of these albedos, I also generated a basic roughness map through assigning each poster a random grayscale value, and using then normal filter in photoshop was able to generate a basic normal texture for the posters. All that was left was to apply them around the scene to clutter the walls around the focal points.

Final Presentation:

Below is a series of high resolution screenshots I took of the completed scene, with the decals applied:

Whole Scene View

Shrine from the Bike Perspective

Shrine and Statue Closeup

I relit the scene since my formative following the advice I received from Kat and Pat - which I think did a great job at accentuating the work I have done to create all the elements of my scene. I also added a couple of extra features, such as adding a simple wind to the lanterns and the foliage so that the scene has some movement, and using the same firefly particle I have used in pretty much every project at this point (Tutorial here) to create a simple but effective dust around the lanterns and foliage. I also vertex painted the floor to add some puddles and dirt variation, which helps break up the obvious tiling repetition I was struggling with before.

I am really happy with the overall result of this project and I think it creates a really comforting atmosphere. If I had more time, I would have liked to create and alternative lighting setup - perhaps with some rain weather system to add some variation. 

Backyard Slice - Related Industry Topics

 Backyard Slice - Related Industry Topics

In this post, I will be tackling the required topics within my brief for the backyard slice project:

1. What is the purpose of highly reusable textures and 3D assets in video game environment art creation? 
2. Investigate techniques and tools that can be utilised to create 3D environment assets faster, and ways in which assets and textures can be given variety to hide obvious repetition.

I want to answer these in a post together since they are both linked together.

1:  What is the purpose of highly reusable textures and 3D assets in video game environment art creation? 

A good starting point to answer this question is to discuss why repetition is beneficial, and this can be summed up as it is extremely performative. Performance in games is a absolutely key aspect; if a player enters a city and the hardware cannot handle a decent frame rate the game is jarring and the player's immersion is broken. The player is incredibly unlikely to touch the game again until they hear the problem is fixed, which is a failed game. 

Draw calls are the key component to poor performance that modular/highly reusable assets can help circumvent. A draw call in a very simple term is a command from the engine to render an object; an object will have a call for the mesh, and then one per material on the object. for example, if you have 5 objects using 1 material, this will still be 10 draw calls. Engines are usually incredibly intelligent at noticing when a material and a object have been copied or "instanced" in the scene multiple times, and thus they batch together these calls to reduce the total amount, making highly reusable assets infinitely more useful than unique assets. 

Another aspect of the use of repeatable textures and assets is that it reduces production time for the artists by allowing them to focus on simple asset types (such as a wall, a beam, a window frame), and allows level builders to have a kit they are comfortable with to produce lots of different building configurations for instance. Here are a couple of examples of projects that have made excellent use of the modular kits:

17th-century-roman-modular-environment-in-ue4

Fantasy Town Modular Kit Breakdown

Within textures, there are a couple of method to produce highly reusable assets, and these are through trim sheets and tiling textures. Tiling textures excel in areas of flat space but require some work to reduce the obvious repetition. Trim sheets are great for assets like wooden trims or simple beams, and normally tile horizontally to be able to hide some repetition also. These textures can be used across a variety of assets and help produce a consistent scene, which when considered with Texel density of the spaces can produce a harmonious environment.

2a: What techniques can be used to create 3D environments faster?

I briefly touched on this before when discussing how reusable assets shorten production time, but what other tools are there to help artists create more efficiently? When modelling these assets, 3D artists will have access to a plethora of modifiers and plug ins that can procedurally or dynamically modify a created mesh. Modifiers in max that are particularly useful for instance are:

• Symmetry: this is exceptionally useful when retopping a character for instance since most characters at least have some degree of symmetry to use (eg. in faces or hands), or when modelling a basic blockout to then add variation to later
• Bend: This allows for easy modification of foliage assets to add a natural curve to a model (Twist also helps here) that can be changed dynamically
• FFD node adjustment: this allows for a whole model to be adjusted if it appears too stiff in a organic way. I used this to add some extra shape to my herbalist house in year 1 for instance.

I haven't looked into plug ins too much, but there are plenty of varied scripts out there that can speed up production time drastically. Companies will usually hire Tech artists or programmers to develop tools for their art teams, which can also speed up time drastically by reducing menial labour. for example, there is a max script that will generate debris in a scene for you after you input some parameters - this debris needs to be optimised to make its way into a game, but it is a great start to cut down time.

2b: What techniques can hide obvious repetition?

Through the backyard slice project, we learnt a couple ways to reduce obvious tiling of our assets. one of these was decals, which act as scene "stickers" to add extra detail to a scene. These can range from being as simple as a concrete pavement seam, to posters as I did on my project. These add areas of interest that can be dynamic and become focal points, which in turn distracts from the obvious repetition of the texture. 

Another method is through vertex painting - which is the method I used to break up my floor texture. This method relies on having extra geometry to an object that can then become a mask for a material in UE4, and is a great tool at giving the artist cheap customisation in engine.

Another method is through world space masking and shaders in unreal; often, a material will be made which will add moss or snow automatically to the top of rocks for instance, which adds a dynamic touch to hide repetition. 

Often, the best materials for tiling are the simplest ones. Textures like bricks or dirt or plaster have simple shapes, and very low texture variation. By limiting features that are noticeable when tiling (such as cracks, rocks, extra dirt), you can make a material more believable.

A modular kit is only ever as good as the pieces you create for it, if you have a limited kit then the seams are going to show more obviously than a detailed kit with many options available.

Pumpkin PBR Refresher Project - Presentation

 Pumpkin PBR Refresher Project - Presentation

I am writing this point quite a bit later on now that I have another project's experience in presenting in unreal, which means that now I can give my previous work justice. I also would now classify myself as confident in designer, so I can make much better materials than just a month ago even. I did an attempt at presentation back around Halloween, with some questionable quality results to show for it. 

The initial attempt

I followed a  tutorial on how to make substance materials - specifically this "Floor with Leaves" tutorial (https://www.youtube.com/watch?v=AMMBK0Xjpf4&t=511s&ab_channel=3dEx) which provided a groundwork into using nodes like the tile sampler node and how to construct elements like leaves using shapes and mirror nodes. 

This original material looked like this:

Original Material + Node Set up

Original Material up close

This material looked good within designer, although I was annoyed at how often the leaves would intersect each other. From what I have now learned I know that to avoid this is through a very laborious process of less dense leaf layers height stacked together, but this gets very intensive and complicated graph wise and just for a presentation stand I was not too concerned.

I then made a simple plane to stick this material on which looked bizarre, but I did not know how to fix this. I would like to preface as I did in my previous post on the tortoise that at this point I believed that I was restricted both in textures AND in tris, so I had to make all my presentation on about 40 tris spare I had managed to scrounge from my tortoise model. I thought I had to keep the presentation limited at a single 1024 sheet, and I made a VERY primitive "trim" material in the end to make layered leaves in a strip at the top of the material that I could use as a border. This masked material was modified to make this combined end result:

Masking step to add the "Bush" element to the texture

This process was still quite amateur at the time - we had not been formally taught designer yet so I was still using this material node in the middle, and I was only just grasping how to output textures. The final result was this, after using a basic "exponential height fog" component and a simple top down lighting set up.

Initial Presentation Attempt

I tried a couple of other things that are slightly noticeable in this image: Light functions and a leaf particle system. I isolated my leaf texture in my designer graph to produce a simple particle texture that I then put in the scene, setting it up by following this tutorial (https://www.youtube.com/watch?v=aJe1ysBGbXs&ab_channel=DeanAshford). The light function was a simple dappled intensity map that I made into a material, set to the light function material mode. I then set this into a spot light and put it above the pumpkin to produce a inconsistent light similar to tree cover. This was a lot more effective in my head, and did not really pan out how I thought it might. I think these light functions would be best used in cases such as stained glass, where you want a very particular shape in light on the floor. It also was not really visible from the camera looking at the tortoise and did not add to the presentation, so I would not implement it again in this project.

As a quick side to this - I posted the picture above of Barty into a discord server called "Experience Points" which has a lot of industry professionals, students and hobbyists, and he actually won the Halloween pumpkin competition they held in the 3D category and I got a free tutorial from it, so that was a real confidence boost!

Fast Forward to Christmas

I now have a separate project's work under my belt, getting me back into prop workflows and improving my artistic eye for what works well in a scene and what doesn't. The presentation I had with Kat and Pat really opened my eyes to how to make a scene work and elevate my project, and this is what I was focussing on:

• Edge dressing/borders
• clean normal maps on designer materials
• Materials that are readable (lighter in value and cleaner in roughness)
• Accent lights to highlight surface qualities
• Not a flat black background, add a simple tone to the back to broaden the scene.

I actually had a issue file wise - initially I did not use OneDrive as a file location on my computer in labs or at home. during each session I would drag and drop files into the drive to back it up because frankly I did not know this was a thing. I found this out a couple weeks into the backyard slice project in fact - and since I am now tackling the presentation away from Leicester while I am home for Christmas I do not have access to my original unreal file or even some of the textures I had saved for it - like the updated "trim" version of the designer file I started. I took this as a blessing in disguise to be honest, and restarted presentation with a fresh view. I knew that I could spend a bit more on making a decent presentation "island" for Barty, and I was actually inspired by a fellow student Iz Green (https://www.artstation.com/artwork/5XQyrz) and her pumpkin project to make a small diorama for him to sit on. I did also consider rigging him for some added posing potential, but since the geometry was not really made for that and I do not have as much time as originally planned I think I will leave him as is.

to begin the rework process I went to my files, and found a very basic version of the original ground file was there that I could use as a starting point. I wanted to add some foliage like small bushes and grass blades to the material, but I was very aware that I did not want aspects of the material reliant heavily on height tessellation. I did not plan on using a displacement method for the floor in ue4, and I did not want this material to appear really flat like my original. The material ended up looking like this:

New Substance Material

I also have learnt now how to publish this material so it could be used within painter, and I planned to use this to my advantage. I made a basic ground plane again with a slight curve in max, and made a simple grass tuft and collection of rocks. These were all unwrapped together, and then imported into painter to make sure these assets could be given more custom treatment. All of these new assets came to about 500 tris in the end, mostly in the grass tuft and the rock which I would instance multiple times in unreal to be budget conscious.

Painter Texturing

Within painter I also added in a translucency to the edge of the island to remove a harsh outline, and I think this really helped soften the whole composition; It helped keep the focus on Barty. I also added more individual strands to each chunky grass blade, which faked detail on each grass tuft and helped immensely at making the model feel softer too. I then imported these all together into unreal, setting up the materials in the basic formation. I added an exponential height fog component again this time basically removing the height fall off to make the colour of the background a consistent light-mid tone orange to complement the colour of the pumpkin. I did not add a particle system in again, partially because I thought the detail of the ground elements was enough interest and also with this orange background they wouldn't show up nearly as obviously. After doing these basic steps, I was happy with the result and chose not to over complicate it. It made Barty look like he was in a warm comfort environment outdoors, and I was really happy with the leap from the first attempt.

Beauty Shots:

Conclusion/final comment:

I am really proud of how I have done with this project. Considering it was just a simple Halloween pumpkin model brief, I think I have done a great job twisting that and making a creature has really served as a rewarding learning experience for the future. There is not much I would change now knowing all the problem solving I did to create this result; I am just happy I persevered and pushed myself where I did. It is actually quite rare that I feel proud of something I have made so I shall enjoy this feeling while it lasts :) 

Backyard Slice - Designer Materials

 Backyard Slice - Designer Materials

Now that the project is in its final stages, I want to breakdown the full catalogue of designer textures I created during this project. I want to preface that before this project I had extremely limited designer experience from only my pumpkin project weeks before, but I have grown to love and prefer designer to create textures. I especially love the publishing feature that lets me send my textures to substance painter, which proved really useful when texturing my shrine to be consistent with the rest of my scene by applying the tiling materials to it. 

Using the isometric drawing of the scene I made at its beginning, I began collecting reference for what types of materials I might need in the scene. 

Original Material Plan

Obviously, aspects of the scene were changed such as the door becoming a shutter, so this plan became more of an adaptive idea. The initial materials I knew I wanted to make were a black brick texture, a road surface texture, a wood planks texture and some form of white plaster texture was required from the brief.

Black Bricks Texture:

This texture took a lot of iterations to create, but in the end I think I produced a semi stylised brick texture that was black but could still be seen. Most of the issues I had with this material was that it was meant to be black, so I was struggling with the value of the bricks. This was something that was even brought up in my formative presentation and an older, darker version can even be seen in screenshots of that version. I found that I could go substantially lighter in value than I first thought, and so created this final texture.

Black Bricks Material in Designer

Black Bricks Node Set up

Road Surface:

This material also had a LOT of iterations - my original idea was to create a black tarmac texture but it looked very lazy, and after looking at references of Tokyo backstreets I found that a lot of them are block paved, so I began making a floor brick texture. I eventually also made an alternative texture, as can be seen below, which became the vertex painting option to reduce the obvious tiling quality of the scene and to add some grunge and wear.

Road Surface Nodes

Road Surface Material in Designer

Alternative Road Nodes

Alternative Road Surface in Designer

Wood Texture:

I got a lot of assistance and advice from Pat about this material, since wood is notoriously difficult to get right. Its a material that requires enough surface information to infer that it has grain and splinters, but too much noise especially in the normal map results in a messy, chaotic material that looks over produced. I got to the final stage here after a lot of trial and error, and this became a material that I put onto my final trim sheet.

Wood Planks Nodes

Wood Planks Material in Designer

White Plaster:

This was ultimately a really simple material that was a result of just a few blends of noises to create the texture. I took extra care into adding detail that looked like splatters and application patterns, but with a material that ultimately just became a white and rough noise map it was hard to make it extraordinary. I had to create this material however to fulfil the brief, and I am happy with the final result.

Plaster Material in Designer

Plaster Material Nodes

Smooth Stone:

This was a very simple material similar to the plaster, but this time I wanted to make sure I had extremely controlled normal and roughness values so as to not create a noisy surface. This material was going to be used sparingly - mostly as a new base to my fox statues and as a base to the shrine. It also became a part of my trim sheet so that I could use it for other assets, although it turned out to not be particularly useful there.

Smooth Stone Nodes

Smooth Stone Material in Designer

Roof tiles:

This was quite a complex material to make, and I used a lot of references of japanese tiled rooves to find a shape that worked. I also initially made the material much too dark, which resulted in it being hard to read in the scene. By making the base colour lighter it meant that the rooves were less of a focal point when contrasted with the white in the plaster.

To make the shape, I used the curve node in combination with many overlaid tiling gradients to create the individual tiles. Each tile also had a incline to make it look like the tiles were overlaying each other in the right pattern.

Roof Tiles Node

Roof Tiles Material in Designer

Trim Sheet:

This was a culmination of 3 materials; The wood planks, roof tiles and the smooth stone. The graph was basically just a masking set up to make sure that all the materials fit right, and I also rotated the wood grain so that it might be better used by my thinner beams. I think my use of the trim sheet was not optimal, but it did save a couple of extra tiling textures being used when I unwrapped. 

Trim Sheet Nodes

Trim Sheet Material in Designer

Shutters:

This was a material that I made quite late, but overall it is a very simple tiled gradient with a curve node attached to get that specific shape per individual shutter slat. Some variation was added through a simple slope blur that was used sparingly, and I think it produced quite a convincing material.

Shutter Material Nodes

Shutter Material in Designer

Foliage:

My final material was this foliage that I set up. I wanted to create a easily adjustable material for my central branch, so I created a simple leaf from reference of the "Japanese Snowball Bush" plant in google, and then created a symmetrical branch. I added some warping to it for variation, and then started building up the branches to produce a final bush cross section. I also created a flower from using a splatter circular node on a single flower shape, and created the flower centre with a shape splatter node. I also created a simple stem from making a shape, transforming it and putting it in the corner of the sheet. Overall I think this produced a great result that became very versatile and adaptable, and in my final scene I think it looked great.

Foliage Nodes

Foliage Material in Designer